A Literature Survey on Recommendation System Based on Sentimental Analysis

Advanced Computational Intelligence: An International Journal (ACII), Vol.3, No.1, January 2016
DOI:10.5121/acii.2016.3103 25
A LITERATURE SURVEY ON RECOMMENDATION
SYSTEM BASED ON SENTIMENTAL ANALYSIS
Achin Jain1
, Vanita Jain2
and Nidhi Kapoor3
BharatiVidyapeeth College of Engineering, New Delhi
ABSTRACT
Recommender systems have grown to be a critical research subject after the emergence of the first paper
on collaborative filtering in the Nineties. Despite the fact that educational studies on recommender systems,
has extended extensively over the last 10 years, there are deficiencies in the complete literature evaluation
and classification of that research. Because of this, we reviewed articles on recommender structures, and
then classified those based on sentiment analysis. The articles are categorized into three techniques of
recommender system, i.e.; collaborative filtering (CF), content based and context based. We have tried to
find out the research papers related to sentimental analysis based recommender system. To classify
research done by authors in this field, we have shown different approaches of recommender system based
on sentimental analysis with the help of tables. Our studies give statistics, approximately trends in
recommender structures research, and gives practitioners and researchers with perception and destiny
route on the recommender system using sentimental analysis. We hope that this paper enables all and
sundry who is interested in recommender systems research with insight for destiny.
KEYWORDS
Recommender systems; Literature review, Sentimental analysis
1. INTRODUCTION
We currently live in an era of information. We are surrounded by a plethora of data in the form of
reviews, blogs, papers and comments on various websites. The number of people around the
world who use the internet has witnessed an increase of approximately 40% since 1995 and
reached a count of 3.2 billion. The increased information flow has opened more avenues, but it
has also led to added confusion for the user. Amidst this huge amount of data, the task of making
certain decisions becomes difficult. It is rightly said that one should make an informed decision,
but too much information can also hinder the decision-making process. Thus, in order to save a
user from this confusion and make the experience of surfing the internet a pleasurable one,
recommender systems were introduced. Francesco Ricci, LiorRokach and BrachaShapira define
the recommender systems as software tools that make relevant suggestions to a user [1], [2].
Depending upon the user profile and the product profile, which are formed using various
techniques and algorithms, suggestions are made. More than 32% of consumers rate a product
online, over 33% writes reviews and nearly 88% trust online reviews [14]. Thus, reviews play an
essential role in affecting the sales of a commodity or a service. Each review posted on the web
consists of the user’s sentiments (positive or negative) and preferences. Sentiment analysis helps
in determining the attitude of the writer by computationally dividing opinions in a piece of text
into positive, negative or neutral [11]. Extracting the sentiments in reviews can largely contribute
to the quality of the recommender system by incorporating in it valuable information present in
the reviews and also help in the understanding that how a particular review affects the consumer

26
[11], [13]. Different techniques such as the weighted algorithm are used to produce scores for
different texts [13].The usage of the process of sentiment analysis paves the way for the
development of personalized recommender system. Extensive research has been done in this field
of recommendation systems. Recommendation systems have been broadly divided into three
categories. These categories are collaborative filtering, content based and context based
recommendation systems.
The rest of the paper is organized as follows: Section 2 gives a brief overview of various types of
Recommender systems. Section 3, 4 and 5 gives a detailed overview about the types of
recommender systems with related work carried out by researchers in tabular form.Finally,
section 6 concludes this paper.
2. TYPES OF RECOMMENDER SYSTEM
Over the years, recommender systems have been studied widely and are divided into different
categories according to the approach being used. The categories are collaborative filtering (CF),
content based and context based.
2.1 Collaboration filtering
Collaborative filtering (CF) uses the numerical reviews given by the user and is mainly based
upon the historical data of the user available to the system [4], [5]. The historical data available
helps to build the user profile and the data available about the item is used to make the item
profile. Both the user profile and the item profile are used to make a recommendation system. The
Netflix Competition has given much popularity to collaborative filtering [7], [5]. Collaborative
filtering is considered the most basic and the easiest method to find recommendations and make
predictions regarding the sales of a product. It does have some disadvantages which has led to the
development of new methods and techniques.
2.2 Content Based Recommender System
Content based [9] systems focus on the features of the products and aim at creating a user profile
depending on the previous reviews and also a profile of the item in accordance with the features it
provides and the reviews it has received [8], [5].It is observed that reviews usually contain
product feature and user opinion in pairs [5], [9], [10]. It is observed that users’ reviews contain a
feature of the product followed by his/her opinion about the product. Content based
recommendation systems help overcome sparsity problem that is faced in collaborative filtering
based recommendation system.
2.3 Context Based Recommender System
Extending the user/item convention to the circumstances of the user to incorporate the contextual
information is what is achieved in context-based recommender systems [15]. This helps to
abandon the cumbersome process of making the user fill a huge number of personal details.

3. COLLABORATIVE FILTERING
Collaboration filtering is a technique for predicting unknown preferences of people by using
already known preferences from many users [6
and the other is the item. It uses cosine and Pearson correlation s
main challenges that Collaborative Filtering deals with are data sparsity, scalability and cold start
problem. CF introduces three main algorithms: memory
which are used to combine CF with oth
with the challenges [5].
Cosine similarity
Given two vectors of attributes, A and B, the cosine similarity, cos(
product and magnitude as[25].
Pearson correlation similarity
Where(x, y) refers to the data objects and N is the total number of
3.1 Memory-Based Collaborative
People with similar interests are combined to form a group and every user is a part of that
[36]. User –based CF and Item-
implement and scales well with correlated items. There is no need
items being recommended. There are many limitations of memory
problem, sparsity and their dependencies on human ratings [37].
3.2 Model-based CF
Complex patters which are based on training data, are
the models (such as data mining algorithms, machine learning) and then intelligent predictions are
made for CF tasks for the real world data which are based on learnt models[38,39]. It
intuitive rationale for recommendations. Model
disadvantage of model-based CF is that it loses useful information for dimensionality reduction
techniques [37].
ILTERING
already known preferences from many users [6]. It computes similarity on two basis:
and the other is the item. It uses cosine and Pearson correlation similarity approach [19]. The
problem. CF introduces three main algorithms: memory-based, model-based, and hybrid CF,
which are used to combine CF with other recommendation techniques and their power to deal
Given two vectors of attributes, A and B, the cosine similarity, cos(θ), is represented using a dot
similarity
) refers to the data objects and N is the total number of attributes [25].
Based Collaborative Filtering (Neighborhood Based)
People with similar interests are combined to form a group and every user is a part of that
-based CF are used to represent memory-based CF. It is easy to
implement and scales well with correlated items. There is no need of considering the content of
items being recommended. There are many limitations of memory-based CF like cold start
problem, sparsity and their dependencies on human ratings [37].
Complex patters which are based on training data, are recognized by designing and developing
made for CF tasks for the real world data which are based on learnt models[38,39]. It
recommendations. Model-building is an expensive procedure. Other
based CF is that it loses useful information for dimensionality reduction
27
]. It computes similarity on two basis: one is user
imilarity approach [19]. The
based, and hybrid CF,
er recommendation techniques and their power to deal
), is represented using a dot
People with similar interests are combined to form a group and every user is a part of that group
based CF. It is easy to
of considering the content of
based CF like cold start
recognized by designing and developing
made for CF tasks for the real world data which are based on learnt models[38,39]. It gives an
building is an expensive procedure. Other
based CF is that it loses useful information for dimensionality reduction

28
3.3 Hybrid Collaborative Filtering Techniques
In hybrid recommender system, different techniques of collaborative approaches and other
recommender techniques (usually content based approaches), combined to get better results.
Various problems like cold-start, data sparsity and scalability can be avoided by using hybrid
approach [40].There are different ways of combining CF with other recommender techniques
which are following:
• Hybrid Recommenders Incorporating CF and Content-Based Features
• Hybrid Recommenders Combining CF and Other Recommender Systems
• Hybrid Recommenders Combining CF Algorithms [37].
Fig. 1. illustrates the flow chart of Collaborative Filtering Recommender System. It shows how
collaborative filtering considers only numerical reviews given by different users and then gives
recommended products as result. The user reviews are stored in a database to make further
refrences and predictions. In the figure User 1 and User 6 show similar behaviour and thus their
profiles lie in the same neighborhood which indicates similar interests. Using this similarity,
review about a product not rated by User 6 can be predicted using the reviews of User 1 that are
available. Thus, a prediction regarding product C’s review by User 6 is made using the available
data. From these predictions, recommendations are extracted and suggested to the user.
Figure 1. Collaborative Filtering Recommender System

29
Table 1 shows different collaborative filering methods based on sentimental analysis for
recommendation system. These methods are based on a variety of different models. Over the
yaers, different models are combined with collaborative filtering technique of recommender
systems in order to increase the accuracy of the predictions given by the results. J. Wang et. Al.
developed the unified relevance model based recommendation system while Parlov et. Al. used
maximum entropy approach and both these models helped reduce data spartsity problem, one of
the commonly faced problem in collaborative filtering based recommendation systems.
Table 1. Collaborative Filtering Methods
Method Description References
Unified relevance
model
It is a probabilistic item-to-user relevance
framework which uses Parzen-window
method for density estimation. This approach
reduces data sparsity problem.
J. Wang, A. P. de
Vries, and M. J. T.
Reinders[28]
Hybrid CF model It introduces effective recommender system
using sequential mixture CF and joint mixture
CF. It also implements advanced Bayes belief
networks.
X. Su, R. Greiner,
T. M.
Khoshgoftaar[27]
Fuzzy Association
Rules and
Multilevel
Similarity
(FARAMS)
It uses fuzzy association rule mining to extend
the existing techniques. FARAMS achieved the
task of generating more qualitative predictions.
C. W. K. Leung, S.
C. F. Chan, and F.
L. Chung[29]
Flexible mixture
model (FMM)
Simultaneous creation of user and item clusters.
It introduces preference nodes to study a
dramatic variation of the rating among users
with similar tastes.
L. Si and R. Jin[26]
Maximum entropy
approach
Clustering of items based on user access path in
order to reduce the apriori probability. This
helps in addressing sparsity and dimensionality
reduction.
D. Y. Pavlov and
D. M. Pennock
[30]
4. CONTENT-BASED RECOMMENDER SYSTEM
It is a technique where individual user profiles are taken into account. It enhances the user’s
interest and predicts whether the user would be interested in eating at any particular restaurant or
interested in seeing any particular movie [22]. It is also known as adaptive Filtering as it provides
suggestions according to user's field of interest and adapts user’s likes and dislikes. It represents
the comparison between the content contained in the item with the content of items of user's
interest. By using Bayesian hierarchical model, better user profiles for upcoming users is made by
collecting feedbacks from the old users [20]. Content based collaborative filtering is more widely
used to compare pure CF and pure Content-base. In CF the problem of sparsity is overcome
(converting sparse user filled matrix into full user rating matrix) by using content-based
prediction [21]. Fig.2 displays the flow of information in a content based recommendation
system. Relevant entities of an item and relations are kept together as input. Main features of
items are extracted from item ontology. Features of items, user's ratings and user modeling data

30
are applied to content based recommender system. After applying, various recommended items
are given as output.
Figure 2. Content-Based Recommender System
Table 2 shows different content based methods using sentimental analysis for recommendation
system. These methods are based on a variety of different models. Different researchers have
studied the relevance of various techniques that can be implemented in content based
recommendation systems. Accuracy and relevance of the recommendation systems have become
better by extensive research. Content-Boosted Collaborative filtering technique was implemented
by Prem Melville et. Al. Recommendation system using content based technique and Bayesian
Hierarchical Model (BHM) was built by Marko Balabanovic.
Table 2. Content-Based Recommender System
Method Description References
Content-Boosted
Collaborative Filtering
It gives an approach to combine
content and collaboration to enhance
existing user data and to give better
performance than a pure content
based predictor.
Prem Melville, Raymond
J. Mooney,
RamadassNagarajan[21]
FAB Technique An adaptive recommendation service
for collection and selection of web
pages. It makes the system more
personalized and combines the
benefits of content analysis with the
shared user interests.
Marko Balabanovic [24]
Bayesian
hierarchical
model(BHM)
Proposes a faster technique to gather
a huge number of individual user
profiles even if feedbacks available
are less. It uses various parameters of
BHM for optimization of joint data
likelihood.
Yi Zhang , Jonathan
Koren
[20]

31
Keyword Map
Algorithm
It captures the relations between
conditions that the learner has been
exposed to, which are used to
represent the knowledge of the
learner. The relevance and
complement of learning resources
recommendation are increased by
keyword map. It is better suited for
e-learning settings and achieves
higher accuracy than common
recommender methods.
Xin Wan, Neil
Rubens,Toshio Okamoto
and Yan Feng
[23]
5. CONTEXT-BASED RECOMMENDER SYSTEM
Recommender systems which use collaborative filtering technique or which are based on the
contents and features of items have achieved great success but can be further improved. The
Higher level of personalization is required in recommender systems to give more appropriate
suggestions. In order to achieve this, the contextual information of users is also taken into
consideration while designing a recommender system. Context refers to the time, location, area
and environment of the user which define a user’s status. Incorporating contextual information in
a recommender system helps to get a clear picture of the situation of any individual, place or
object which is of relevance to the system for prediction [17], [18].
It aids in extracting information about a particular community of individuals and this information
proves to be of high importance to improve the suggestions provided to a user and makes the
system more efficient. Recommender systems require situational information of the user and
context based recommender system accesses this information directly using various techniques
(such as GPS) [16] and does not bother the user with this. Fig. 3 displays the flow of contextual
information in a context based recommender system. The user's location data, social data, current
time, weather data is taken into consideration as the contextual data and is given as input to the
system. An approximate address of the user is determined and the location is saved. Social data of
a user can be accessed by requesting permission to a social account of the user.
The device the person is using can be used to get the time of the day which play a role in finding
suitable recommendations for the user. Accessing the location can also help in finding the
weather of the place. All the information gathered is processed in a system and sentiment analysis
is done. After processing the system gives an output of a list of attractions for the user. Depending
on the application of the system, desired results are produced. If the application is for a tourist,
destinations or shortest routes or hotels might be given as recommendations.
Contextual factors are of two types: Dynamic and static, depending on whether they change with
time or not. .
1. Dynamic: When the contextual factors change over time and hence unstable. They may
change by explicit user feedback. User feedback is generally used for refining the profile
of user to get better results of recommendations. The biggest challenge is that if a system
is considered to be dynamic then the system should be able to find out when to switch to
a different underlying context model.

32
2. Static: The contextual factors don’t change over time and hence stable. For e.g. to buy a
cell phone the contextual factors can be Time, purpose of purchasing and only them while
entire purchasing purpose recommendation application runs.
Figure 3. Context-Based Recommender System
Contextual factors are of three types: Fully observable, partially observable and Unobservable,
depending on what is being observed (or what system exactly knows).
1. Fully observable: Complete structure and values of contextual factors are known
explicitly, at the time when recommendations are made.
2. Partially observable: Some of the information is known explicitly about the contextual
factors.
3. Unobservable: There is no information of contextual factors explicitly available in
it.[34].

33
Figure 4. Contextual Information Dimensions
Table 3 shows different context based methods with sentimental analysis for recommendation
system.The table gives recommender system by Norma Saiph Savage in which context based
technique using Hidden Markov Model is used. It helps improve location
recommendations.Maciej Baranski et al. used a multidimensional database to help increase
credibility of a system. Gediminaset. Al. used Fuzzy Bayesian Networks. . Human memory
model was used as the base for a context based system by Sarabjot Singh Anand et al.
VíctorCodina et al. created a recommendation system based on Matrix-factorization Predictive
Context Model and observed more accurate results.
Table 3. Context-Based Recommender System
METHOD DESCRIPTION REFERENCES
Hidden Markov Model Improved version of a location
recommender system by
implementing Decision Tree (DT)
along with discrete Hidden Markov
Model (HMM).Together HMM and
DT differentiate between transport
modes and reduce noise.
Norma Saiph Savage
Maciej Baranski
Norma Elva Chavez
Tobias Höllerer[34]
Multidimensional
approach
It provides additional contextual
information on user and item, and
also supports multiple
Dimensions, profiling information,
and hierarchical aggregation of
recommender system.
GediminasAdomavicius
Ramesh
Sankaranarayanan
Shahana Sen
Alexander Tuzhilin
[31]
Fuzzy Bayesian
Networks
It gives a recommender system
which exploits the fuzzy system,
Han-SaemPark,Ji-Oh
YooSung-Bae Cho

34
Bayesian Networks in order to get
appropriate recommendation with
respect to the context.
[33]
Human memory model A recommender system is proposed
which retrieves relevant preference
information from long term
memory and uses
it in conjunction with the
information stored in short term
memory.
Sarabjot Singh Anand
and BamshadMobasher
[32]
Matrix-factorization
Predictive Context
based
Model
Distributional-Semantics Pre-
filtering (DSPF) approach is used
to build more precise context aware
rating prediction models, by
exploiting, in a novel way, the
distributional Semantics of
contextual conditions. It also shows
how DSPF can be improved by
using clustering techniques.
Víctor Codina
FrancescoRicci
LuigiCeccaroni
[35]
6. CONCLUSIONS
Recommender structures are proving to be a useful device for addressing a part of the records
overload phenomenon from the internet. Its evolution has followed the evolution of the internet.
The primary technology of recommender system used conventional web sites to gather
information from the following sources: (a) content material-primarily based records (b)
demographic statistics, and (c) memory-primarily based information.
Latest research shows the use of Sentimental Analysis in developing of more accurate
recommender system. These types of methods are commonly used in e-commerce business. In
this paper we have classify various approaches of recommender system that are based on
Sentiemental analysis.
Future research will deal with advancing the prevailingtechniques and algorithms to enhance the
nice of recommenderstructures predictions and hints.
REFERENCES
[1] Francesco Ricci, LiorRokach, BrachaShapira and Paul B. Kantor - Recommender Systems
Handbook; First Edition; Springer-Verlag New York, Inc. New York, NY, USA, 2010.
[2] Tariq Mahmood and Francesco Ricci,” Improving recommender systems with adaptive
conversational strategies”, 20th
ACM conference on Hypertext and Hypermedia, pp. 73–82, ACM,
July 2009.
[3] Tariq Mahmood, Francesco Ricci, Adriano Venturini and Wolfram Höpken, “Adaptive
recommender systems for travel planning”, Information and Communication Technologies in
Tourism 2008, Proceedings of the International Conference,Innsbruck Austria, pp. 1 - 11, 2008.
[4] X. Su and T. Khoshgoftaar, “A survey of collaborative filtering techniques,” Advances in Artificial
Intelligence, vol. 2009, pp. 19, August 2009.

35
[5] Yongfeng Zhang, Min Zhang and Yiqun Liu, “Incorporating Phrase-level Sentiment Analysis on
Textual Reviews for Personalized Recommendation”,Eighth ACM International Conference on
Web Search and Data Mining, pp. 435 – 440, February 2015.
[6] P. Resnick and H. R. Varian, “Recommender systems,” Communications of the ACM, vol. 40, no.
3, pp. 56–58, 1997.
[7] J. Bennett and S. Lanning, “The Netflix Prize”, ACM SIGKDD Explorations Newsletter - Special
issue on visual analytics, Vol. 9 Issue 2, pp. 51 – 52, December 2007.
[8] Titov and R. McDonald, “A joint model of text and aspect ratings for sentiment summarization”,
Annual Meeting of the Association for Computational Linguistics, pp. 308 – 316, June 2008.
[9] P. Lops, M. de Gemmis and G. Semeraro, “Content-based recommender systems: State of the art
and trends”, Recommender Systems Handbook, pp. 73 - 105, 2011.
[10] X. Ding, B. Liu, and P. S. Yu, “A Holistic Lexicon-Based Approach to Opinion Mining”, Web
Search and Data Mining, pp. 231 - 239, February 2008.
[11] A. Kennedy and D. Inkpen, “Sentiment Classification of Movie and Product Reviews Using
Contextual Valence Shifters”, Computational Intelligence, Vol. 22, No. 2, pp. 110 – 125, May
2006.
[12] A. Andreevskaia, S. Bergler and M. Urseanu “All Blogs Are Not Made Equal: Exploring Genre
Differences in Sentiment Tagging of Blogs”, International Conference on Weblogs and Social
Media (ICWSM-2007), Boulder, CO, March 2007.
[13] AntonisKoukourikos, GiannisStoitsis and Pythagoras Karampiperis, “Sentiment Analysis: A tool
for Rating Attribution to Content in Recommender Systems”, 7th
European Conference on
Technology Enhanced Learning, September 2012.
[14] Bo Pang and Lillian Lee, “Opinion Mining and Sentiment Analysis”, Foundations and Trends in
Information Retrieval, Vol. 2, Issue 1 – 2, pp. 1 – 135, January 2008.
[15] Umberto Panniello, Alexander Tuzhilin, Michele Gorgoglione, CosimoPalmisano and AntoPedone,
“ Experimental Comparison of Pre- vs. Post-Filtering approaches in Context-Aware Recommender
Systems”, ACM Conference on Recommender Systems, October 2009.
[16] Fan Yang and Zhi-Meiwang, “A Mobile Location-based Information Recommendation System
Based on GPS and WEB2.0 Services”, WSEAS Transactions on Computers, Vol. 8, Issue 4, pp.
725 – 734, April 2009.
[17] Jae Sik Lee and Jin Chun Lee, “Context Awareness by Case-Based Reasoning in a Music
Recommendation System”, UCS, Vol. 4836, pp. 45 – 58, 2007.
[18] Asher Levi, Osnat (Ossi) Mokryn, Christophe Diot and Nina Taft, “Finding a Needle in a Haystack
of Reviews: Cold Start Context-Based Hotel Recommender System”,6th ACM conference on
Recommender Systems, September 2012.
[19] Hyung Jun Ahn, “A new similarity measure for collaborative filtering to alleviate the new user
cold-starting problem”, Information Sciences: an International Journal, Vol. 178, Issue 1, pp 37 –
51, January 2008.
[20] Yi Zhang and Jonathan Koren, “Efficient Bayesian Hierarchical User Modeling for
Recommendation Systems”,30th annual International ACM SIGIR Conference on Research and
Development, pp. 47 – 54, July 2007.
[21] Prem Melville, Raymond J. Mooney and RamadassNagarajan, “Content-Boosted Collaborative
Filtering for Improved Recommendations”, 18th
National Conference for Artificial Intelligence
(AAAI), pp. 187 – 192, August 2002.
[22] ChumkiBasu, Haym Hirsh and William Cohen, “Recommendation as Classification: Using Social
and Content-Based Information in Recommendation”, American Association for Artificial
Intelligence (AAAI), 1998.
[23] Xin Wan, Neil Rubens, Toshio Okamoto and Yan Feng, “Content Filtering Based on Keyword
Map”, 2nd International Conference on Electrical, Computer Engineering and Electronics
(ICECEE ), pp. 484 – 489, May 2015.
[24] Marko Balabanovic, “An Adaptive Web Page Recommendation Service”, First International
Conference on Autonomous Agents, pp. 378 – 385, February 1997.

36
[25] Paul Marx - Providing Actionable Recommendations: A Movie Recommendation Algorithm with
Explanatory Capability, Joseph EulVerlag, 2013.
[26] L. Si and R. Jin, “Flexible mixture model for collaborative filtering,” 20th International Conference
on Machine Learning, vol. 2, pp. 704–711, August 2003.
[27] X. Su, R. Greiner, T. M. Khoshgoftaar and X. Zhu, “Hybrid collaborative filtering algorithms using
a mixture of experts,” IEEE/WIC/ACM International Conference on Web Intelligence, pp. 645–649,
November 2007.
[28] J. Wang, A. P. de Vries, and M. J. T. Reinders, “Unified relevance models for rating prediction in
collaborative filtering,” ACM Transactions on Information Systems, vol. 26, no. 3, pp. 1–42, June
2008.
[29] W. K. Leung, S. C. F. Chan, and F. L. Chung, “A collaborative filtering framework based on fuzzy
association rules and multi-level similarity,”Knowledge and Information Systems, vol. 10, no. 3, pp.
357–381, 2006.
[30] D. Y. Pavlov and D. M. Pennock, “A maximum entropy approach to collaborative filtering in
dynamic, sparse, high-dimensional domains,” Neural Information Processing Systems, pp. 1441–
1448, MIT Press, 2002.
[31] GeminasAdomavicius, Ramesh Sankaranarayanan, Shahana Sen and Alexander Tuzhilin,
“Incorporating Contextual Information in Recommender Systems Using a Multidimensional
Approach”, ACM Transactions on Information Systems, Vol. 23, No. 1, pp. 103 – 145, January
2005.
[32] Sarabjot Singh Anand and BamshadMobasher -From Web to Social Web: Discovering and
Deploying User and Content Profiles - Contextual Recommendation,Springer-Verlag Berlin,
Heidelberg, 2007.
[33] Han-Saem Park, Ji-Oh Yoo, and Sung-Bae Cho, “A Context-Aware Music Recommendation
System Using Fuzzy Bayesian Networks with Utility Theory”, 3rd
International Conference on
Fuzzy Systems and Knowledge Discovery, pp. 970-979, September 2006.
[34] Norma Saiph Savage, MaciejBaranski, Norma Elva Chavez and Tobias Höllerer, “Advances in
Location-Based Services I’m feeling LoCo: A Location Based Context Aware Recommendation
System”, Springer, USA, 2011.
[35] Victor Codina, Francesco Ricci and Luigi Ceccaroni, “Distributional Semantic Pre-filtering in
Context-Aware Recommender Systems”, 3rd
ACM Conference on Recommender Systems, pp. 265 –
268, 2015.
[36] B. M. Sarwar, G. Karypis, J. A. Konstan, and J. Riedl, “Itembased collaborative filtering
recommendation algorithms,” in Proceedings of the 10th International Conference on World Wide
Web (WWW ’01), pp. 285–295, May 2001.
[37] Xiaoyuan Su and Taghi M. Khoshgoftaar,” A Survey of Collaborative Filtering Techniques,’’
Department of Computer Science and Engineering, Florida Atlantic University, 777 Glades Road,
Boca Raton, FL 33431, USA, August 2009.
[38] J. Breese, D. Heckerman, and C. Kadie, “Empirical analysis of predictive algorithms for
collaborative filtering,” in Proceedings of the 14th Conference on Uncertainty in Artificial
Intelligence (UAI ’98), 1998.
[39] C. Basu, H. Hirsh, and W. Cohen, “Recommendation as classification: using social and content-
based information in recommendation,” in Proceedings of the 15th National Conference on
Artificial Intelligence (AAAI ’98), pp. 714–720, Madison, Wis, USA, July 1998.
[40] G. Adomavicius and A. Tuzhilin, “Toward the next generation of recommender systems: A survey
of the state-of-the-artand possible extensions,” IEEE Trans. on Knowledge and Data Eng., vol. 17,
pp. 734–749, June 2005.

A Literature Survey on Recommendation System Based on Sentimental Analysis

More Related Content

Similar to A Literature Survey on Recommendation System Based on Sentimental Analysis (20)

Recently uploaded (20)

A Literature Survey on Recommendation System Based on Sentimental Analysis