Deep Learning Relevance: Creating Relevant Information (as Opposed to Retrieving it)

TL;DR

This paper explores a novel approach where IR systems generate synthetic, relevant documents by deep learning, aiming to understand and synthesize information rather than just retrieve it, showing promising relevance results.

Contribution

It introduces a method using RNNs to generate synthetic relevant documents for queries, advancing beyond traditional retrieval to understanding and synthesis.

Findings

Synthetic documents ranked most relevant in crowdsourcing tests

Deep learning can generate relevant, coherent documents

Method outperforms baseline relevance in user assessments

Abstract

What if Information Retrieval (IR) systems did not just retrieve relevant information that is stored in their indices, but could also "understand" it and synthesise it into a single document? We present a preliminary study that makes a first step towards answering this question. Given a query, we train a Recurrent Neural Network (RNN) on existing relevant information to that query. We then use the RNN to "deep learn" a single, synthetic, and we assume, relevant document for that query. We design a crowdsourcing experiment to assess how relevant the "deep learned" document is, compared to existing relevant documents. Users are shown a query and four wordclouds (of three existing relevant documents and our deep learned synthetic document). The synthetic document is ranked on average most relevant of all.