ICTACT Journal on Soft Computing

K. Latha ¹, R. Rajaram ²

Affiliations
1 Department of Computer Science and Engineering, Anna University of Technology, Tiruchirappalli, IN
2 Department of Information Technology, Thiagarajar College of Engineering, Tamil Nadu, IN

Abstract

One popular Community question answering (CQA) site, Yahoo! Answers, had attracted 120 million users worldwide, and had 400 million answers to questions available. A typical characteristic of such sites is that they allow anyone to post or answer any questions on any subject. Question Answering Community has emerged as popular, and often effective, means of information seeking on the web. By posting questions, for other participants to answer, information seekers can obtain specific answers to their questions. However, CQA is not always effective: in some cases, a user may obtain a perfect answer within minutes, and in others it may require hours and sometimes days until a satisfactory answer is contributed. We investigate the problem of predicting information seeker satisfaction in yahoo collaborative question answering communities, where we attempt to predict whether a question author will be satisfied with the answers submitted by the community participants. Our experimental results, obtained from a large scale evaluation over thousands of real questions and user ratings, demonstrate the feasibility of modeling and predicting asker satisfaction. We complement our results with a thorough investigation of the interactions and information seeking patterns in question answering communities that correlate with information seeker satisfaction. We also explore automatic ranking, creating abstract from retrieved answers, and history updation, which aims to provide users with what they want or need without explicitly ask them for user satisfaction. Our system could be useful for a variety of applications, such as answer selection, user feedback analysis, and ranking.

Keywords

Social Media, Community Question Answering, Information Seeker Satisfaction, Ranking, History Updation.

Full Text

     

K. Latha ¹, P. Ramya ¹, V. Sita ¹, R. Rajaram ²

Affiliations
1 Department of Computer Science and Engineering, Anna University, Tiruchirappalli, IN
2 Department of Information Technology, Thiagarajar College of Engineering, Madurai, IN

Abstract

Web browsers are provided with complex information space where the volume of information available to them is huge. There comes the Recommender system which effectively recommends web pages that are related to the current webpage, to provide the user with further customized reading material. To enhance the performance of the recommender systems, we include an elegant proposed web based recommendation system; Truth Discovery based Content and Collaborative RECommender (TDCCREC) which is capable of addressing scalability. Existing approaches such as Learning automata deals with usage and navigational patterns of users. On the other hand, Weighted Association Rule is applied for recommending web pages by assigning weights to each page in all the transactions. Both of them have their own disadvantages. The websites recommended by the search engines have no guarantee for information correctness and often delivers conflicting information. To solve them, content based filtering and collaborative filtering techniques are introduced for recommending web pages to the active user along with the trustworthiness of the website and confidence of facts which outperforms the existing methods. Our results show how the proposed recommender system performs better in predicting the next request of web users.

Keywords

Recommendation, Content, Collaborative Filtering, Learning Automata, Navigation.

Full Text

     

K. Latha ¹, B. Bhargavi ¹, C. Dharani ¹, R. Rajaram ²

Affiliations
1 Department of Computer Science and Engineering, Anna University of Technology, Tiruchirappalli, Tamil Nadu, IN
2 Department of Information Technology, Thiagarajar College of Engineering, Madurai, Tamil Nadu, IN

Abstract

Ranking search results is essential for information retrieval and Web search. Search engines need to not only return highly relevant results, but also be fast to satisfy users. As a result, not all available features can be used for ranking, and in fact only a small percentage of these features can be used. Thus, it is crucial to have a feature selection mechanism that can find a subset of features that both meets latency requirements and achieves high relevance. In this paper we describe a 0/1 knapsack procedure for automatically selecting features to use within Generalization model for Document Ranking. We propose an approach for Relevance Feedback using Expectation Maximization method and evaluate the algorithm on the TREC Collection for describing classes of feedback textual information retrieval features. Experimental results, evaluated on standard TREC-9 part of the OHSUMED collections, show that our feature selection algorithm produces models that are either significantly more effective than, or equally effective as, models such as Markov Random Field model, Correlation Co-efficient and Count Difference method.

Keywords

Feature Selection, Expectation Maximization, Markov Random Field, Generalization, Document Ranking.

Full Text