Title :
Study on Topic Tracking System Based on SVM
Author :
Li, Shuping ; Zhao, Jie ; Song, Zhichao ; Li, Shengdong
Author_Institution :
Dept. of Comput. Sci., Mudanjiang Normal Univ., Mudanjiang, China
Abstract :
Text classification is the key technology for topic tracking, and vector space model (VSM) is one of the most simple and effective model for topics representation. Feature selection algorithm in VSM is an important means of data pre-processing, and it can reduce vector space dimension and improve the generalization ability of the algorithm. So we develop a topic tracking system based on SVM to study how feature dimension and the value of K-neighbors affect topic tracking. Then we get the variation law that they affect topic tracking, and add up their optimal values in topic tracking. And finally we analyze topic tracking system performance according to TDT evaluation results to find the reasons for this results.
Keywords :
generalisation (artificial intelligence); pattern classification; support vector machines; text analysis; data preprocessing; feature dimension; feature selection algorithm; generalization; support vector machines; text classification; topic representation; topic tracking system; vector space model; Classification algorithms; Computers; Educational institutions; Support vector machine classification; Text categorization; Training; svm; tdt evaluation; topic tracking; weight of evidence for text;
Conference_Titel :
Knowledge Acquisition and Modeling (KAM), 2011 Fourth International Symposium on
Conference_Location :
Sanya
Print_ISBN :
978-1-4577-1788-8
DOI :
10.1109/KAM.2011.30
