An evaluation of relevancy ranking techniques used by internet search engines


  • Jonathan Back



relevancy ranking, algorithm, retrived information, internet search engines, search techniques, comparative evaluation, accuracy


A relevancy ranking algorithm aims to sort retrieved information resources so that those most likely to be relevant are shown flrst. Experimentation reveals that ranking techniques employed by lnternet search engines do not facilitate effective retrieval. The methodology adopted allows for a comparative evaluation of rival search engines. Additionally, different search techniques utilised by experiment participants were analysed. Results show that if more than four search terms are used,the accuracy of a relevancy ranking will increase.

Author Biography

Jonathan Back

Jonathan Back is now a full-time Ph.D. research student at Loughborough University, Loughborough, Leicestershire, LEI I 3TU. His current research involves investigating how to capture'user relevance feedback' in order to improveWeb retrieval performance. The 1998/99 Undergraduate Prize was awarded to Jonathan for his final year project, "An Evaluation of Relevancy Ranking Techniques used by lnternet Search Engines".