Enhance web search results using user feedback sessions

Abstract

Most search engines in use today present the user a single ordered list of documents matching the search query. Since most users do not browse past the first page of search results, this method of displaying results often limits the effectiveness of the search to the relevance of the first several documents. An alternative to a single ordered list is to cluster the search results in that cluster are then displayed in a list. Under the assumption that documents which are similar to each other are likely to be relevant to the same query, the clustering of search results are easier to browse than a single ordered list. The inference and analysis of user search goals for a query can be very useful in improving search engine relevance and user experience. A novel approach is used to infer the user search goals by analyzing the user click through logs. So we first propose a framework to find out different user search goals for a query by clustering the proposed feedback sessions. The Feedback sessions are constructed from user click-through data which can efficiently reflect the information needs of users. Then a novel approach is used to generate the pseudo documents by considering the feedback sessions. The pseudodocuments are clustered using the suffix tree clustering algorithm to restructure the search results. In order to evaluate the restructured results, we use a method called “Classified Average Precision (CAP)” to measure the performance of inferring user search goals.

Authors and Affiliations

S. R. Sri Abirami, Dr. C. Nalini, A. P. Ponselvakumar

Keywords

Related Articles

Synthesis and Characterization of Some Nicotinic Acid Complexes

The research deals with the preparation of some complexes of transition and non transition elements of Co(II), Ni(II), Cu(II),Zn(II), Cd(II) and uranium (VI) with nicotinic acid. The synthesized complexes were character...

Vehicular Handover Approach Based On LTE In Urban Road Scenario

The development of wireless technology has great impacted on various phases of communication system. In vehicular system the use of wireless technology is taking place at rapid pace. Intelligent Transport systems are ev...

Adaptive Cloud Downloading Service Using AMS Algorithm

Video content downloading using the P2P approach is scalable, but does not always give good performance. Recently, subscription-based premium services have emerged, referred to as cloud downloading. In this service, the...

Prediction of Corrosion Rates in Structural Steel Using Artificial Neural Networks

A phenomenal outcome for the prediction of corrosion in steel was proposed with the learning ability of artificial neural network using MATLAB software. The prediction of corrosion rate has become an important challenge...

Study about the Behaviour of Concrete Filled Tubular Column with GFRP Wrapping

In this paper, the tubular columns are filled with normal and replacement concrete under M25 grade. The replacement involves 5 % of metakaolin replaced for cement and 25 % of green sand replaced for fine aggregate. The...

Download PDF file
  • EP ID EP20254
  • DOI -
  • Views 255
  • Downloads 4

How To Cite

S. R. Sri Abirami, Dr. C. Nalini, A. P. Ponselvakumar (2015). Enhance web search results using user feedback sessions. International Journal for Research in Applied Science and Engineering Technology (IJRASET), 3(4), -. https://europub.co.uk./articles/-A-20254