Finding People's Professions and Nationalities Using Distant Supervision - The FMI@SU "goosefoot" team at the WSDM Cup 2017 Triple Scoring Task

TL;DR

This paper presents a system that uses distant supervision, Wikipedia data, and machine learning to accurately score the relevance of profession and nationality triples, achieving top rankings in the WSDM Cup 2017 competition.

Contribution

The paper introduces a novel distant supervision method combined with word embeddings and regression modeling for triple scoring in profession and nationality classification.

Findings

Ranked 1st on Kendall's Tau in the competition

Achieved high accuracy and low score difference

Effective use of Wikipedia and DBpedia data

Abstract

We describe the system that our FMI@SU student's team built for participating in the Triple Scoring task at the WSDM Cup 2017. Given a triple from a "type-like" relation, profession or nationality, the goal is to produce a score, on a scale from 0 to 7, that measures the relevance of the statement expressed by the triple: e.g., how well does the profession of an Actor fit for Quentin Tarantino? We propose a distant supervision approach using information crawled from Wikipedia, DeletionPedia, and DBpedia, together with task-specific word embeddings, TF-IDF weights, and role occurrence order, which we combine in a linear regression model. The official evaluation ranked our submission 1st on Kendall's Tau, 7th on Average score difference, and 9th on Accuracy, out of 21 participating teams.