# Professional Network Matters: Connections Empower Person-Job Fit

**Authors:** Hao Chen, Lun Du, Yuxuan Lu, Qiang Fu, Xu Chen, Shi Han, Yanbin Kang, Guangming Lu, Zi Li  
**arXiv:** [2401.00010](https://arxiv.org/abs/2401.00010)  
**Date:** 2023-12-19  
**Summary page:** https://www.summarizepaper.com/en/paper/2401.00010v1/

## Key points

- Person-Job Fit models are commonly used in online recruitment platforms to match job seekers with suitable positions.
- Existing methodologies often overlook the importance of job seekers' social relationships within their professional networks.
- The integration of professional networks into the Person-Job Fit model is advocated to enhance its efficacy.
- The proposed approach involves two key stages: establishing a Workplace Heterogeneous Information Network (WHIN) and crafting a Contextual Social Attention Graph Neural Network (CSAGNN).
- CSAGNN supplements users' missing information with contextual insights from their professional connections and introduces a job-specific attention mechanism to address noisy networks.
- Experimental evaluations across real-world datasets show that this approach outperforms baseline models, emphasizing the significance of incorporating professional networks into the Person-Job Fit model.

## Summary

, , , , 

In the realm of online recruitment platforms, the utilization of Person-Job Fit models is a common practice to facilitate the automatic matching of job seekers with suitable job positions. However, existing methodologies often overlook the importance of job seekers' social relationships within their professional networks. This paper advocates for the integration of professional networks into the Person-Job Fit model to enhance its efficacy. The proposed approach consists of two key stages: firstly, establishing a Workplace Heterogeneous Information Network (WHIN) that encompasses diverse knowledge elements, including professional connections and pre-training representations of various entities through a heterogeneous graph neural network. Secondly, crafting a Contextual Social Attention Graph Neural Network (CSAGNN) designed to supplement users' missing information with contextual insights derived from their professional connections. To address noisy professional networks, a job-specific attention mechanism is introduced in CSAGNN, leveraging pre-trained entity representations from WHIN. Through rigorous experimental evaluations conducted across three real-world recruitment datasets sourced from LinkedIn, this innovative approach demonstrates superior performance compared to baseline models. This highlights the significance and impact of incorporating professional networks into the Person-Job Fit model. This research sheds light on how connections within professional networks can empower and optimize the process of matching job seekers with appropriate job positions on online recruitment platforms.