Abstract—Recommender systems (RSs) provide personalised
suggestions of information or products relevant to users’ needs.
Although RSs have made substantial progresses in theory and
algorithm development and have achieved many commercial
successes, how to utilise the widely available information in
Online Social Networks (OSNs) has been largely overlooked.
Noticing such a gap in the existing research in RSs and taking
into account a user’s selection being greatly influenced by
his/her trusted friends and their opinions, this paper proposes a
framework of Implicit Social Trust and Sentiment (ISTS) based
RSs, which improves the existing recommendation approaches
by exploring a new source of data from friends’ short posts in
microbloggings as micro-reviews. The impact degree of friends’
sentiment and level being trusted to a user’s selection are
identified by using machine learning methods including Naive
Bayes, Logistic Regression and Decision Trees. As the
verification of the proposed framework, experiments using real
social data from Twitter microblogger are presented and results
show the effectiveness and promising of the proposed approach.
Index Terms—Recommender systems, machine learning,
trust, sentiment analysis, microblogging.
The authors are with School of Computer Science, The University of
Manchester, Manchester, M13 9PL, UK (e-mail:
dimah.alahmadi@postgrad.manchester.ac.uk, x.zeng@manchester.ac.uk).
[PDF]
Cite: Dimah Alahmadi and Xaio-Jun Zeng, "Improving Recommendation Using Trust and Sentiment Inference from OSNs," International Journal of Knowledge Engineering vol. 1, no. 1, pp. 9-17, 2015.
