ECIR 2016 Paper Beyond Factoid QA: Effective Methods for Non-factoid Answer Sentence Retrieval

中文简介：本文对non-factoid 问题的答案句子检索进行了研究，基于learning to ranking的框架，在传统文本匹配特征的基础上提出了给予语义匹配和上下文信息的特征，并且证明了这些特征对于答案句子检索的有效性。本文使用TREC GOV2数据集，并且开源了code和标注数据集，下载链接参见论文脚注。

论文出处：ECIR'16.

英文摘要： Retrieving finer grained text units such as passages or sentences as answers for non-factoid Web queries is becoming increasingly important for applications such as mobile Web search. In this work, we introduce the answer sentence retrieval task for non-factoid Web queries, and investigate how this task can be effectively solved under a learning to rank framework. We design two types of features, namely semantic and context features, beyond traditional text matching features. We compare learning to rank methods with multiple baseline methods including query likelihood and the state-of-the-art convolutional neural network based method, using
an answer-annotated version of the TREC GOV2 collection. Results show that features used previously to retrieve topical sentences and factoid answer sentences are not sufficient for retrieving answer sentences for non-factoid queries, but with semantic and context features, we can significantly outperform the baseline methods.

下载链接：https://yangliuy.github.io/publication/2016-03-20-ecir-proactive-ranking

开源Code Github链接：http://rmit-ir.github.io/SummaryRank/

Data 链接：https://ciir.cs.umass.edu/downloads/WebAP/