Ozumutlu, H.C.; Cavdur, F.: ¬Application of automatic topic identification on Excite Web search engine data logs (2005)
0.02
0.018582305 = product of:
0.03716461 = sum of:
0.03716461 = product of:
0.07432922 = sum of:
0.07432922 = weight(_text_:web in 1047) [ClassicSimilarity], result of:
0.07432922 = score(doc=1047,freq=6.0), product of:
0.17002425 = queryWeight, product of:
3.2635105 = idf(docFreq=4597, maxDocs=44218)
0.052098576 = queryNorm
0.43716836 = fieldWeight in 1047, product of:
2.4494898 = tf(freq=6.0), with freq of:
6.0 = termFreq=6.0
3.2635105 = idf(docFreq=4597, maxDocs=44218)
0.0546875 = fieldNorm(doc=1047)
0.5 = coord(1/2)
0.5 = coord(1/2)
- Abstract
- The analysis of contextual information in search engine query logs enhances the understanding of Web users' search patterns. Obtaining contextual information on Web search engine logs is a difficult task, since users submit few number of queries, and search multiple topics. Identification of topic changes within a search session is an important branch of search engine user behavior analysis. The purpose of this study is to investigate the properties of a specific topic identification methodology in detail, and to test its validity. The topic identification algorithm's performance becomes doubtful in various cases. These cases are explored and the reasons underlying the inconsistent performance of automatic topic identification are investigated with statistical analysis and experimental design techniques.
