Experiments on multistrategy learning by meta-learning
CIKM '93 Proceedings of the second international conference on Information and knowledge management
Scaling question answering to the Web
Proceedings of the 10th international conference on World Wide Web
On the MSE robustness of batching estimators
Proceedings of the 33nd conference on Winter simulation
Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data
ICML '01 Proceedings of the Eighteenth International Conference on Machine Learning
Relational Learning for NLP using Linear Threshold Elements
IJCAI '99 Proceedings of the Sixteenth International Joint Conference on Artificial Intelligence
Question classification using support vector machines
Proceedings of the 26th annual international ACM SIGIR conference on Research and development in informaion retrieval
Web-scale information extraction in knowitall: (preliminary results)
Proceedings of the 13th international conference on World Wide Web
Is question answering an acquired skill?
Proceedings of the 13th international conference on World Wide Web
COLING '02 Proceedings of the 19th international conference on Computational linguistics - Volume 1
Shallow parsing with conditional random fields
NAACL '03 Proceedings of the 2003 Conference of the North American Chapter of the Association for Computational Linguistics on Human Language Technology - Volume 1
Question classification with support vector machines and error correcting codes
NAACL-Short '03 Proceedings of the 2003 Conference of the North American Chapter of the Association for Computational Linguistics on Human Language Technology: companion volume of the Proceedings of HLT-NAACL 2003--short papers - Volume 2
Accurate unlexicalized parsing
ACL '03 Proceedings of the 41st Annual Meeting on Association for Computational Linguistics - Volume 1
AnswerBus question answering system
HLT '02 Proceedings of the second international conference on Human Language Technology Research
Optimizing scoring functions and indexes for proximity search in type-annotated corpora
Proceedings of the 15th international conference on World Wide Web
Tamil Question Classification Using Morpheme Features
GoTAL '08 Proceedings of the 6th international conference on Advances in Natural Language Processing
Question classification using head words and their hypernyms
EMNLP '08 Proceedings of the Conference on Empirical Methods in Natural Language Processing
Semantically Expanding Questions for Supervised Automatic Classification
FQAS '09 Proceedings of the 8th International Conference on Flexible Query Answering Systems
Investigation of question classifier in question answering
EMNLP '09 Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing: Volume 2 - Volume 2
From symbolic to sub-symbolic information in question classification
Artificial Intelligence Review
Enhanced semantic expansion for question classification
International Journal of Internet Technology and Secured Transactions
Question type classification using a part-of-speech hierarchy
AIS'11 Proceedings of the Second international conference on Autonomous and intelligent systems
A support vector machine-based context-ranking model for question answering
Information Sciences: an International Journal
Locational relativity and domain constraints in spatial questions
Proceedings of the 20th International Conference on Advances in Geographic Information Systems
What is it we are asking: interpreting problem-solving questions in computer science and linguistics
Proceeding of the 44th ACM technical symposium on Computer science education
Hi-index | 0.00 |
Question classification is an important step in factual question answering (QA) and other dialog systems. Several attempts have been made to apply statistical machine learning approaches, including Support Vector Machines (SVMs) with sophisticated features and kernels. Curiously, the payoff beyond a simple bag-of-words representation has been small. We show that most questions reveal their class through a short contiguous token subsequence, which we call its informer span. Perfect knowledge of informer spans can enhance accuracy from 79.4% to 88% using linear SVMs on standard benchmarks. In contrast, standard heuristics based on shallow pattern-matching give only a 3% improvement, showing that the notion of an informer is non-trivial. Using a novel multi-resolution encoding of the question's parse tree, we induce a Conditional Random Field (CRF) to identify informer spans with about 85% accuracy. Then we build a meta-classifier using a linear SVM on the CRF output, enhancing accuracy to 86.2%, which is better than all published numbers.