𝔖 Bobbio Scriptorium
✦   LIBER   ✦

Using support vector machines for automatic new topic identification

✍ Scribed by Seda Ozmutlu; H. Cenk Ozmutlu; Amanda Spink

Publisher
Wiley (John Wiley & Sons)
Year
2008
Tongue
English
Weight
162 KB
Volume
44
Category
Article
ISSN
0044-7870

No coin nor oath required. For personal study only.

✦ Synopsis

Abstract

Recent studies on automatic new topic identification in Web search engine user sessions demonstrated that learning algorithms such as neural networks and regression have been fairly successful in automatic new topic identification. In this study, we investigate whether another learning algorithm, Support Vector Machines (SVM) are successful in terms of identifying topic shifts and continuations. Sample data logs from the Norwegian search engine FAST (currently owned by Overture) and Excite are used in this study. Findings of this study suggest that support vector machines' performance depends on the characteristics of the dataset it is applied on.

πŸ“œ SIMILAR VOLUMES

A hierarchical classifier using new supp
A hierarchical classifier using new support vector machines for automatic target recognition
✍ David Casasent; Yu-Chiang Wang πŸ“‚ Article πŸ“… 2005 πŸ› Elsevier Science 🌐 English βš– 204 KB

A binary hierarchical classifier is proposed for automatic target recognition. We also require rejection of non-object (non-target) inputs, which are not seen during training or validation, thus producing a very difficult problem. The SVRDM (support vector representation and discrimination machine)

Cross validation of neural network appli
Cross validation of neural network applications for automatic new topic identification
✍ H. Cenk Γ–zmutlu; Fatih Γ‡avdur; Amanda Spink; Seda Γ–zmutlu πŸ“‚ Article πŸ“… 2006 πŸ› Wiley (John Wiley & Sons) 🌐 English βš– 970 KB

## Abstract There are recent studies in the literature on automatic topic‐shift identification in Web search engine user sessions; however most of this work applied their topic‐shift identification algorithms on data logs from a single search engine. The purpose of this study is to provide the cros