Cognitive-linguistics-based request answer system
AMR'09 Proceedings of the 7th international conference on Adaptive multimedia retrieval: understanding media and adapting to the user
Query classification using Wikipedia
International Journal of Intelligent Information and Database Systems
Exploring wikipedia's category graph for query classification
AIS'11 Proceedings of the Second international conference on Autonomous and intelligent systems
| Hi-index | 0.00 |
In the paper, we introduce a music knowledge question answering system based on ontology knowledge base. Users can propose a question about music knowledge in natural language, the system automatically extract relative knowledge to form answer based on FAQ (Frequently Asked Questions) and ontology knowledge base, and return the answer to users in natural language. The important modules in the question answering system are introduced in the paper. The domain ontology is base stone of the system. The improved Bayesian method based on ontology is adopted to classify the questions. The questions are located by index and search technology in FAQ firstly; it can rapidly locate the question and return answer. For question outside of FAQ, semantic meaning of questions need to be analyze, the question semantic representation method is adopted in the process. Two strategies are adopted to extract answer knowledge, and return the answer in natural language. Experimental results show that the system is effective in answering the question in music knowledge domain.