Learning in Text Streams: Discovery and Disambiguation of Entity and Relation Instances

TL;DR

This paper introduces an end-to-end trainable memory network that enables an artificial agent to learn, disambiguate, and discover entities and relations in streaming text, improving its knowledge base through online, unsupervised learning.

Contribution

It presents a novel memory network model capable of online, one-shot learning for entity and relation discovery and disambiguation in text streams with minimal supervision.

Findings

Strong disambiguation and discovery performance on Wikipedia corpus

Effective handling of abrupt narration changes for coreference resolution

Model improves its knowledge base while reading unsupervised text

Abstract

We consider a scenario where an artificial agent is reading a stream of text composed of a set of narrations, and it is informed about the identity of some of the individuals that are mentioned in the text portion that is currently being read. The agent is expected to learn to follow the narrations, thus disambiguating mentions and discovering new individuals. We focus on the case in which individuals are entities and relations, and we propose an end-to-end trainable memory network that learns to discover and disambiguate them in an online manner, performing one-shot learning, and dealing with a small number of sparse supervisions. Our system builds a not-given-in-advance knowledge base, and it improves its skills while reading unsupervised text. The model deals with abrupt changes in the narration, taking into account their effects when resolving co-references. We showcase the strong disambiguation and discovery skills of our model on a corpus of Wikipedia documents and on a newly introduced dataset, that we make publicly available.