Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

TIMBRE: Efficient Job Recommendation On Heterogeneous Graphs For Professional Recruiters

Eric Behar, Julien Romero, Amel Bouzeghoub, Katarzyna Wegrzyn-Wolska

TL;DR

A temporal graph-based method for job recommendation: TIMBRE (Temporal Integrated Model for Better REcommendations), adapted to allow efficient temporal recommendation and evaluation, which is later done using a graph neural network.

Abstract

Job recommendation gathers many challenges well-known in recommender systems. First, it suffers from the cold start problem, with the user (the candidate) and the item (the job) having a very limited lifespan. It makes the learning of good user and item representations hard. Second, the temporal aspect is crucial: We cannot recommend an item in the future or too much in the past. Therefore, using solely collaborative filtering barely works. Finally, it is essential to integrate information about the users and the items, as we cannot rely only on previous interactions. This paper proposes a temporal graph-based method for job recommendation: TIMBRE (Temporal Integrated Model for Better REcommendations). TIMBRE integrates user and item information into a heterogeneous graph. This graph is adapted to allow efficient temporal recommendation and evaluation, which is later done using a graph neural network. Finally, we evaluate our approach with recommender system metrics, rarely computed on graph-based recommender systems.

TIMBRE: Efficient Job Recommendation On Heterogeneous Graphs For Professional Recruiters

TL;DR

A temporal graph-based method for job recommendation: TIMBRE (Temporal Integrated Model for Better REcommendations), adapted to allow efficient temporal recommendation and evaluation, which is later done using a graph neural network.

Abstract

Job recommendation gathers many challenges well-known in recommender systems. First, it suffers from the cold start problem, with the user (the candidate) and the item (the job) having a very limited lifespan. It makes the learning of good user and item representations hard. Second, the temporal aspect is crucial: We cannot recommend an item in the future or too much in the past. Therefore, using solely collaborative filtering barely works. Finally, it is essential to integrate information about the users and the items, as we cannot rely only on previous interactions. This paper proposes a temporal graph-based method for job recommendation: TIMBRE (Temporal Integrated Model for Better REcommendations). TIMBRE integrates user and item information into a heterogeneous graph. This graph is adapted to allow efficient temporal recommendation and evaluation, which is later done using a graph neural network. Finally, we evaluate our approach with recommender system metrics, rarely computed on graph-based recommender systems.

Paper Structure

This paper contains 30 sections, 2 equations, 4 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Previous Work
  3. Job Recommendation
  4. Graph-Based Recommendation
  5. Temporal Recommendation
  6. Temporal Graph Neural Networks
  7. Datasets for temporal recommendation with side information
  8. Problem
  9. Background
  10. Our Problem
  11. Methodology
  12. Graph Construction
  13. Basic Structure
  14. Shortlist Nodes
  15. Temporal Nodes
  16. ...and 15 more sections

Figures (4)

  • Figure 1: Our complete job recommender system pipeline. 1. We turn our input data into a heterogeneous graph. 2. We replace relations between candidates and jobs with a shortlist node. 3. We add temporal nodes. 4; We apply our temporal sampling algorithm. 5. We apply a graph convolution network and make a prediction based on the representations of the shortlist node and the job node.
  • Figure 2: Sampling Distance For CF. C=Candidate, J=Job, SL=Shortlist.
  • Figure 3: Classification metrics computed on the negative sample
  • Figure 4: Recommendation system metrics