Performance of question-based vs keyword-based search engines and effect of web user characteristics on search engine performance

Abstract

Purpose - The purpose of this study is to investigate whether question and keyword-format queries are more successfully processed by search engines encouraging answers to searching and keyword-format querying, respectively. This study aims to investigate whether web user characteristics and choice of search engine affects the relevancy scores and precision of the results. Design/methodology/approach - The results of two search engines, Google and AskJeeves, were compared for question and keyword-format queries. It was observed that AskJeeves was slightly more successful in processing question-format queries, but this finding was not statistically supported. However, Google provided results on keyword-format queries and the entire set of queries, which were statistically superior to those of AskJeeves. Findings - Analysis of variance (ANOVA) showed that the age of web user is not as affective on the relevancy score and precision of results as other factors. Interactions of the main factors were also affective on the relevancy scores and precision, meaning that the different combinations of various factors cause a synergy in terms of relevancy scores and precision. Research limitations/implications - This was a preliminary work on the effect of user characteristics on comprehension and evaluation of search query results. Future work includes expanding this study to include more web user characteristics, more levels of the web user characteristics, and inclusion of more search engines. Originality/value - The findings of this study provide statistical proof for the relationship between the characteristics of web users, choice of search engine and the relevancy scores and precision of search results.