Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

Query Recommendations and its Evaluation in Web Information Retrieval


Affiliations
1 Department of Computer Technology, Kongunadu Arts and Science College, India
2 Department of Computer Applications, Hindustan College of Arts and Science, India
     

   Subscribe/Renew Journal


Search Engine retrieves significant and essential information from the web based on query terms given by the user. Due to the lack of background knowledge about the information required, shorter length queries posed by the user, the ambiguity of query keywords and dynamic growth of the web, irrelevant and redundant results are also retrieved by the search engine. Query recommendations is an important technique which analyze the real search intent of the user and suggests the alternative queries to be used by the user in future to satisfies their information need. The proposed method recommends and ranks the alternative queries and evaluates the ranking order of the recommendations with the help of the ranking measures Normalized Discounted Cumulative Gain (NDCG) and Coefficient of Variance (CV). These measures identify the relationship between the ranking techniques. The proposed strategies are experimentally evaluated using a real time search engine query log.

Keywords

Queries, PrefixSpan, NDCG, CV, Kappa Measure.
Subscription Login to verify subscription
User
Notifications
Font Size

Abstract Views: 216

PDF Views: 0




  • Query Recommendations and its Evaluation in Web Information Retrieval

Abstract Views: 216  |  PDF Views: 0

Authors

R. Umagandhi
Department of Computer Technology, Kongunadu Arts and Science College, India
A. V. Senthil Kumar
Department of Computer Applications, Hindustan College of Arts and Science, India

Abstract


Search Engine retrieves significant and essential information from the web based on query terms given by the user. Due to the lack of background knowledge about the information required, shorter length queries posed by the user, the ambiguity of query keywords and dynamic growth of the web, irrelevant and redundant results are also retrieved by the search engine. Query recommendations is an important technique which analyze the real search intent of the user and suggests the alternative queries to be used by the user in future to satisfies their information need. The proposed method recommends and ranks the alternative queries and evaluates the ranking order of the recommendations with the help of the ranking measures Normalized Discounted Cumulative Gain (NDCG) and Coefficient of Variance (CV). These measures identify the relationship between the ranking techniques. The proposed strategies are experimentally evaluated using a real time search engine query log.

Keywords


Queries, PrefixSpan, NDCG, CV, Kappa Measure.