Mesut Kaya
Welcome! I am a Postdoctoral Researcher at Jobindex A/S and IT University Copenhagen, working on the FairMatch: Multi-stakeholder Fairness for Algorithmic Hiring project.
Previously I was part of the JobMatch project, where the goal was to use recommender systems to match job seekers with jobs. Our candidate recommendation algorithm has been integrated into Jobindex's recruiter workflow.
Before that I was a Postdoc / Research Engineer at TU Delft, Delft Data Science, Web Information Systems (Epsilon group), working on Fairness-Aware Group Recommendations and Conversational Health in Social Media Platforms.
I obtained my PhD from Insight Centre for Data Analytics, University College Cork, Ireland, supervised by Dr. Derek Bridge, focusing on diversity-aware personalized recommender algorithms. My MSc is from Middle East Technical University, where I researched sentiment analysis of Turkish political news.
The fastest way to reach me is via my e-mail address.
News
April 3, 2026
Our paper "Human, Algorithm, or Both? Gender Bias in Human-Augmented Recruiting" accepted at the 9th ACM Conference on Fairness, Accountability, and Transparency.
April 3, 2026
Our paper "Analyzing the Effects of a Human-in-the-Loop Candidate Recommendation Algorithm at Jobindex" accepted at ACM Transactions on Recommender Systems Journal.
Publications
2026
Analyzing the Effects of a Human-in-the-Loop Candidate Recommendation Algorithm at Jobindex
ACM Transactions on Recommender Systems
2025
Mapping Stakeholder Needs to Multi-Sided Fairness in Candidate Recommendation for Algorithmic Hiring
Proceedings of the 19th ACM Conference on Recommender Systems
From Queries to Candidates: Exploring Search and Source Interaction Behavior of Recruiters
Proceedings of the 2025 ACM SIGIR Conference on Human Information Interaction and Retrieval
Fifth workshop on recommender systems for human resources (RecSys in HR 2025)
Proceedings of the 19th ACM Conference on Recommender Systems
2024
Fourth workshop on recommender systems for human resources (RecSys in HR 2024)
Proceedings of the 18th ACM Conference on Recommender Systems
2023
An Exploration of Sentence-Pair Classification for Algorithmic Recruiting
Proceedings of the 17th ACM Conference on Recommender Systems
Understanding Recruiters' Information Seeking Behavior in Talent Search
CHIIR '23: Proceedings of the 2023 Conference on Human Information Interaction and Retrieval
Third Workshop on Recommender Systems for Human Resources (RecSys in HR 2023)
Proceedings of the 17th ACM Conference on Recommender Systems
2022
Second Workshop on Recommender Systems for Human Resources (RecSys in HR 2022)
Proceedings of the 16th ACM Conference on Recommender Systems
Towards Comparing Recommendation to Multiple-Query Search Sessions for Talent Search
2022
Report on the 1st Workshop on Recommender Systems for Human Resources (RecSys in HR 2021) at RecSys 2021
ACM SIGIR Forum, 2022
2021
Effectiveness of Job Title Based Embeddings on Résumé to Job Ad Recommendation
2021
An Exploration of the Information Seeking Behavior of Recruiters
2021
Recommenders with a Mission: Assessing Diversity in News Recommendations
Proceedings of the 2021 Conference on Human Information Interaction and Retrieval
2020
Ensuring Fairness in Group Recommendations by Rank-Sensitive Balancing of Relevance
Fourteenth ACM Conference on Recommender Systems
▸ Abstract
For group recommendations, one objective is to recommend an ordered set of items to a group such that each individual recommendation is relevant for everyone. This paper defines fairness that 'balances' relevance across group members in a rank-sensitive way and provides a greedy algorithm (GFAR). Evaluated on two datasets, GFAR performs significantly better than all other algorithms around 43% of the time, particularly for large groups with diverging interests.
Contextual Personalized Re-Ranking of Music Recommendations through Audio Features
CARS 2020, Workshop on Context-Aware Recommender Systems (RecSys 2020)
2019
A Comparison of Calibrated and Intent-Aware Recommendations
Proceedings of the 13th ACM Conference on Recommender Systems
▸ Abstract
Calibrated and intent-aware recommendation are recent approaches with apparent similarities. This paper compares them in detail on two datasets, showing the extent to which intent-aware recommendations are calibrated and calibrated ones are diverse. Defining interests in terms of subprofiles yields highest precision and best relevance/diversity trade-off.
Subprofile-Aware Diversification of Recommendations
User Modeling and User-Adapted Interaction, 2019
▸ Abstract
Presents SPAD (Subprofile-Aware Diversification), a novel intent-aware diversification method where aspects are subprofiles of the user's profile rather than item features. On three datasets (movies, music artists, books), SPAD achieves higher precision and diversity than competing methods across more configurations.
Community-Aware Diversification of Recommendations
Proceedings of the 34th ACM/SIGAPP Symposium on Applied Computing
Sudden Death: A New Way to Compare Recommendation Diversification
REVEAL 2019 Workshop, RecSys 2019
2018
Accurate and Diverse Recommendations Using Item-Based SubProfiles
31st International Florida Artificial Intelligence Research Society Conference
Automatic Playlist Continuation Using Subprofile-Aware Diversification
Proceedings of the ACM Recommender Systems Challenge 2018
▸ Abstract
Describes teamrozik's approach to ACM RecSys Challenge 2018 on automatic playlist continuation. Uses Cold-Start-APC for short playlists and SPAD-APC for others, combining implicit matrix factorization with SPAD re-ranking. Demonstrates that subprofiles exist within playlists and that SPAD achieves higher precision than matrix factorization alone.
2017
Intent-Aware Diversification using Item-Based SubProfiles
Poster Track, 11th ACM Conference on Recommender Systems
2016
Improved Recommendation of Photo-Taking Locations Using Virtual Ratings
CEUR Workshop Proceedings
2013
Transfer Learning Using Twitter Data for Improving Sentiment Classification of Turkish Political News
2013
Sentiment Analysis of Turkish Political Columns with Transfer Learning
2013
2012
Sentiment Analysis of Turkish Political News
2012 IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology