publications

2023

1. An Exploration of Sentence-Pair Classification for Algorithmic Recruiting

   Kaya, Mesut, and Bogers, Toine

   In Proceedings of the 17th ACM Conference on Recommender Systems 2023
2. Understanding Recruiters’ Information Seeking Behavior in Talent Search

   Kaya, Mesut, and Bogers, Toine

   In CHIIR ’23: Proceedings of the 2023 Conference on Human Information Interaction and Retrieval 2023
3. Third Workshop on Recommender Systems for Human Resources (RecSys in HR 2023)

   Bogers, Toine, Graus, David, Kaya, Mesut, Johnson, Chris, and Decorte, Jens-Joris

   In Proceedings of the 17th ACM Conference on Recommender Systems 2023

2022

1. Second Workshop on Recommender Systems for Human Resources (RecSys in HR 2022)

   Bogers, Toine, Graus, David, Kaya, Mesut, Gutiérrez, Francisco, Mesbah, Sepideh, and Johnson, Chris

   In Proceedings of the 16th ACM Conference on Recommender Systems 2022
2. Towards Comparing Recommendation to Multiple-Query Search Sessions for Talent Search

   Kaya, Mesut, and Bogers, Toine

   2022
3. Report on the 1st workshop on recommender systems for human resources (RecSys in HR 2021) at RecSys 2021

   Graus, David, Bogers, Toine, Kaya, Mesut, Gutiérrez, Francisco, and Verbert, Katrien

   In ACM SIGIR Forum 2022

2021

1. Effectiveness of job title based embeddings on résumé to job ad recommendation

   Kaya, Mesut, and Bogers, Toine

   2021
2. An Exploration of the Information Seeking Behavior of Recruiters

   Bogers, Toine, and Kaya, Mesut

   2021
3. Recommenders with a mission: assessing diversity in news recommendations

   Vrijenhoek, Sanne, Kaya, Mesut, Metoui, Nadia, Möller, Judith, Odijk, Daan, and Helberger, Natali

   In Proceedings of the 2021 Conference on Human Information Interaction and Retrieval 2021

2020

1. Ensuring Fairness in Group Recommendations by Rank-Sensitive Balancing of Relevance

   Kaya, Mesut, Bridge, Derek, and Tintarev, Nava

   In Fourteenth ACM Conference on Recommender Systems 2020

   Abs

   For group recommendations, one objective is to recommend an ordered set of items, a top-N, to a group such that each individual recommendation is relevant for everyone. A common way to do this is to select items on which the group can agree, using so-called ‘aggregation strategies’. One weakness of these aggregation strategies is that they select items independently of each other. They therefore cannot guarantee properties such as fairness, that apply to the set of recommendations as a whole. In this paper, we give a definition of fairness that ‘balances’ the relevance of the recommended items across the group members in a rank-sensitive way. Informally, an ordered set of recommended items is considered fair to a group if the relevance of the items in the top-N is balanced across the group members for each prefix of the top-N. In other words, the first item in the top-N should, as far as possible, balance the interests of all group members; the first two items taken together must do the same; also the first three; and so on up to N. In this paper, we formalize this notion of rank-sensitive balance and provide a greedy algorithm (GFAR) for finding a top-N set of group recommendations that satisfies our definition. We compare the performance of GFAR to five approaches from the literature on two datasets, one from each of the movie and music domains. We evaluate performance for 42 different configurations (two datasets, seven different group sizes, three different group types) and for ten evaluation metrics. We find that GFAR performs significantly better than all other algorithms around 43% of the time; in only 10% of cases are there algorithms that are significantly better than GFAR. Furthermore, GFAR performs particularly well in the most difficult cases, where groups are large and interests within the group diverge. We attribute GFAR’s success both to its rank-sensitivity and its way of balancing relevance. Current methods do not define fairness in a rank-sensitive way (although some achieve a degree of rank-sensitivity through the use of greedy algorithms) and none define balance in the way that we do.

2. Contextual Personalized Re-Ranking of Music Recommendations through Audio Features

   Gong, Boning, Kaya, Mesut, and Tintarev, Nava

   In Procs. of CARS 2020, the Workshop on Context-Aware Recommender Systems (Workshop Programme of the Fourteenth ACM Conference on Recommender Systems) 2020

2019

1. A Comparison of Calibrated and Intent-Aware Recommendations

   Kaya, Mesut, and Bridge, Derek

   In Proceedings of the 13th ACM Conference on Recommender Systems 2019

   Abs

   Calibrated and intent-aware recommendation are recent approaches to recommendation that have apparent similarities. Both try, to a certain extent, to cover the user’s interests, as revealed by her user profile. In this paper, we compare them in detail. On two datasets, we show the extent to which intent-aware recommendations are calibrated and the extent to which calibrated recommendations are diverse. We consider two ways of defining a user’s interests, one based on item features, the other based on subprofiles of the user’s profile. We find that defining interests in terms of subprofiles results in highest precision and the best relevance/diversity trade-off. Along the way, we define a new version of calibrated recommendation and three new evaluation metrics.

2. Subprofile-Aware Diversification of Recommendations

   Kaya, Mesut, and Bridge, Derek

   User Modeling and User-Adapted Interaction 2019

   Abs

   A user of a recommender system is more likely to be satisfied by one or more of the recommendations if each individual recommendation is relevant to her but additionally if the set of recommendations is diverse. The most common approach to recommendation diversification uses re-ranking: the recommender system scores a set of candidate items for relevance to the user; it then re-ranks the candidates so that the subset that it will recommend achieves a balance between relevance and diversity. Ordinarily, we expect a trade-off between relevance and diversity: the diversity of the set of recommendations increases by including items that have lower relevance scores but which are different from the items already in the set. In early work, the diversity of a set of recommendations was given by the average of their distances from one another, according to some semantic distance metric defined on item features such as movie genres. More recent intent-aware diversification methods formulate diversity in terms of coverage and relevance of aspects. The aspects are most commonly defined in terms of item features. By trying to ensure that the aspects of a set of recommended items cover the aspects of the items in the user’s profile, the level of diversity is more personalized. In offline experiments on pre-collected datasets, intent-aware diversification using item features as aspects sometimes defies the relevance/diversity trade-off: there are configurations in which the recommendations exhibits increases in both relevance and diversity. In this paper, we present a new form of intent-aware diversification, which we call SPAD (Subprofile-Aware Diversification), and a variant called RSPAD (Relevance-based SPAD). In SPAD, the aspects are not item features; they are subprofiles of the user’s profile. We present and compare a number of different ways to extract subprofiles from a user’s profile. None of them is defined in terms of item features. Therefore, SPAD is useful even in domains where item features are not available or are of low quality. On three pre-collected datasets from three different domains (movies, music artists and books), we compare SPAD and RSPAD to intent-aware methods in which aspects are item features. We find on these datasets that SPAD and RSPAD suffer even less from the relevance/diversity trade-off: across all three datasets, they increase both relevance and diversity for even more configurations than other approaches to diversification. Moreover, we find that SPAD and RSPAD are the most accurate systems across all three datasets.

3. Community-Aware Diversification of Recommendations

   Kaya, Mesut, and Bridge, Derek

   In Proceedings of the 34th ACM/SIGAPP Symposium on Applied Computing 2019

   Abs

   Intent-aware methods for recommendation diversification seek to ensure that the recommended items cover so-called aspects, which are assumed to define the user’s tastes and interests. Most typically, aspects are item features such as movie or music genres. In recent work, we presented a novel intent-aware diversification method, called Subprofile-Aware Diversification (SPAD). In SPAD, aspects are subprofiles of the active user’s profile, detected using an item-item similarity method. In this paper, we propose Community-Aware Diversification (CAD), in which aspects are again subprofiles but are detected indirectly through users who are similar to the active user. We evaluate CAD’s precision and diversity on four different datasets, and compare it with SPAD and an intent-aware diversification method called xQuAD. We show that on two of the datasets SPAD outperforms CAD, but for the other two CAD outperforms SPAD. For all datasets, both CAD and SPAD achieve higher precision than xQuAD. When it comes to diversity, xQuAD sometimes results in more diverse recommendations but it is more prone to paying for this diversity with decreases in precision. Arguably, SPAD and CAD strike a better balance between the two.

4. Sudden Death: A New Way to Compare Recommendation Diversification

   Bridge, Derek, Kaya, Mesut, and Castells, Pablo

   In Procs. of REVEAL 2019, the Workshop on Reinforcement and Robust Estimators for Recommendation (Workshop Programme of the Thirteenth ACM Conference on Recommender Systems) 2019

2018

1. Accurate and Diverse Recommendations Using Item-Based SubProfiles

   Kaya, Mesut, and Bridge, Derek

   In Procs. of the Thirty-First International Florida Artificial Intelligence Research Society Conference 2018
2. Automatic Playlist Continuation Using Subprofile-Aware Diversification

   Kaya, Mesut, and Bridge, Derek

   In Proceedings of the ACM Recommender Systems Challenge 2018 2018

   Abs

   The ACM RecSys Challenge 2018 involves the task of automatic playlist continuation (APC), aiming to help users to create and extend their own music playlists. In this paper, we explain teamrozik’s approach to the Challenge. Our approach to APC is twofold: Cold-Start-APC for short playlists and SPAD-APC for other playlists. Cold-Start-APC is a rudimentary popularity-based recommender. SPAD-APC treats playlists as if they were user profiles. It builds an implicit matrix factorization model to generate initial recommendations. But it re-ranks those recommendations using SubProfile-Aware Diversification (SPAD), which is a personalized intent-aware diversification method. The SPAD re-ranking method aims to ensure that the final set of recommendations covers different interests or tastes in the playlists of the users, which we refer to as subprofiles. We show that such subprofiles do exist within playlists and we show that the SPAD method achieves higher precision than matrix factorization alone.

2017

1. Intent-Aware Diversification using Item-Based SubProfiles

   Kaya, Mesut, and Bridge, Derek

   In Procs. of the Poster Track of the 11th ACM Conference on Recommender Systems 2017

2016

1. Improved recommendation of photo-taking locations using virtual ratings

   Kaya, Mesut, and Bridge, Derek G

   In CEUR Workshop Proceedings 2016

2013

1. Transfer learning using Twitter data for improving sentiment classification of Turkish political news

   Kaya, Mesut, Fidan, Guven, and Toroslu, I Hakkı

   2013
2. Sentiment analysis of Turkish political columns with transfer learning

   Kaya, Mesut

   2013

2012

1. Sentiment analysis of turkish political news

   Kaya, Mesut, Fidan, Güven, and Toroslu, Ismail H

   In 2012 IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology 2012