Title :
Incorporating Temporal Information for Document Classification
Author :
Luo, Xiao ; Zincir-Heywood, Nur
Author_Institution :
Dalhousie Univ., Halifax
Abstract :
In this paper, we propose a novel document classification system where the Recurrent Linear Genetic Programming is employed to classify documents that are represented in encoded word sequences by Self Organizing feature Maps. The results using different feature selection techniques on Reuters 21578 data set show that the proposed system can analyze the temporal sequence patterns of a document and achieve competitive performance on classification.
Keywords :
genetic algorithms; pattern classification; self-organising feature maps; text analysis; document classification system; encoded word sequences; feature selection techniques; recurrent linear genetic programming; self organizing feature maps; temporal information; temporal sequence patterns; Computer science; Encoding; Genetic programming; Kernel; Pattern analysis; Performance analysis; Support vector machine classification; Support vector machines; Text analysis; Text categorization;
Conference_Titel :
Data Engineering Workshop, 2007 IEEE 23rd International Conference on
Conference_Location :
Istanbul
Print_ISBN :
978-1-4244-0832-0
Electronic_ISBN :
978-1-4244-0832-0
DOI :
10.1109/ICDEW.2007.4401067
