作者: Valerie J. Gillet , Andrew D. Madden , Nigel Ford , Barry Eaglestone , Martin Whittle
DOI:
关键词:
摘要: Introduction. This paper reports preliminary research in a primarily experimental study of how the general public search for information on Web. The focus is query transformation patterns that characterise searching. Method. In this work, we have used transaction logs from Excite engine to develop methods analysing transformations should aid analysis our ongoing work. Our involve use similarity techniques link queries with most similar previous train. resulting are represented as list codes representing whole search. Analysis. It shown sequences can be graphical networks and some basic statistical results shown. A correlation performed examine co-occurrence Boolean quotation mark changes syntactic changes. Results. frequency occurrence presented. connectivity graphs obtained investigated found follow an exponential scaling law. reveals number provide interesting insights into Web searching by public. Conclusion. We developed analytical based applied current work volunteer subjects. these will form part database aim developing improved understanding