An Approach to Text Topic Identification
Jing Ju Zhu · Journal of Northeastern University · 2002
A knowledge based approach to text topic identification was presented. The approach applies statistic based and knowledge based techniques in topic feature identification. The aggregation formulas are used in topic feature aggregation. Domain knowledge base is used to upgrade topic identification technique from lexical analysis to domain knowledge analysis. Experiment results show that the average accuracy of topic identification reaches 70% by this approach.