A DYNAMIC FEATURE SELECTION METHOD FOR DOCUMENT RANKING WITH RELEVANCE FEEDBACK APPROACH
Journal Title: ICTACT Journal on Soft Computing - Year 2010, Vol 1, Issue 1
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.
Authors and Affiliations
Latha K, Bhargavi B, Dharani C, Rajaram R
Keywords
Related Articles
- EP ID EP198963
- DOI 10.21917/ijsc.2010.0001
- Views 70
- Downloads 0
A HIGH QUALITY EMBEDDED SYSTEM FOR ASSESSING FOOD QUALITY USING HISTOGRAM OF ORIENTED GRADIENTS
A low cost high quality system for accessing quality of food samples by finding the presence of fungus is proposed. Most of the food items kept for long intervals will have fungal infection in them. The proposed system u...
AN IMPLEMENTATION OF EIS-SVM CLASSIFIER USING RESEARCH ARTICLES FOR TEXT CLASSIFICATION
Automatic text classification is a prominent research topic in text mining. The text pre-processing is a major role in text classifier. The efficiency of pre-processing techniques is increasing the performance of text cl...
RICH SEMANTIC SENTIMENT ANALYSIS USING LEXICON BASED APPROACH
Web is a huge repository of information, and a massive amount of data is generated everyday on online platforms. Information, can be facts and opinions, facts are objective statements about an event, and opinions are sub...
A METHOD FOR FORECASTING WEATHER CONDITION BY USING ARTIFICIAL NEURAL NETWORK ALGORITHM
This article presents a method to forecast and make decision on weather condition. In most of the cities around the world, people try to decide on leisure activities on their spare time but weather condition would not be...
MISSING VALUE IMPUTATION AND NORMALIZATION TECHNIQUES IN MYOCARDIAL INFARCTION
Missing Data imputation is an important research topic in data mining. In general, real data contains missing values. The presence of the missing value in the data set has a major problem for precise prediction. The obje...