Mandarin Chinese broadcast news retrieval and summarization using probabilistic generative models

This paper presents our recent research work on applying probabilistic generative models to Mandarin Chinese broadcast news retrieval and summarization. Most models can be trained in either a supervised or unsupervised manner. In addition, both literal term matching and concept matching strategies have been intensively investigated. This paper also presents a prototype web-based Mandarin Chinese broadcast news retrieval system, which is based on technologies such as automatic story segmentation, automatic speech recognition, spoken document retrieval and summarization.