Browsing by Author "Vahab, Febin A"

Now showing 1 - 2 of 2

ISD-An Intelligent Service Desk
(CIIT, 2013) Vahab, Febin A
A knowledge base is where an organization documents communal knowledge that the teams are acquiring through hard experience. Customers are the main reason an organization need to use a knowledge base. The turn-around time of the query resolution and correctness is of utmost importance. It is also important to be able to retain the knowledge the employees acquire, rather than letting it walk out the door with them when they eventually move on to another job. The information in a knowledge base can be used to solve the issues which were earlier solved with customer representative help. Many companies use text based or case-based service desk systems to improve customer service quality. But most of the existing knowledge base systems use matching based on the keywords in the cases and rank the cases based on those keyword matches. This method of case retrieval in inefficient and has difficulty in understanding the exact meanings of the cases. The results based on keyword-based retrieval, are inaccurate and incomplete in cases where different keywords are used for the description of similar concepts in artifacts and queries. To address this challenge, ISD, an Intelligent Service Desk, is proposed, to find problem–solution patterns from the past customer–representative interactions automatically. The main aim of the paper is to bring in semantic analysis of the cases in case retrieval. When a new query from the customer arrives, ISD searches the previous cases in the knowledge base and ranks it based on the semantic relevance of the incoming request and the knowledge base cases. A new way is formulated to understand the semantic meanings of the cases. This method can be used to trance the exact meanings of the cases. The proposed system uses tokenization to remove the stop words, part of speech tagging, word sense disambiguation and finally a path length based similarity measurement to capture the semantic similarity between the sentences. ISD calculates a score for the sentence searched for and the reference solutions in the knowledge base using the proposed method and displays the results in the decreasing order of the score. The experimental result and case studies presented in the paper show that the proposed method has high precision of retrieval when compared to case based systems.
Predicting Post Importance in Question Answer Forums Based on Topic-Wise User Expertise
(Springer, 2015) Vahab, Febin A
Q & A forums on the web are aplenty and the content produced through such crowd-sourced efforts is generally of good quality and highly beneficial to novices and experts alike. As the community matures, however, the explosion in the number of posts/answers leads to the information overload problem. Many a times users having expertise in a particular area are not able to address quality issues raised in the area maybe due to the positioning of the question in the list displayed to the user. A good mechanism to assess the quality of questions and to display it to the users depending on their area of expertise, if devised, may lead to a higher quality answers and faster resolutions to the questions posted. In this paper we present the results of our investigations into the effectiveness of various mechanisms to represent user expertise to estimate a post score reflecting its quality/utility of the post. We follow three different approaches to building a user profile representing the user’s areas of expertise: topic models based approach, tag-based approach and semantic user profiling approaches. We present the results of experiments performed on the popular Q&A Forum Stack Overflow, exploring the value add offered by these approaches. The preliminary experiments support our hypothesis that considering additional features in terms of user expertise does offer an increase in the classification accuracy even while ignoring features computable only after the first 24 hours. However, the proposed method to individually leverage on the semantic tag relations to construct an enhanced user profile did not prove beneficial.