# Should we Embed? A Study on the Online Performance of Utilizing Embeddings for Real-Time Job Recommendations

## TL;DR

This study evaluates the effectiveness of embedding-based recommendations for online job platforms, showing that recent interaction embeddings excel for similar job suggestions, while combined frequency and recency embeddings improve homepage personalization.

## Contribution

It provides an empirical analysis of embedding strategies for real-time job recommendations, highlighting the importance of interaction recency and frequency in online performance.

## Key findings

- Recent interaction embeddings improve similar job recommendations.
- Combining frequency and recency embeddings enhances homepage personalization.
- Embedding strategies significantly impact click-through rates.

## Abstract

In this work, we present the findings of an online study, where we explore the impact of utilizing embeddings to recommend job postings under real-time constraints. On the Austrian job platform Studo Jobs, we evaluate two popular recommendation scenarios: (i) providing similar jobs and, (ii) personalizing the job postings that are shown on the homepage. Our results show that for recommending similar jobs, we achieve the best online performance in terms of Click-Through Rate when we employ embeddings based on the most recent interaction. To personalize the job postings shown on a user's homepage, however, combining embeddings based on the frequency and recency with which a user interacts with job postings results in the best online performance.