Simple Question Answering by Attentive Convolutional Neural Network

TL;DR

This paper introduces an effective approach for simple question answering over Freebase using attentive convolutional neural networks, improving entity linking and predicate matching for better accuracy.

Contribution

It proposes a new attentive maxpooling technique over CNNs and a simple entity linker, achieving state-of-the-art results in simple question answering.

Findings

Outperforms previous entity linkers on SimpleQA

Achieves new state-of-the-art accuracy in fact selection

Attentive maxpooling improves predicate-question matching

Abstract

This work focuses on answering single-relation factoid questions over Freebase. Each question can acquire the answer from a single fact of form (subject, predicate, object) in Freebase. This task, simple question answering (SimpleQA), can be addressed via a two-step pipeline: entity linking and fact selection. In fact selection, we match the subject entity in a fact candidate with the entity mention in the question by a character-level convolutional neural network (char-CNN), and match the predicate in that fact with the question by a word-level CNN (word-CNN). This work makes two main contributions. (i) A simple and effective entity linker over Freebase is proposed. Our entity linker outperforms the state-of-the-art entity linker over SimpleQA task. (ii) A novel attentive maxpooling is stacked over word-CNN, so that the predicate representation can be matched with the predicate-focused question representation more effectively. Experiments show that our system sets new state-of-the-art in this task.