Company Name Matching Using Job Market Data Enrichment

Abstract

This article contributes to the field of matching techniques by introducing a new algorithm based on labor market data enrichment. This approach is able to collect and balance the training and test samples for data integration purposes. By setting thresholds for textual matching and geographic proximity, it simplifies the process of finding suitable company matches. Based on insufficiently studied datasets, the experimental findings show that the performance evaluation of proposed models differs depending on the similarity thresholds used.