![BOOLEAN MODEL
ď Each index term is either present or absent
ď Documents are either Relevant or Not Relevant(no
ranking)
ď A document is represented as a set of keywords.
ď Queries are Boolean expressions of
keywords, connected by AND, OR, and
NOT, including the use of brackets to indicate scope.
ď [[Rio & Brazil] | [Hilo & Hawaii]] & hotel & !Hilton]
ď Output: Document is relevant or not. No partial
matches or ranking.](https://image.slidesharecdn.com/modelofinformationretrieval3-130929115715-phpapp01/85/Model-of-information-retrieval-3-9-320.jpg)
Model of information retrieval (3)
- 1. BY N. SUMANJALI DPT OF LIS PONDICHERRY UNIVERSITY
- 2. INFORMATION RETRIEVAL ď Information retrieval is the activity of obtaining information resources relevant to an information need from a collection of information resources. ď Searches can be based on metadata or on full-text (or other content-based) indexing. ď Goal: Find the documents most relevant to a certain Query ď Dealing with notions of: ď Collection of documents ď Query (Userâs information need) ď Notion of Relevancy
- 3. MODEL ď A model is a construct designed help us understand a complex system ď A particular way of âlooking at thingsâ ď Models inevitably make simplifying assumptions ď What are the limitations of the model? ď Different types of models: ď Conceptual models ď Physical analog models ď Mathematical models
- 4. Retrieval Models ďA retrieval model specifies the details of: ď Document representation ď Query representation ď Retrieval function ďDetermines a notion of relevance. ďNotion of relevance can be binary or continuous (i.e. ranked retrieval).
- 5. CLASSES OF RM ďBoolean models (set theoretic) ď Extended Boolean ďVector space models (statistical/algebraic) ď Generalized VS ď Latent Semantic Indexing ďProbabilistic models
- 6. MODELS OF IR ď Boolean model ď Based on the notion of sets ď Documents are retrieved only if they satisfy Boolean conditions specified in the query ď Does not impose a ranking on retrieved documents ď Exact match ď Vector space model ď Based on geometry, the notion of vectors in high dimensional space ď Documents are ranked based on their similarity to the query (ranked retrieval) ď Best/partial match
- 7. ď Language models ď Based on the notion of probabilities and processes for generating text ď Documents are ranked based on the probability that they generated the query ď Best/partial match
- 8. BOOLEAN MODEL ď Invented by George Boole (1815-1864) ď He devised a system of symbolic logic in which he used three operators (+, , - ) to combine statements in symbolic form. ď John Venn named to this operators of Boolean logic are the logical sum(+), logical product(), and logical difference(-). ď IR systems allow the users to express their queries by using this operators.
- 9. BOOLEAN MODEL ď Each index term is either present or absent ď Documents are either Relevant or Not Relevant(no ranking) ď A document is represented as a set of keywords. ď Queries are Boolean expressions of keywords, connected by AND, OR, and NOT, including the use of brackets to indicate scope. ď [[Rio & Brazil] | [Hilo & Hawaii]] & hotel & !Hilton] ď Output: Document is relevant or not. No partial matches or ranking.
- 10. BOOLEAN RETRIEVAL MODEL ď Popular retrieval model because: ď Easy to understand for simple queries. ď Clean formalism. ď Boolean models can be extended to include ranking. ď Reasonably efficient implementations possible for normal queries.
- 11. BOOLEAN MODEL ď Weights assigned to terms are either â0â or â1â ď â0â represents âabsenceâ: term isnât in the document ď â1â represents âpresenceâ: term is in the document ď Build queries by combining terms with Boolean operators ď AND, OR, NOT ď The system returns all documents that satisfy the query
- 12. AND/OR/NOT A B C
- 13. Why Boolean Retrieval Works ď Boolean operators approximate natural language ď Find documents about a good party that is not over ď AND can discover relationships between concepts ď good party ď OR can discover alternate terminology ď excellent party, wild party, etc. ď NOT can discover alternate meanings ď Democratic party
- 14. The Perfect Query Paradox ď Every information need has a perfect set of documents ď If not, there would be no sense doing retrieval ď Every document set has a perfect query ď AND every word in a document to get a query for it ď Repeat for each document in the set ď OR every document query to get the set query ď But can users realistically be expected to formulate this perfect query? ď Boolean query formulation is hard!
- 15. Why Boolean Retrieval Fails ⢠Natural language is way more complex ⢠AND âdiscoversâ nonexistent relationships â Terms in different sentences, paragraphs, ⌠⢠Guessing terminology for OR is hard â good, nice, excellent, outstanding, awesome, ⌠⢠Guessing terms to exclude is even harder! â Democratic party, party to a lawsuit, âŚ
- 16. BOOLEAN MODEL ď Strengths ď Precise, if you know the right strategies ď Precise, if you have an idea of what youâre looking for ď Efficient for the computer ď Simple ď Weaknesses ď Users must learn Boolean logic ď Boolean logic insufficient to capture the richness of language ď No control over size of result set: either too many documents or none ď When do you stop reading? All documents in the result set are considered âequally goodâ ď What about partial matches? Documents that âdonât quite matchâ the query may be useful also ď No notion of ranking (exact matching only) ď All index terms have equal weight
- 17. PROBLEMS ď Very rigid: AND means all; OR means any. ď Difficult to express complex user requests. ď Difficult to control the number of documents retrieved. ď All matched documents will be returned. ď Difficult to rank output. ď All matched documents logically satisfy the query. ď Difficult to perform relevance feedback. ď If a document is identified by the user as relevant or irrelevant, how should the query be modified?
- 18. ADVANTAGES & DISADVANTAGES ď Advantages ď Results are predictable, relatively easy to explain ď Many different features can be incorporated ď Efficient processing since many documents can be eliminated from search ď Disadvantages ď Effectiveness depends entirely on user ď Simple queries usually donât work well ď Complex queries are difficult.
- 19. LIMITATIONS ď The first relates to the formulation of search statements. ď It has been noted that users are not able to formulate an exact search statement by the combination of AND, OR and NOT operators, especially when several query terms are involved. ď In such cases either the search statement becomes too narrow or too broad. ď The second limitation relates to the number of retrieval items. ď It has been noted that users cannot predict a priori exactly how many items are to be retrieved to satisfy a given query. ď If the search statement is broad, the number of retrieved items may sometimes be several hundreds and thus it may be quite difficult to find out the exact information required. ď The third limitation is that it identifies an item as relevant by finding out whether a given query term is present or not in a given record in the database.