Sheetal Girase

Identification of topic-specific Opinion Leader using SPEAR algorithm in Online Knowledge communities

Currently Internet usage has increased a lot due to bandwidth availaility and technology advancements. Internet is widely used for knowledge sharing, online review of products etc. Many open forums, blogs are used for this purpose. Since many users are contributing their opinions towards any query submitted by information seeker, there is a possibility of confusion. Often opinions contradict with each other creating confusion in information seekers mind. In these cases role of Opinion Leader(s) is very prominent. Opinion Leader is a person who has knowledge in the particular field, whos opinion makes difference and who can influence others opinions. Identification of a person who has great experiences and/or knowledge, in a particular domain, is very helpful and useful in decision making, product marketing etc. This paper presents an approach for identification of Opinion Leader(s) using modified SPEAR (Spamming Resistant Expertise Analysis and Ranking) algorithm. The expertise of user is found out on different topics. Modified SPEAR algorithm effectively identifies Opinion Leader(s) by making use of additional influence measures in the form of credit score functions. It also analyses these measures and studies their effects while ranking the Opinion Leader(s) effectively.

Performance analysis of classification and ranking techniques

Recommendation systems aim at recommending relevant items to the users of the system. Recommendation Systems provide efficient recommendations based on algorithms used for classification and ranking. There exist various ways by which classification can be achieved in a supervised or unsupervised manner. Since the sample datasets that are used for experiments are large and also contain more number of feature sets, it is essential to understand dataset beforehand. Also when results are shown to the user, big challenge is how well data can be ranked so that user satisfaction is guaranteed. When data sets are large, some ranking algorithms perform poorly in terms of computation and storage. Thus, these kinds of algorithms are quite expensive. We aim at developing classification and ranking algorithm which will reduce computational cost and dimensionality of data without affecting the diversity of the feature set. Dimensionality of data can be handled by SVM (Support Vector Machine). AUC (Area under the Curve) and WARP (Weighted Approximately Ranked Pairwise) algorithms are efficient for ranking of the items which are of user interest.