Abstract image of a woman sitting on books and studying on a laptop

Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes

Kim Ahlstrøm, profile picture
Kim Ahlstrøm

The document discusses methods for answering provenance-enabled SPARQL queries over RDF data cubes, focusing on the importance of data provenance for improving query accuracy and trustworthiness. It explores the representation of provenance data, the challenges involved in querying, and presents a solution involving materialization strategies and context indexing. The authors provide examples and evaluate their approach, highlighting the implementation of provenance queries to filter and analyze trusted data sources.

Science
1 of 62
Download to read offline
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes February 7, 2017 Kim Ahlstrøm, Katja Hose, and Torben Bach Pedersen {kah|khose|tbp}@cs.aau.dk Department of Computer Science Aalborg University Denmark
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Agenda Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm 2 Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Preliminaries
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries 3 Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Resource Description Framework What is RDF RDF is a standard model for data interchange on the Web
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries 3 Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Resource Description Framework What is RDF RDF is a standard model for data interchange on the Web What is it good for? Linked data Hetrogenious data Publish data in a searchable manner
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries 3 Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Resource Description Framework What is RDF RDF is a standard model for data interchange on the Web What is it good for? Linked data Hetrogenious data Publish data in a searchable manner What does it look like? subject object subject literal predicate predicate
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries 4 Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark http://lod-cloud.net/
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries 5 Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark RDF Example sale1 7 book1Frank_Herbert “Dune” product nameauthor price product nameauthor price Graph A
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries 5 Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark RDF Example sale1 7 book1Frank_Herbert “Dune” product nameauthor price product nameauthor price Graph A
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries 5 Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark RDF Example sale1 7 book1Frank_Herbert “Dune” product nameauthor price product nameauthor price Graph A
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries 5 Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark RDF Example sale1 7 book1Frank_Herbert “Dune” product nameauthor price product nameauthor price Graph A
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries 5 Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark RDF Example sale1 7 book1Frank_Herbert “Dune” product nameauthor price product nameauthor price Graph A Subject Predicate Object Graph sale1 price 7 A sale1 product book1 A book1 name “Dune” A book1 author Frank_Herbert A
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries 6 Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark SPARQL SELECT ( count (? sale ) as ?sales ) FROM A WHERE { ?sale product ?book . ?book author Frank_Herbert . }
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework 7 RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark RDF Data Cube 9 7 12 4 85 3 6 product location 1 88 9 8 1 5 6 2 4 9 7 1 5 6 88 2 3 9 tim e
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework 7 RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark RDF Data Cube 9 7 12 4 85 3 6 product location 1 88 9 8 1 5 6 2 4 9 7 1 5 6 88 2 3 9 tim e Observations, Measures, Dimensions, and Levels
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework 7 RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark RDF Data Cube 9 7 12 4 85 3 6 product location 1 88 9 8 1 5 6 2 4 9 7 1 5 6 88 2 3 9 tim e Observations, Measures, Dimensions, and Levels Pure RDF data cubesab hybrid RDF data cubesc aRDF Analytics: Lenses over Semantic Graphs, WWW’14. bModeling and querying data warehouses on the semantic web using QB4OLAP, DaWaK’14. cFusion Cubes: Towards Self-Service Business Intelligence, IJDWM’13.
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework 8 RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Example :sale1:shop1 “Barnes & Noble” :book1“Dune” 7 “January” “2016” :january2016 :2016 :date31012016 :shop :product :name :title :price :date skos:broader skos:broader :month :year
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube 9 Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Provenance Data Provenance Provenance (from the French provenir, "to come from"), is the chronology of the ownership, custody or location of a historical object1 1Oxfort Dictionary. 2Automated metadata generation for linked data generation and publishing workflows, LDOW’16 . 3https://www.w3.org/TR/prov-o/.
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube 9 Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Provenance Data Provenance Provenance (from the French provenir, "to come from"), is the chronology of the ownership, custody or location of a historical object1 Where does it come from? Generated when data is manipulated2 1Oxfort Dictionary. 2Automated metadata generation for linked data generation and publishing workflows, LDOW’16 . 3https://www.w3.org/TR/prov-o/.
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube 9 Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Provenance Data Provenance Provenance (from the French provenir, "to come from"), is the chronology of the ownership, custody or location of a historical object1 Where does it come from? Generated when data is manipulated2 What do we use it for? Quality & Trust Access control Debugging 1Oxfort Dictionary. 2Automated metadata generation for linked data generation and publishing workflows, LDOW’16 . 3https://www.w3.org/TR/prov-o/.
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube 9 Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Provenance Data Provenance Provenance (from the French provenir, "to come from"), is the chronology of the ownership, custody or location of a historical object1 Where does it come from? Generated when data is manipulated2 What do we use it for? Quality & Trust Access control Debugging How do we represent provenance data? Prov-o vocabulary3 1Oxfort Dictionary. 2Automated metadata generation for linked data generation and publishing workflows, LDOW’16 . 3https://www.w3.org/TR/prov-o/.
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube 10 Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark RDF and provenance data Reification Named graphs Singleton properties n-ary Nano publications
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube 11 Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Named graphs sale1 7 book1Frank_Herbert “Dune” product nameauthor price Graph A
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube 11 Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Named graphs sale1 7 book1Frank_Herbert “Dune” product nameauthor price Graph A Subject Predicate Object :A rdf:type prov:entity :A prov:wasGeneratedBy :DataExtractor :DataExtractor prov:used :DBpedia :DBpedia :status "trusted"
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance 12 Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Provenance-enabled Query Provenance query Filters the provenance data. Analytical query Query the RDF data cube only containing allowed provenance data.
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query 13 The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark The Problem
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem 14 Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Provenance-enabled Queries How do we answer provenance-enabled queries over RDF data cubes?
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 15 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Cube Instance Data :sale1:shop1 “Barnes & Noble” :book1 :sale2 “Dune” :shop2“Amazon” :book2“Dracula” 12 7 “January” “2016” :january2016 :2016 :date31012016 :shop:shop :shop:shop :product:product :product:product :name:name :name:name :title:title :title:title :price:price :price:price :date:date :date:date skos:broaderskos:broader skos:broaderskos:broader :month:month :year:year
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 15 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Cube Instance Data :sale1:shop1 “Barnes & Noble” :book1 :sale2 “Dune” :shop2“Amazon” :book2“Dracula” 12 7 “January” “2016” :january2016 :2016 :date31012016 :shop:shop :shop:shop :product:product :product:product :name:name :name:name :title:title :title:title :price:price :price:price :date:date :date:date skos:broaderskos:broader skos:broaderskos:broader :month:month :year:year
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 16 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Provenance Data Subject Predicate Object :Blue prov:wasGeneratedBy :BookExtractor :BookExtractor prov:used :DBpedia :DBpedia :status "trusted" :Purple prov:wasGeneratedBy :ShopExtractor :ShopExtractor prov:used :Database :Database :status "trusted" :Teal prov:wasGeneratedBy :CalenderExtractor :CalenderExtractor prov:used :CSVFile :CSVFile :status "trusted" :Brown prov:wasGeneratedBy :WebTableExtractor :WebTableExtractor prov:used :WebTable :WebTable :status "unknown"
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 17 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Provenance-enabled Query Provenance Query All sources that are trusted SELECT ?provenanceGraph WHERE { ?provenanceGraph prov : wasGeneratedBy ? a c t i v i t y . ? a c t i v i t y prov : used ? e n t i t y . ? e n t i t y : status " trusted " . }
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 17 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Provenance-enabled Query Provenance Query All sources that are trusted SELECT ?provenanceGraph WHERE { ?provenanceGraph prov : wasGeneratedBy ? a c t i v i t y . ? a c t i v i t y prov : used ? e n t i t y . ? e n t i t y : status " trusted " . } Analytical Query Revenue of books sold in shop “Barnes & Noble” SELECT sum(? price ) WHERE { ?observation : price ? price . ?observation : shop ?shop . ?shop : t i t l e " Barnes & Noble " . }
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 18 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Naive Strategy 1. Evaluate Provenance Query 2. Reformulate Analytical Query 3. Execute Analytical Query
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 19 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Naive Strategy Step 1 Provenance Query All sources that are trusted SELECT ?provenanceGraph WHERE { ?provenanceGraph prov : wasGeneratedBy ? a c t i v i t y . ? a c t i v i t y prov : used ? e n t i t y . ? e n t i t y : status " trusted " . }
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 20 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Naive Strategy Step 1 Subject Predicate Object :Blue prov:wasGeneratedBy :BookExtractor :BookExtractor prov:used :DBpedia :DBpedia :status "trusted" :Purple prov:wasGeneratedBy :ShopExtractor :ShopExtractor prov:used :Database :Database :status "trusted" :Teal prov:wasGeneratedBy :CalenderExtractor :CalenderExtractor prov:used :CSVFile :CSVFile :status "trusted" :Brown prov:wasGeneratedBy :WebTableExtractor :WebTableExtractor prov:used :WebTable :WebTable :status "unknown"
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 20 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Naive Strategy Step 1 Subject Predicate Object :Blue prov:wasGeneratedBy :BookExtractor :BookExtractor prov:used :DBpedia :DBpedia :status "trusted" :Purple prov:wasGeneratedBy :ShopExtractor :ShopExtractor prov:used :Database :Database :status "trusted" :Teal prov:wasGeneratedBy :CalenderExtractor :CalenderExtractor prov:used :CSVFile :CSVFile :status "trusted" :Brown prov:wasGeneratedBy :WebTableExtractor :WebTableExtractor prov:used :WebTable :WebTable :status "unknown"
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 20 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Naive Strategy Step 1 Subject Predicate Object :Blue prov:wasGeneratedBy :BookExtractor :BookExtractor prov:used :DBpedia :DBpedia :status "trusted" :Purple prov:wasGeneratedBy :ShopExtractor :ShopExtractor prov:used :Database :Database :status "trusted" :Teal prov:wasGeneratedBy :CalenderExtractor :CalenderExtractor prov:used :CSVFile :CSVFile :status "trusted"
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 21 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Naive Strategy Step 1 Provenance Query All sources that are trusted SELECT ?provenanceGraph WHERE { ?provenanceGraph prov : wasGeneratedBy ? a c t i v i t y . ? a c t i v i t y prov : used ? e n t i t y . ? e n t i t y : status " trusted " . } Result :Blue, :Purple, and :Teal
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 22 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Naive Strategy Step 2 Analytical Query Revenue of books sold in shop “Barnes & Noble” SELECT sum(? price ) FROM :Blue FROM :Purple FROM :Teal WHERE { ?observation : price ? price . ?observation : shop ?shop . ?shop : t i t l e " Barnes & Noble " . }
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 23 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Naive Strategy Step 3 :sale1:shop1 “Barnes & Noble” :book1 :sale2 “Dune” :shop2“Amazon” :book2“Dracula” 12 7 “January” “2016” :january2016 :2016 :date31012016 :shop:shop :shop:shop :product:product :product:product :name:name :name:name :title:title :title:title :price:price :price:price :date:date :date:date skos:broaderskos:broader skos:broaderskos:broader :month:month :year:year
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 23 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Naive Strategy Step 3 :sale1:shop1 “Barnes & Noble” :book1 :sale2 “Dune” :shop2“Amazon” :book2“Dracula” 12 7 “January” “2016” :january2016 :2016 :date31012016 :shop:shop :shop:shop :product:product :product:product :name:name :name:name :title:title :title:title :price:price :price:price :date:date :date:date skos:broaderskos:broader skos:broaderskos:broader :month:month :year:year
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 24 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Query Tree σ(?shop :name “Barns & Nobels”) σ(?obs :shop ?shop) σ(?obs :price ?price) AGG(?price) Π (?price) :Purple:Teal :Blue :Purple:Teal :Blue :Purple:Teal :Blue Overhead per join
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example 25 Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Our Solution
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution 26 Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Materialization Strategy 1. Evaluate Provenance Query 2. Materialize Provenance Graphs 3. Execute Analytical Query
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution 27 Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Materialization Step 1 Provenance Query All sources that are trusted SELECT ?provenanceGraph WHERE { ?provenanceGraph prov : wasGeneratedBy ? a c t i v i t y . ? a c t i v i t y prov : used ? e n t i t y . ? e n t i t y : status " trusted " . } Result :Blue, :Purple, and :Teal
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution 28 Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Materialization Step 2 :sale1:shop1 “Barnes & Noble” :book1 :sale2 “Dune” :shop2“Amazon” :book2“Dracula” 12 7 “January” “2016” :january2016 :2016 :date31012016 :shop:shop :shop:shop :product:product :product:product :name:name :name:name :title:title :title:title :price:price :price:price :date:date :date:date skos:broaderskos:broader skos:broaderskos:broader :month:month :year:year
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution 28 Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Materialization Step 2 :sale1:shop1 “Barnes & Noble” :book1 :sale2 “Dune” :shop2“Amazon” :book2“Dracula” 12 7 “January” “2016” :january2016 :2016 :date31012016 :shop:shop :shop:shop :product:product :product:product :name:name :name:name :title:title :title:title :price:price :price:price :date:date :date:date skos:broaderskos:broader skos:broaderskos:broader :month:month :year:year
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution 29 Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Materialization Step 2 Materialize Cube’ Revenue of books sold in shop “Barnes & Noble” CONSTRUCT { ? subject ? predicate ? object } FROM :Blue FROM :Purple FROM :Teal WHERE { ? subject ? predicate ? object }
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution 30 Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Materialization Step 3 Analytical Query Revenue of books sold in shop “Barnes & Noble” SELECT sum(? price ) FROM :CUBE’ WHERE { ?observation : price ? price . ?observation : shop ?shop . ?shop : t i t l e " Barnes & Noble " . }
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy 31 Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Context Index 1. Analyze analytical query 2. Lookup predicates 3. Filter provenance graphs
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy 32 Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Context Index Query Analysis Analytical Query Revenue of books sold in shop “Barnes & Noble” SELECT sum(? price ) WHERE { ?observation :price ? price . ?observation : shop ?shop . ?shop :title " Barnes & Noble " . }
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy 33 Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Context Index Query Analysis :sale1:shop1 “Barnes & Noble” :book1 :sale2 “Dune” :shop2“Amazon” :book2“Dracula” 12 7 “January” “2016” :january2016 :2016 :date31012016 :shop :shop :product :product :name :name :title :title :price :price :date :date skos:broader skos:broader :month :year
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy 34 Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Context Index Lookup Root :product :name :shop :title :price :date skos:broader :month skos:broader :year :Teal:Teal:Blue:Brown:Purple:Purple observations Level 1 Level 2 Level 3 provenance :title :shop Root :price
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy 34 Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Context Index Lookup Root :product :name :shop :title :price :date skos:broader :month skos:broader :year :Teal:Teal:Blue:Brown:Purple:Purple observations Level 1 Level 2 Level 3 provenance :title :shop Root :price
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy 35 Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Context Index Combining Provenance Query & Context Index :sale1:shop1 “Barnes & Noble” :book1 :sale2 “Dune” :shop2“Amazon” :book2“Dracula” 12 7 “January” “2016” :january2016 :2016 :date31012016 :shop :shop :product :product :name :name :title :title :price :price :date :date skos:broader skos:broader :month :year
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy 35 Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Context Index Combining Provenance Query & Context Index :sale1:shop1 “Barnes & Noble” :book1 :sale2 “Dune” :shop2“Amazon” :book2“Dracula” 12 7 “January” “2016” :january2016 :2016 :date31012016 :shop :shop :product :product :name :name :title :title :price :price :date :date skos:broader skos:broader :month :year
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy 35 Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Context Index Combining Provenance Query & Context Index :sale1:shop1 “Barnes & Noble” :book1 :sale2 “Dune” :shop2“Amazon” :book2“Dracula” 12 7 “January” “2016” :january2016 :2016 :date31012016 :shop :shop :product :product :name :name :title :title :price :price :date :date skos:broader skos:broader :month :year
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index 36 Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Experiments
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments 37 Setup Evaluation Dept. of Computer Science Aalborg University Denmark Setup Dataset
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup 38 Evaluation Dept. of Computer Science Aalborg University Denmark Evaluation 1.0 1.8 8.0 13.5 101 102 103 104 Million of triples Evaluationtime(sec) Naive Naive+Index Materialized Materialized+Index
39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup 39 Evaluation Dept. of Computer Science Aalborg University Denmark Evaluation 10 30 50 70 90 50 100 150 200 Percentage size of provenance query Evaluationtime(sec) Materialized Materialized+Index
Thank you for your attention

More Related Content

PDF
Learning Commonalities in RDF
Sara EL HASSAD
 
PDF
Learning Commonalities in RDF and SPARQL
Sara EL HASSAD
 
PDF
Classification and Clustering of arXiv Documents, Sections, and Abstracts, Co...
Scientific Information Analytics Group, Prof. Gipp
 
PDF
ParlBench: a SPARQL-benchmark for electronic publishing applications.
Tatiana Tarasova
 
PDF
RDF Stream Processing Models (RSP2014)
Daniele Dell'Aglio
 
PDF
IPython Notebook as a Unified Data Science Interface for Hadoop
DataWorks Summit
 
PDF
Framester: A Wide Coverage Linguistic Linked Data Hub
Mehwish Alam
 
PDF
Heuristic based Query Optimisation for SPARQL
PlanetData Network of Excellence
 
Learning Commonalities in RDF
Sara EL HASSAD
 
Learning Commonalities in RDF and SPARQL
Sara EL HASSAD
 
Classification and Clustering of arXiv Documents, Sections, and Abstracts, Co...
Scientific Information Analytics Group, Prof. Gipp
 
ParlBench: a SPARQL-benchmark for electronic publishing applications.
Tatiana Tarasova
 
RDF Stream Processing Models (RSP2014)
Daniele Dell'Aglio
 
IPython Notebook as a Unified Data Science Interface for Hadoop
DataWorks Summit
 
Framester: A Wide Coverage Linguistic Linked Data Hub
Mehwish Alam
 
Heuristic based Query Optimisation for SPARQL
PlanetData Network of Excellence
 

What's hot (11)

PDF
Interactive Knowledge Discovery over Web of Data.
Mehwish Alam
 
PDF
LDQL: A Query Language for the Web of Linked Data
Olaf Hartig
 
PPTX
Programming with Semantic Broad Data
Steffen Staab
 
PDF
Sharing massive data analysis: from provenance to linked experiment reports
Gaignard Alban
 
PPTX
Democratizing Big Semantic Data management
WU (Vienna University of Economics and Business)
 
PDF
Rethinking Online SPARQL Querying to Support Incremental Result Visualization
Olaf Hartig
 
PPTX
(Semi-)Automatic analysis of online contents
Steffen Staab
 
PPTX
EDF2012 Peter Boncz - LOD benchmarking SRbench
European Data Forum
 
PDF
Apache Spark NLP for Healthcare: Lessons Learned Building Real-World Healthca...
Databricks
 
PDF
HARE: An Engine for Enhancing Answer Completeness of SPARQL Queries via Crowd...
Maribel Acosta Deibe
 
PDF
Navigating and Exploring RDF Data using Formal Concept Analysis
Mehwish Alam
 
Interactive Knowledge Discovery over Web of Data.
Mehwish Alam
 
LDQL: A Query Language for the Web of Linked Data
Olaf Hartig
 
Programming with Semantic Broad Data
Steffen Staab
 
Sharing massive data analysis: from provenance to linked experiment reports
Gaignard Alban
 
Democratizing Big Semantic Data management
WU (Vienna University of Economics and Business)
 
Rethinking Online SPARQL Querying to Support Incremental Result Visualization
Olaf Hartig
 
(Semi-)Automatic analysis of online contents
Steffen Staab
 
EDF2012 Peter Boncz - LOD benchmarking SRbench
European Data Forum
 
Apache Spark NLP for Healthcare: Lessons Learned Building Real-World Healthca...
Databricks
 
HARE: An Engine for Enhancing Answer Completeness of SPARQL Queries via Crowd...
Maribel Acosta Deibe
 
Navigating and Exploring RDF Data using Formal Concept Analysis
Mehwish Alam
 

Viewers also liked (20)

PDF
Using PostgreSQL with Bibliographic Data
Jimmy Angelakos
 
PDF
Bringing the Semantic Web closer to reality: PostgreSQL as RDF Graph Database
Jimmy Angelakos
 
PDF
Marco Gold Standard
MarcoTechnologies
 
PDF
10 Ways Your Boss Kills Employee Motivation
Officevibe
 
PDF
Recovery: Job Growth and Education Requirements Through 2020
CEW Georgetown
 
PDF
Beyond the Gig Economy
Jon Lieber
 
PDF
African Americans: College Majors and Earnings
CEW Georgetown
 
PDF
Creative Traction Methodology - For Early Stage Startups
Tommaso Di Bartolo
 
PDF
The Online College Labor Market
CEW Georgetown
 
PPTX
3 hard facts shaping higher education thinking and behavior
Grant Thornton LLP
 
PDF
8 Tips for Scaling Mobile Users in China by Edith Yeung
Edith Yeung
 
PDF
BigWeatherGear Group and Corporate Services Brochure 2013
Kristin Matson
 
PPTX
The French Revolution of 1789
Tom Richey
 
PDF
What's Trending in Talent and Learning for 2016?
Skillsoft
 
PDF
GAME ON! Integrating Games and Simulations in the Classroom
Brian Housand
 
PDF
Game Based Learning for Language Learners
Shelly Sanchez Terrell
 
PDF
SXSW 2016: The Need To Knows
Ogilvy Consulting
 
PDF
Digitized Student Development, Social Media, and Identity
Paul Brown
 
PDF
Bill Aulet GEC2016 keynote speech March 16 2016 Medellin Colombia
Massachusetts Institute of Technology
 
PDF
SXSW 2016 takeaways
Havas
 
Using PostgreSQL with Bibliographic Data
Jimmy Angelakos
 
Bringing the Semantic Web closer to reality: PostgreSQL as RDF Graph Database
Jimmy Angelakos
 
Marco Gold Standard
MarcoTechnologies
 
10 Ways Your Boss Kills Employee Motivation
Officevibe
 
Recovery: Job Growth and Education Requirements Through 2020
CEW Georgetown
 
Beyond the Gig Economy
Jon Lieber
 
African Americans: College Majors and Earnings
CEW Georgetown
 
Creative Traction Methodology - For Early Stage Startups
Tommaso Di Bartolo
 
The Online College Labor Market
CEW Georgetown
 
3 hard facts shaping higher education thinking and behavior
Grant Thornton LLP
 
8 Tips for Scaling Mobile Users in China by Edith Yeung
Edith Yeung
 
BigWeatherGear Group and Corporate Services Brochure 2013
Kristin Matson
 
The French Revolution of 1789
Tom Richey
 
What's Trending in Talent and Learning for 2016?
Skillsoft
 
GAME ON! Integrating Games and Simulations in the Classroom
Brian Housand
 
Game Based Learning for Language Learners
Shelly Sanchez Terrell
 
SXSW 2016: The Need To Knows
Ogilvy Consulting
 
Digitized Student Development, Social Media, and Identity
Paul Brown
 
Bill Aulet GEC2016 keynote speech March 16 2016 Medellin Colombia
Massachusetts Institute of Technology
 
SXSW 2016 takeaways
Havas
 

Similar to Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes (20)

PDF
Executing Provenance-Enabled Queries over Web Data
eXascale Infolab
 
PDF
Efficient, Scalable, and Provenance-Aware Management of Linked Data
eXascale Infolab
 
PDF
IRJET- Data Retrieval using Master Resource Description Framework
IRJET Journal
 
PPTX
Timbuctoo 2 EASY
henkvandenberg16
 
PPTX
Provenance for Data Munging Environments
Paul Groth
 
PPTX
Provenance for Reproducible Data Science
Andreas Schreiber
 
PDF
Enabling Citizen-empowered Apps over Linked Data
Diego López-de-Ipiña González-de-Artaza
 
PDF
Citizen-centric Linked Data Services for Smarter Cities
Diego López-de-Ipiña González-de-Artaza
 
PDF
Provenance Analysis and RDF Query Processing: W3C PROV for Data Quality and T...
satyasanket
 
ODP
2010 06 ipaw_prv
Jun Zhao
 
PPTX
Lotico oct 2010
dallemang
 
PPTX
RDF-Gen: Generating RDF from streaming and archival data
Giorgos Santipantakis
 
PDF
2010 06 rdf_next
Jun Zhao
 
PPT
Re-using Media on the Web: Media fragment re-mixing and playout
MediaMixerCommunity
 
PPTX
Triplestore and SPARQL
Lino Valdivia
 
PPTX
Elastic Meetup - Elasticsearch and Linked Data
Quentin Reul
 
PDF
IDEAS 2013 Presentation
Muntazir Mehdi
 
PPTX
Publishing "5 star" data: the case for RDF
PeterWinstanley1
 
PDF
Tutorial "An Introduction to SPARQL and Queries over Linked Data" Chapter 1 (...
Olaf Hartig
 
PPTX
Validating statistical Index Data represented in RDF using SPARQL Queries: Co...
Jose Emilio Labra Gayo
 
Executing Provenance-Enabled Queries over Web Data
eXascale Infolab
 
Efficient, Scalable, and Provenance-Aware Management of Linked Data
eXascale Infolab
 
IRJET- Data Retrieval using Master Resource Description Framework
IRJET Journal
 
Timbuctoo 2 EASY
henkvandenberg16
 
Provenance for Data Munging Environments
Paul Groth
 
Provenance for Reproducible Data Science
Andreas Schreiber
 
Enabling Citizen-empowered Apps over Linked Data
Diego López-de-Ipiña González-de-Artaza
 
Citizen-centric Linked Data Services for Smarter Cities
Diego López-de-Ipiña González-de-Artaza
 
Provenance Analysis and RDF Query Processing: W3C PROV for Data Quality and T...
satyasanket
 
2010 06 ipaw_prv
Jun Zhao
 
Lotico oct 2010
dallemang
 
RDF-Gen: Generating RDF from streaming and archival data
Giorgos Santipantakis
 
2010 06 rdf_next
Jun Zhao
 
Re-using Media on the Web: Media fragment re-mixing and playout
MediaMixerCommunity
 
Triplestore and SPARQL
Lino Valdivia
 
Elastic Meetup - Elasticsearch and Linked Data
Quentin Reul
 
IDEAS 2013 Presentation
Muntazir Mehdi
 
Publishing "5 star" data: the case for RDF
PeterWinstanley1
 
Tutorial "An Introduction to SPARQL and Queries over Linked Data" Chapter 1 (...
Olaf Hartig
 
Validating statistical Index Data represented in RDF using SPARQL Queries: Co...
Jose Emilio Labra Gayo
 

Recently uploaded (20)

PPT
what do you want to know about myeloprolifritive disorders .ppt
lasheseye676
 
PPTX
Personality for guidance related to theories
ananyakashyap127
 
PDF
Thyroid Hormone by Iqra Nasir detail.pdf
NasirIqra1
 
PPT
plant growth and development after seeding plant .ppt
ghaithalsmadi1997
 
PDF
Human Anatomy (Anatomy and Physiology A)
mockerburdstella
 
PPTX
Antihypertensive Medicinal Chemistry Unit II BP501T.pptx
Dhanashri Dupade
 
PDF
Physics of Bitcoin #30 Perrenod Santostasi.pdf
Stephen Perrenod
 
PDF
software engineering for computer science
shouryapratikofficia
 
PPT
dcs-computertraningbasics-170826004702.ppt
sujatasonar1975
 
PDF
2024_PohleJellKlug_CambrianPlectronoceratidsAustralia.pdf
Aridelabarrera
 
PPT
INSTRUMENTAL ANALYSIS (Electrochemical processes )-1.ppt
TomizUddin1
 
DOCX
Introduction , chapter 1 , Nahid Fatema thesis
NahidFatema3
 
PPTX
Bacterial and protozoal infections in pregnancy.pptx
aishuansari
 
PDF
Sujay Rao Mandavilli Variable logic FINAL FINAL FINAL FINAL FINAL.pdf
Sujay Rao Mandavilli
 
PPTX
Contact Lens Dr Hari.pptx presentation powerpoint
h4c9hnqgk5
 
DOCX
lesson plan on learning disabilities in children
pavin2
 
PDF
Sujay Rao Mandavilli Degrowth delusion FINAL FINAL FINAL FINAL FINAL.pdf
Sujay Rao Mandavilli
 
PDF
SWAG Research Lab Scientific Publications
swagresearchlab
 
PPTX
Morphology of Bacteria & Structure of Bacterial Cell
Ayantika17
 
PPTX
Cutaneous tuberculosis Dermatology
ShanmathiSethumadhav1
 
what do you want to know about myeloprolifritive disorders .ppt
lasheseye676
 
Personality for guidance related to theories
ananyakashyap127
 
Thyroid Hormone by Iqra Nasir detail.pdf
NasirIqra1
 
plant growth and development after seeding plant .ppt
ghaithalsmadi1997
 
Human Anatomy (Anatomy and Physiology A)
mockerburdstella
 
Antihypertensive Medicinal Chemistry Unit II BP501T.pptx
Dhanashri Dupade
 
Physics of Bitcoin #30 Perrenod Santostasi.pdf
Stephen Perrenod
 
software engineering for computer science
shouryapratikofficia
 
dcs-computertraningbasics-170826004702.ppt
sujatasonar1975
 
2024_PohleJellKlug_CambrianPlectronoceratidsAustralia.pdf
Aridelabarrera
 
INSTRUMENTAL ANALYSIS (Electrochemical processes )-1.ppt
TomizUddin1
 
Introduction , chapter 1 , Nahid Fatema thesis
NahidFatema3
 
Bacterial and protozoal infections in pregnancy.pptx
aishuansari
 
Sujay Rao Mandavilli Variable logic FINAL FINAL FINAL FINAL FINAL.pdf
Sujay Rao Mandavilli
 
Contact Lens Dr Hari.pptx presentation powerpoint
h4c9hnqgk5
 
lesson plan on learning disabilities in children
pavin2
 
Sujay Rao Mandavilli Degrowth delusion FINAL FINAL FINAL FINAL FINAL.pdf
Sujay Rao Mandavilli
 
SWAG Research Lab Scientific Publications
swagresearchlab
 
Morphology of Bacteria & Structure of Bacterial Cell
Ayantika17
 
Cutaneous tuberculosis Dermatology
ShanmathiSethumadhav1
 

Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes

  • 1. Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes February 7, 2017 Kim Ahlstrøm, Katja Hose, and Torben Bach Pedersen {kah|khose|tbp}@cs.aau.dk Department of Computer Science Aalborg University Denmark
  • 2. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Agenda Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation
  • 3. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm 2 Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Preliminaries
  • 4. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries 3 Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Resource Description Framework What is RDF RDF is a standard model for data interchange on the Web
  • 5. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries 3 Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Resource Description Framework What is RDF RDF is a standard model for data interchange on the Web What is it good for? Linked data Hetrogenious data Publish data in a searchable manner
  • 6. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries 3 Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Resource Description Framework What is RDF RDF is a standard model for data interchange on the Web What is it good for? Linked data Hetrogenious data Publish data in a searchable manner What does it look like? subject object subject literal predicate predicate
  • 7. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries 4 Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark http://lod-cloud.net/
  • 8. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries 5 Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark RDF Example sale1 7 book1Frank_Herbert “Dune” product nameauthor price product nameauthor price Graph A
  • 9. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries 5 Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark RDF Example sale1 7 book1Frank_Herbert “Dune” product nameauthor price product nameauthor price Graph A
  • 10. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries 5 Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark RDF Example sale1 7 book1Frank_Herbert “Dune” product nameauthor price product nameauthor price Graph A
  • 11. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries 5 Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark RDF Example sale1 7 book1Frank_Herbert “Dune” product nameauthor price product nameauthor price Graph A
  • 12. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries 5 Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark RDF Example sale1 7 book1Frank_Herbert “Dune” product nameauthor price product nameauthor price Graph A Subject Predicate Object Graph sale1 price 7 A sale1 product book1 A book1 name “Dune” A book1 author Frank_Herbert A
  • 13. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries 6 Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark SPARQL SELECT ( count (? sale ) as ?sales ) FROM A WHERE { ?sale product ?book . ?book author Frank_Herbert . }
  • 14. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework 7 RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark RDF Data Cube 9 7 12 4 85 3 6 product location 1 88 9 8 1 5 6 2 4 9 7 1 5 6 88 2 3 9 tim e
  • 15. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework 7 RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark RDF Data Cube 9 7 12 4 85 3 6 product location 1 88 9 8 1 5 6 2 4 9 7 1 5 6 88 2 3 9 tim e Observations, Measures, Dimensions, and Levels
  • 16. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework 7 RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark RDF Data Cube 9 7 12 4 85 3 6 product location 1 88 9 8 1 5 6 2 4 9 7 1 5 6 88 2 3 9 tim e Observations, Measures, Dimensions, and Levels Pure RDF data cubesab hybrid RDF data cubesc aRDF Analytics: Lenses over Semantic Graphs, WWW’14. bModeling and querying data warehouses on the semantic web using QB4OLAP, DaWaK’14. cFusion Cubes: Towards Self-Service Business Intelligence, IJDWM’13.
  • 17. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework 8 RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Example :sale1:shop1 “Barnes & Noble” :book1“Dune” 7 “January” “2016” :january2016 :2016 :date31012016 :shop :product :name :title :price :date skos:broader skos:broader :month :year
  • 18. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube 9 Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Provenance Data Provenance Provenance (from the French provenir, "to come from"), is the chronology of the ownership, custody or location of a historical object1 1Oxfort Dictionary. 2Automated metadata generation for linked data generation and publishing workflows, LDOW’16 . 3https://www.w3.org/TR/prov-o/.
  • 19. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube 9 Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Provenance Data Provenance Provenance (from the French provenir, "to come from"), is the chronology of the ownership, custody or location of a historical object1 Where does it come from? Generated when data is manipulated2 1Oxfort Dictionary. 2Automated metadata generation for linked data generation and publishing workflows, LDOW’16 . 3https://www.w3.org/TR/prov-o/.
  • 20. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube 9 Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Provenance Data Provenance Provenance (from the French provenir, "to come from"), is the chronology of the ownership, custody or location of a historical object1 Where does it come from? Generated when data is manipulated2 What do we use it for? Quality & Trust Access control Debugging 1Oxfort Dictionary. 2Automated metadata generation for linked data generation and publishing workflows, LDOW’16 . 3https://www.w3.org/TR/prov-o/.
  • 21. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube 9 Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Provenance Data Provenance Provenance (from the French provenir, "to come from"), is the chronology of the ownership, custody or location of a historical object1 Where does it come from? Generated when data is manipulated2 What do we use it for? Quality & Trust Access control Debugging How do we represent provenance data? Prov-o vocabulary3 1Oxfort Dictionary. 2Automated metadata generation for linked data generation and publishing workflows, LDOW’16 . 3https://www.w3.org/TR/prov-o/.
  • 22. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube 10 Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark RDF and provenance data Reification Named graphs Singleton properties n-ary Nano publications
  • 23. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube 11 Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Named graphs sale1 7 book1Frank_Herbert “Dune” product nameauthor price Graph A
  • 24. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube 11 Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Named graphs sale1 7 book1Frank_Herbert “Dune” product nameauthor price Graph A Subject Predicate Object :A rdf:type prov:entity :A prov:wasGeneratedBy :DataExtractor :DataExtractor prov:used :DBpedia :DBpedia :status "trusted"
  • 25. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance 12 Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Provenance-enabled Query Provenance query Filters the provenance data. Analytical query Query the RDF data cube only containing allowed provenance data.
  • 26. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query 13 The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark The Problem
  • 27. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem 14 Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Provenance-enabled Queries How do we answer provenance-enabled queries over RDF data cubes?
  • 28. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 15 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Cube Instance Data :sale1:shop1 “Barnes & Noble” :book1 :sale2 “Dune” :shop2“Amazon” :book2“Dracula” 12 7 “January” “2016” :january2016 :2016 :date31012016 :shop:shop :shop:shop :product:product :product:product :name:name :name:name :title:title :title:title :price:price :price:price :date:date :date:date skos:broaderskos:broader skos:broaderskos:broader :month:month :year:year
  • 29. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 15 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Cube Instance Data :sale1:shop1 “Barnes & Noble” :book1 :sale2 “Dune” :shop2“Amazon” :book2“Dracula” 12 7 “January” “2016” :january2016 :2016 :date31012016 :shop:shop :shop:shop :product:product :product:product :name:name :name:name :title:title :title:title :price:price :price:price :date:date :date:date skos:broaderskos:broader skos:broaderskos:broader :month:month :year:year
  • 30. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 16 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Provenance Data Subject Predicate Object :Blue prov:wasGeneratedBy :BookExtractor :BookExtractor prov:used :DBpedia :DBpedia :status "trusted" :Purple prov:wasGeneratedBy :ShopExtractor :ShopExtractor prov:used :Database :Database :status "trusted" :Teal prov:wasGeneratedBy :CalenderExtractor :CalenderExtractor prov:used :CSVFile :CSVFile :status "trusted" :Brown prov:wasGeneratedBy :WebTableExtractor :WebTableExtractor prov:used :WebTable :WebTable :status "unknown"
  • 31. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 17 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Provenance-enabled Query Provenance Query All sources that are trusted SELECT ?provenanceGraph WHERE { ?provenanceGraph prov : wasGeneratedBy ? a c t i v i t y . ? a c t i v i t y prov : used ? e n t i t y . ? e n t i t y : status " trusted " . }
  • 32. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 17 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Provenance-enabled Query Provenance Query All sources that are trusted SELECT ?provenanceGraph WHERE { ?provenanceGraph prov : wasGeneratedBy ? a c t i v i t y . ? a c t i v i t y prov : used ? e n t i t y . ? e n t i t y : status " trusted " . } Analytical Query Revenue of books sold in shop “Barnes & Noble” SELECT sum(? price ) WHERE { ?observation : price ? price . ?observation : shop ?shop . ?shop : t i t l e " Barnes & Noble " . }
  • 33. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 18 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Naive Strategy 1. Evaluate Provenance Query 2. Reformulate Analytical Query 3. Execute Analytical Query
  • 34. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 19 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Naive Strategy Step 1 Provenance Query All sources that are trusted SELECT ?provenanceGraph WHERE { ?provenanceGraph prov : wasGeneratedBy ? a c t i v i t y . ? a c t i v i t y prov : used ? e n t i t y . ? e n t i t y : status " trusted " . }
  • 35. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 20 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Naive Strategy Step 1 Subject Predicate Object :Blue prov:wasGeneratedBy :BookExtractor :BookExtractor prov:used :DBpedia :DBpedia :status "trusted" :Purple prov:wasGeneratedBy :ShopExtractor :ShopExtractor prov:used :Database :Database :status "trusted" :Teal prov:wasGeneratedBy :CalenderExtractor :CalenderExtractor prov:used :CSVFile :CSVFile :status "trusted" :Brown prov:wasGeneratedBy :WebTableExtractor :WebTableExtractor prov:used :WebTable :WebTable :status "unknown"
  • 36. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 20 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Naive Strategy Step 1 Subject Predicate Object :Blue prov:wasGeneratedBy :BookExtractor :BookExtractor prov:used :DBpedia :DBpedia :status "trusted" :Purple prov:wasGeneratedBy :ShopExtractor :ShopExtractor prov:used :Database :Database :status "trusted" :Teal prov:wasGeneratedBy :CalenderExtractor :CalenderExtractor prov:used :CSVFile :CSVFile :status "trusted" :Brown prov:wasGeneratedBy :WebTableExtractor :WebTableExtractor prov:used :WebTable :WebTable :status "unknown"
  • 37. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 20 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Naive Strategy Step 1 Subject Predicate Object :Blue prov:wasGeneratedBy :BookExtractor :BookExtractor prov:used :DBpedia :DBpedia :status "trusted" :Purple prov:wasGeneratedBy :ShopExtractor :ShopExtractor prov:used :Database :Database :status "trusted" :Teal prov:wasGeneratedBy :CalenderExtractor :CalenderExtractor prov:used :CSVFile :CSVFile :status "trusted"
  • 38. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 21 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Naive Strategy Step 1 Provenance Query All sources that are trusted SELECT ?provenanceGraph WHERE { ?provenanceGraph prov : wasGeneratedBy ? a c t i v i t y . ? a c t i v i t y prov : used ? e n t i t y . ? e n t i t y : status " trusted " . } Result :Blue, :Purple, and :Teal
  • 39. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 22 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Naive Strategy Step 2 Analytical Query Revenue of books sold in shop “Barnes & Noble” SELECT sum(? price ) FROM :Blue FROM :Purple FROM :Teal WHERE { ?observation : price ? price . ?observation : shop ?shop . ?shop : t i t l e " Barnes & Noble " . }
  • 40. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 23 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Naive Strategy Step 3 :sale1:shop1 “Barnes & Noble” :book1 :sale2 “Dune” :shop2“Amazon” :book2“Dracula” 12 7 “January” “2016” :january2016 :2016 :date31012016 :shop:shop :shop:shop :product:product :product:product :name:name :name:name :title:title :title:title :price:price :price:price :date:date :date:date skos:broaderskos:broader skos:broaderskos:broader :month:month :year:year
  • 41. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 23 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Naive Strategy Step 3 :sale1:shop1 “Barnes & Noble” :book1 :sale2 “Dune” :shop2“Amazon” :book2“Dracula” 12 7 “January” “2016” :january2016 :2016 :date31012016 :shop:shop :shop:shop :product:product :product:product :name:name :name:name :title:title :title:title :price:price :price:price :date:date :date:date skos:broaderskos:broader skos:broaderskos:broader :month:month :year:year
  • 42. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries 24 Example Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Query Tree σ(?shop :name “Barns & Nobels”) σ(?obs :shop ?shop) σ(?obs :price ?price) AGG(?price) Π (?price) :Purple:Teal :Blue :Purple:Teal :Blue :Purple:Teal :Blue Overhead per join
  • 43. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example 25 Our Solution Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Our Solution
  • 44. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution 26 Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Materialization Strategy 1. Evaluate Provenance Query 2. Materialize Provenance Graphs 3. Execute Analytical Query
  • 45. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution 27 Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Materialization Step 1 Provenance Query All sources that are trusted SELECT ?provenanceGraph WHERE { ?provenanceGraph prov : wasGeneratedBy ? a c t i v i t y . ? a c t i v i t y prov : used ? e n t i t y . ? e n t i t y : status " trusted " . } Result :Blue, :Purple, and :Teal
  • 46. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution 28 Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Materialization Step 2 :sale1:shop1 “Barnes & Noble” :book1 :sale2 “Dune” :shop2“Amazon” :book2“Dracula” 12 7 “January” “2016” :january2016 :2016 :date31012016 :shop:shop :shop:shop :product:product :product:product :name:name :name:name :title:title :title:title :price:price :price:price :date:date :date:date skos:broaderskos:broader skos:broaderskos:broader :month:month :year:year
  • 47. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution 28 Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Materialization Step 2 :sale1:shop1 “Barnes & Noble” :book1 :sale2 “Dune” :shop2“Amazon” :book2“Dracula” 12 7 “January” “2016” :january2016 :2016 :date31012016 :shop:shop :shop:shop :product:product :product:product :name:name :name:name :title:title :title:title :price:price :price:price :date:date :date:date skos:broaderskos:broader skos:broaderskos:broader :month:month :year:year
  • 48. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution 29 Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Materialization Step 2 Materialize Cube’ Revenue of books sold in shop “Barnes & Noble” CONSTRUCT { ? subject ? predicate ? object } FROM :Blue FROM :Purple FROM :Teal WHERE { ? subject ? predicate ? object }
  • 49. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution 30 Materialization Strategy Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Materialization Step 3 Analytical Query Revenue of books sold in shop “Barnes & Noble” SELECT sum(? price ) FROM :CUBE’ WHERE { ?observation : price ? price . ?observation : shop ?shop . ?shop : t i t l e " Barnes & Noble " . }
  • 50. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy 31 Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Context Index 1. Analyze analytical query 2. Lookup predicates 3. Filter provenance graphs
  • 51. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy 32 Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Context Index Query Analysis Analytical Query Revenue of books sold in shop “Barnes & Noble” SELECT sum(? price ) WHERE { ?observation :price ? price . ?observation : shop ?shop . ?shop :title " Barnes & Noble " . }
  • 52. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy 33 Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Context Index Query Analysis :sale1:shop1 “Barnes & Noble” :book1 :sale2 “Dune” :shop2“Amazon” :book2“Dracula” 12 7 “January” “2016” :january2016 :2016 :date31012016 :shop :shop :product :product :name :name :title :title :price :price :date :date skos:broader skos:broader :month :year
  • 53. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy 34 Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Context Index Lookup Root :product :name :shop :title :price :date skos:broader :month skos:broader :year :Teal:Teal:Blue:Brown:Purple:Purple observations Level 1 Level 2 Level 3 provenance :title :shop Root :price
  • 54. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy 34 Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Context Index Lookup Root :product :name :shop :title :price :date skos:broader :month skos:broader :year :Teal:Teal:Blue:Brown:Purple:Purple observations Level 1 Level 2 Level 3 provenance :title :shop Root :price
  • 55. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy 35 Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Context Index Combining Provenance Query & Context Index :sale1:shop1 “Barnes & Noble” :book1 :sale2 “Dune” :shop2“Amazon” :book2“Dracula” 12 7 “January” “2016” :january2016 :2016 :date31012016 :shop :shop :product :product :name :name :title :title :price :price :date :date skos:broader skos:broader :month :year
  • 56. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy 35 Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Context Index Combining Provenance Query & Context Index :sale1:shop1 “Barnes & Noble” :book1 :sale2 “Dune” :shop2“Amazon” :book2“Dracula” 12 7 “January” “2016” :january2016 :2016 :date31012016 :shop :shop :product :product :name :name :title :title :price :price :date :date skos:broader skos:broader :month :year
  • 57. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy 35 Context Index Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Context Index Combining Provenance Query & Context Index :sale1:shop1 “Barnes & Noble” :book1 :sale2 “Dune” :shop2“Amazon” :book2“Dracula” 12 7 “January” “2016” :january2016 :2016 :date31012016 :shop :shop :product :product :name :name :title :title :price :price :date :date skos:broader skos:broader :month :year
  • 58. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index 36 Experiments Setup Evaluation Dept. of Computer Science Aalborg University Denmark Experiments
  • 59. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments 37 Setup Evaluation Dept. of Computer Science Aalborg University Denmark Setup Dataset
  • 60. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup 38 Evaluation Dept. of Computer Science Aalborg University Denmark Evaluation 1.0 1.8 8.0 13.5 101 102 103 104 Million of triples Evaluationtime(sec) Naive Naive+Index Materialized Materialized+Index
  • 61. 39 Towards Answering Provenance-Enabled SPARQL Queries over RDF Data Cubes Kim Ahlstrøm Preliminaries Resource Description Framework RDF Data Cube Provenance Provenance-enabled Query The Problem Provenance-enabled Queries Example Our Solution Materialization Strategy Context Index Experiments Setup 39 Evaluation Dept. of Computer Science Aalborg University Denmark Evaluation 10 30 50 70 90 50 100 150 200 Percentage size of provenance query Evaluationtime(sec) Materialized Materialized+Index
  • 62. Thank you for your attention