What question did this study set out to answer?

This research aims to improve the process of ranking candidate answers for extraction by exploring the correlation of dependency relation paths.

January 1, 2006Open Access

Exploring correlation of dependency relation paths for answer extraction

View Full Paper

Ask AI

Bookmark

Authors

Randomized trial explores correlation of dependency relation paths for answer extraction, suggesting enhanced ranking performance.

This research aims to improve the process of ranking candidate answers for extraction by exploring the correlation of dependency relation paths.
Proposed an approximate phrase mapping algorithm to improve comparison accuracy.
Incorporated mapping scores into a correlation measure for better candidate ranking.
Used a Maximum Entropy-based ranking model to estimate path weights based on training data.
The proposed method shows a significant improvement, outperforming state-of-the-art methods by up to 20% in mean reciprocal rank (MRR).

Shen et al. (2006) studied this question.

View Full Paper

Ask AI

Bookmark