Learning Question Classifiers

Experimental Data for Question Classification

This data collection contains all the data used in our learning question classification experiments(see [1]), which has question class definitions, the training and testing question sets, examples of preprocessing the questions, feature definition scripts and examples of semantically related word features. This work has been done by Xin Li and Dan Roth and supported by [2].

Question Classification Taxonomy
Training and Test sets

Training set 1(1000 labeled questions)
Training set 2(2000 labeled questions)
Training set 3(3000 labeled questions)
Training set 4(4000 labeled questions)
Training set 5(5500 labeled questions)
Test set: TREC 10 questions

Examples of Preprocessing
Script of Feature description
Examples of semantically related word lists

Profession
Mountain
Food

All the data for extracting features by Fex

[1] Xin Li, Dan Roth, Learning Question Classifiers. COLING'02, Aug., 2002.

[2] Research supported by NSF grants IIS-9801638 and ITR IIS-0085836 and an ONR MURI Award.

Questions or comments can be directed to xli1@uiuc.edu