Semi-supervised Learning of Domain-Specific Language Models from General Domain Data

We present a semi-supervised learning method for building domain-specific language models (LM) from general-domain data. This method is aimed to use small amount of domain-specific data as seeds to tap domain-specific resources residing in larger amount of general-domain data with the help of topic modeling technologies. The proposed algorithm first performs topic decomposition (TD) on the combined dataset of domain-specific and general-domain data using probabilistic latent semantic analysis (PLSA). Then it derives domain-specific word n-gram counts with mixture modeling scheme of PLSA. Finally, it uses traditional n-gram modeling approach to construct domain-specific LMs from the domain-specific word n-gram counts. Experimental results show that this approach can outperform both stat-of-the-art methods and the simulated supervised learning method with our data sets. In particular, the semi-supervised learning method can achieve better performance even with very small amount of domain-specific data.