Semantic Chunk Annotation for complex questions using Conditional Random Field

This paper presents a CRF (Conditional Random Field) model for Semantic Chunk Annotation in a Chinese Question and Answering System (SCACQA). The model was derived from a corpus of real world questions, which are collected from some discussion groups on the Internet. The questions are supposed to be answered by other people, so some of the questions are very complex. Mutual information was adopted for feature selection. The training data collection consists of 14000 sentences and the testing data collection consists of 4000 sentences. The result shows an F-score of 93.07%. © 2008. Licensed under the Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported license (http://creativecommons.org/licenses/by-ncsa/3.0/). Some rights reserved.