A survey on session detection methods in query logs and a proposal for future evaluation
Information Sciences: an International Journal
Identification of factors predicting clickthrough in Web searching using neural network analysis
Journal of the American Society for Information Science and Technology
How to define searching sessions on web search engines
WebKDD'06 Proceedings of the 8th Knowledge discovery on the web international conference on Advances in web mining and web usage analysis
Identifying the optimal set of parameters for new topic identification through experimental design
Expert Systems with Applications: An International Journal
Hi-index | 0.00 |
The purpose of this study is to provide results from experiments designed to investigate the cross-validation of an artificial neural network application to automatically identify topic changes in Web search engine user sessions by using data logs of different Web search engines for training and testing the neural network. Sample data logs from the FAST and Excite search engines are used in this study. The results of the study show that identification of topic shifts and continuations on a particular Web search engine user session can be achieved with neural networks that are trained on a different Web search engine data log. Although FAST and Excite search engine users differ with respect to some user characteristics (e.g., number of queries per session, number of topics per session), the results of this study demonstrate that both search engine users display similar characteristics as they shift from one topic to another during a single search session. The key finding of this study is that a neural network that is trained on a selected data log could be universal; that is, it can be applicable on all Web search engine transaction logs regardless of the source of the training data log. © 2008 Wiley Periodicals, Inc.