An Explanatory Query-Based Framework for Exploring Academic Expertise

TL;DR

This paper introduces a query-based system that automatically identifies academic researchers' expertise from publication abstracts, aiding collaboration and resource matching with explanations, evaluated on university data.

Contribution

It presents a novel, efficient framework that uses natural language queries and domain knowledge to find relevant researchers, outperforming neural network baselines.

Findings

Effective in matching researchers to queries

Outperforms neural network baselines

Works across different academic domains

Abstract

The success of research institutions heavily relies upon identifying the right researchers "for the job": researchers may need to identify appropriate collaborators, often from across disciplines; students may need to identify suitable supervisors for projects of their interest; administrators may need to match funding opportunities with relevant researchers, and so on. Usually, finding potential collaborators in institutions is a time-consuming manual search task prone to bias. In this paper, we propose a novel query-based framework for searching, scoring, and exploring research expertise automatically, based upon processing abstracts of academic publications. Given user queries in natural language, our framework finds researchers with relevant expertise, making use of domain-specific knowledge bases and word embeddings. It also generates explanations for its recommendations. We evaluate our framework with an institutional repository of papers from a leading university, using, as baselines, artificial neural networks and transformer-based models for a multilabel classification task to identify authors of publication abstracts. We also assess the cross-domain effectiveness of our framework with a (separate) research funding repository for the same institution. We show that our simple method is effective in identifying matches, while satisfying desirable properties and being efficient.