Improving Automatic Query Classification via Semi-Supervised Learning
ICDM '05 Proceedings of the Fifth IEEE International Conference on Data Mining
Web Search: Public Searching of the Web (Information Science and Knowledge Management)
Web Search: Public Searching of the Web (Information Science and Knowledge Management)
Presentation bias is significant in determining user preference for search results—A user study
Journal of the American Society for Information Science and Technology
News vertical search: when and what to display to users
Proceedings of the 36th international ACM SIGIR conference on Research and development in information retrieval
Hi-index | 0.00 |
The analysis of search engine logs is important in order to understand how users interact with a search engine. Conventional analysis of search engine log data looks at various metrics such as query and session length aggregated over the full data set. Here we segment the data according to a top-level ontology of web search and compute the metrics on a topic by topic basis. Our results show that although for a given metric, such as query length, the statistics of most classes are similar to the aggregate statistic; there are usually some outlier classes which exhibit deviant behavior.