An End-to-End Model for Question Answering over Knowledge Base with Cross-Attention Combining Global

An End-to-End Model for Question Answering over Knowledge Base with Cross-Attention Combining Global Knowledge

2017 ACL WQ42.9
Candidate Generation
Freebase API method is able to resolve as many as 86% questions if we use the top1 result
After getting the topic entity, we collect all the entities directly connected to it and the ones connected with 2-hop5

Overview

1）从问题当中，确认topic entity，从Freebase党章产生候选答案
2）一个交叉神经网络去表示，问题和答案
3）相似度函数，计算相似度

we employ four kinds of answer aspects:
answer entity ae
answer relation ar,
answer type at
answer context ac