Marjan Celikik, Ana Peleteiro-Ramallo, Jacek Wasilewski
A large number of empirical studies on applying self-attention models in the domain of recommender systems are based on offline evaluation and metrics computed on standardized datasets. Moreover, many of them do not consider side information such as item and customer metadata although deep-learning recommenders live up to their full potential only when numerous features of heterogeneous type are included. Also, normally the model is used only for a single use case. Due to these shortcomings, even if relevant, previous works are not always representative of their actual effectiveness in real-world industry applications. In this talk, we contribute to bridging this gap by presenting live experimental results demonstrating improvements in user retention of up to 30%. Moreover, we share our learnings and challenges from building a re-usable and configurable recommender system for various applications from the fashion industry. In particular, we focus on fashion inspiration use-cases, such as outfit ranking, outfit recommendation and real-time personalized outfit generation.
{"title":"Reusable Self-Attention Recommender Systems in Fashion Industry Applications","authors":"Marjan Celikik, Ana Peleteiro-Ramallo, Jacek Wasilewski","doi":"10.1145/3523227.3547377","DOIUrl":"https://doi.org/10.1145/3523227.3547377","url":null,"abstract":"A large number of empirical studies on applying self-attention models in the domain of recommender systems are based on offline evaluation and metrics computed on standardized datasets. Moreover, many of them do not consider side information such as item and customer metadata although deep-learning recommenders live up to their full potential only when numerous features of heterogeneous type are included. Also, normally the model is used only for a single use case. Due to these shortcomings, even if relevant, previous works are not always representative of their actual effectiveness in real-world industry applications. In this talk, we contribute to bridging this gap by presenting live experimental results demonstrating improvements in user retention of up to 30%. Moreover, we share our learnings and challenges from building a re-usable and configurable recommender system for various applications from the fashion industry. In particular, we focus on fashion inspiration use-cases, such as outfit ranking, outfit recommendation and real-time personalized outfit generation.","PeriodicalId":443279,"journal":{"name":"Proceedings of the 16th ACM Conference on Recommender Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116259294","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Farfetch, the leading online platform for luxury fashion, has spent several years developing a recommender system. In fact, recommendations have been quite successful in improving both the user experience and the company’s own business metrics [3–9]. In this talk we will shed some light on how we built our recommender system at Farfetch, the main obstacles we faced, and some plans for the future. Recommendations started their journey at Farfetch somewhere around 2015. At the time, we had a single model that trained once per day that updated the users’ recommendations with the same frequency. Currently, we have around 20 models in production and the majority of them are designed to handle streaming data from the users and adapt in realtime to user actions. How can we balance training and improving existing models, creating new models, serving them in real time and still keep our code in check, our tests up to date and our pipelines moving? We will discuss the three main components that we created in order to tackle our real world issue of providing ever-improving recommendations to our customers: The Gym, The Recommenders and The API.
{"title":"Recommendations: They’re in fashion","authors":"C. Carvalheira, Tiago Lacerda, Diogo Gonçalves","doi":"10.1145/3523227.3547389","DOIUrl":"https://doi.org/10.1145/3523227.3547389","url":null,"abstract":"Farfetch, the leading online platform for luxury fashion, has spent several years developing a recommender system. In fact, recommendations have been quite successful in improving both the user experience and the company’s own business metrics [3–9]. In this talk we will shed some light on how we built our recommender system at Farfetch, the main obstacles we faced, and some plans for the future. Recommendations started their journey at Farfetch somewhere around 2015. At the time, we had a single model that trained once per day that updated the users’ recommendations with the same frequency. Currently, we have around 20 models in production and the majority of them are designed to handle streaming data from the users and adapt in realtime to user actions. How can we balance training and improving existing models, creating new models, serving them in real time and still keep our code in check, our tests up to date and our pipelines moving? We will discuss the three main components that we created in order to tackle our real world issue of providing ever-improving recommendations to our customers: The Gym, The Recommenders and The API.","PeriodicalId":443279,"journal":{"name":"Proceedings of the 16th ACM Conference on Recommender Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129486794","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Ori Katz, Oren Barkan, Noam Koenigstein, Nir Zabari
The problem of Next Basket Recommendation (NBR) addresses the challenge of recommending items for the next basket of a user, based on her sequence of prior baskets. In this paper, we focus on a variation of this problem in which we aim to predict repurchases, i.e. we wish to recommend a user only items she had purchased before. We coin this problem Next Basket Repurchase Recommendation (NBRR). Over the years, a variety of models have been proposed to address the problem of NBR, however, the problem of NBRR has been overlooked. Although being highly related problems, which are often solved by the same methods, the problem of repurchase recommendation calls for a different approach. In this paper, we share insights from our experience of facing the challenge of NBRR. In light of these insights, we propose a novel hyper-convolutional model to leverage the behavioral patterns of repeated purchases. We demonstrate the effectiveness of the proposed model on three publicly available datasets, where it is shown to outperform other existing methods across multiple metrics.
{"title":"Learning to Ride a Buy-Cycle: A Hyper-Convolutional Model for Next Basket Repurchase Recommendation","authors":"Ori Katz, Oren Barkan, Noam Koenigstein, Nir Zabari","doi":"10.1145/3523227.3546763","DOIUrl":"https://doi.org/10.1145/3523227.3546763","url":null,"abstract":"The problem of Next Basket Recommendation (NBR) addresses the challenge of recommending items for the next basket of a user, based on her sequence of prior baskets. In this paper, we focus on a variation of this problem in which we aim to predict repurchases, i.e. we wish to recommend a user only items she had purchased before. We coin this problem Next Basket Repurchase Recommendation (NBRR). Over the years, a variety of models have been proposed to address the problem of NBR, however, the problem of NBRR has been overlooked. Although being highly related problems, which are often solved by the same methods, the problem of repurchase recommendation calls for a different approach. In this paper, we share insights from our experience of facing the challenge of NBRR. In light of these insights, we propose a novel hyper-convolutional model to leverage the behavioral patterns of repeated purchases. We demonstrate the effectiveness of the proposed model on three publicly available datasets, where it is shown to outperform other existing methods across multiple metrics.","PeriodicalId":443279,"journal":{"name":"Proceedings of the 16th ACM Conference on Recommender Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124138771","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Group recommender systems (GRS) focus on recommending items to groups of users. GRS need to tackle the heterogeneity of group members’ preferences and produce recommendations of high overall utility while also considering some sense of fairness among group members. This work plans to aim for novel applications of GRS involving construction of large-scale groups of users and focusing on the long-term fairness of these groups which is in contrast with current research that concentrates on small groups of ephemeral nature. We believe that these directions could bring results of significant societal impact and scope of the effect expanding beyond currently considered GRS domains, e.g., helping to mitigate the filter bubble problem
{"title":"Long-term fairness for Group Recommender Systems with Large Groups","authors":"Patrik Dokoupil","doi":"10.1145/3523227.3547424","DOIUrl":"https://doi.org/10.1145/3523227.3547424","url":null,"abstract":"Group recommender systems (GRS) focus on recommending items to groups of users. GRS need to tackle the heterogeneity of group members’ preferences and produce recommendations of high overall utility while also considering some sense of fairness among group members. This work plans to aim for novel applications of GRS involving construction of large-scale groups of users and focusing on the long-term fairness of these groups which is in contrast with current research that concentrates on small groups of ephemeral nature. We believe that these directions could bring results of significant societal impact and scope of the effect expanding beyond currently considered GRS domains, e.g., helping to mitigate the filter bubble problem","PeriodicalId":443279,"journal":{"name":"Proceedings of the 16th ACM Conference on Recommender Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127709126","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Blaž Škrlj, A. Schwartz, Jure Ferlez, Davorin Kopic, Naama Ziporin
Hyperparameter optimization is the process of identifying the appropriate hyperparameter configuration of a given machine learning model with regard to a given learning task. For smaller data sets, an exhaustive search is possible; However, when the data size and model complexity increase, the number of configuration evaluations becomes the main computational bottleneck. A promising paradigm for tackling this type of problem is surrogate-based optimization. The main idea underlying this paradigm considers an incrementally updated model of the relation between the hyperparameter space and the output (target) space; the data for this model are obtained by evaluating the main learning engine, which is, for example, a factorization machine-based model. By learning to approximate the hyperparameter-target relation, the surrogate (machine learning) model can be used to score large amounts of hyperparameter configurations, exploring parts of the configuration space beyond the reach of direct machine learning engine evaluation. Commonly, a surrogate is selected prior to optimization initialization and remains the same during the search. We investigated whether dynamic switching of surrogates during the optimization itself is a sensible idea of practical relevance for selecting the most appropriate factorization machine-based models for large-scale online recommendation. We conducted benchmarks on data sets containing hundreds of millions of instances against established baselines such as Random Forest- and Gaussian process-based surrogates. The results indicate that surrogate switching can offer good performance while considering fewer learning engine evaluations.
{"title":"Dynamic Surrogate Switching: Sample-Efficient Search for Factorization Machine Configurations in Online Recommendations","authors":"Blaž Škrlj, A. Schwartz, Jure Ferlez, Davorin Kopic, Naama Ziporin","doi":"10.1145/3523227.3547384","DOIUrl":"https://doi.org/10.1145/3523227.3547384","url":null,"abstract":"Hyperparameter optimization is the process of identifying the appropriate hyperparameter configuration of a given machine learning model with regard to a given learning task. For smaller data sets, an exhaustive search is possible; However, when the data size and model complexity increase, the number of configuration evaluations becomes the main computational bottleneck. A promising paradigm for tackling this type of problem is surrogate-based optimization. The main idea underlying this paradigm considers an incrementally updated model of the relation between the hyperparameter space and the output (target) space; the data for this model are obtained by evaluating the main learning engine, which is, for example, a factorization machine-based model. By learning to approximate the hyperparameter-target relation, the surrogate (machine learning) model can be used to score large amounts of hyperparameter configurations, exploring parts of the configuration space beyond the reach of direct machine learning engine evaluation. Commonly, a surrogate is selected prior to optimization initialization and remains the same during the search. We investigated whether dynamic switching of surrogates during the optimization itself is a sensible idea of practical relevance for selecting the most appropriate factorization machine-based models for large-scale online recommendation. We conducted benchmarks on data sets containing hundreds of millions of instances against established baselines such as Random Forest- and Gaussian process-based surrogates. The results indicate that surrogate switching can offer good performance while considering fewer learning engine evaluations.","PeriodicalId":443279,"journal":{"name":"Proceedings of the 16th ACM Conference on Recommender Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128751699","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
DAGFiNN is a conversational conference assistant that can be made available for a given conference both as a chatbot on the website and as a Furhat robot physically exhibited at the conference venue. Conference participants can interact with the assistant to get advice on various questions, ranging from where to eat in the city or how to get to the airport to which sessions we recommend them to attend based on the information we have about them. The overall objective is to provide a personalized and engaging experience and allow users to ask a broad range of questions that naturally arise before and during the conference.
{"title":"DAGFiNN: A Conversational Conference Assistant","authors":"Ivica Kostric, K. Balog, Tølløv Alexander Aresvik, Nolwenn Bernard, Eyvinn Thu Dørheim, Pholit Hantula, Sander Havn-Sørensen, Rune Henriksen, Hengameh Hosseini, Ekaterina Khlybova, Weronika Lajewska, Sindre Ekrheim Mosand, Narmin Orujova","doi":"10.1145/3523227.3551467","DOIUrl":"https://doi.org/10.1145/3523227.3551467","url":null,"abstract":"DAGFiNN is a conversational conference assistant that can be made available for a given conference both as a chatbot on the website and as a Furhat robot physically exhibited at the conference venue. Conference participants can interact with the assistant to get advice on various questions, ranging from where to eat in the city or how to get to the airport to which sessions we recommend them to attend based on the information we have about them. The overall objective is to provide a personalized and engaging experience and allow users to ask a broad range of questions that naturally arise before and during the conference.","PeriodicalId":443279,"journal":{"name":"Proceedings of the 16th ACM Conference on Recommender Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126991115","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Siyong Ma, Puja Das, S. M. Nikolakaki, Qifeng Chen, Humeyra Topcu Altintas
Online commercial app marketplaces serve millions of apps to billions of users in an efficient manner. Bandit optimization algorithms are used to ensure that the recommendations are relevant, and converge to the best performing content over time. However, directly applying bandits to real-world systems, where the catalog of items is dynamic and continuously refreshed, is not straightforward. One of the challenges we face is the existence of several competing content surfacing components, a phenomenon not unusual in large-scale recommender systems. This often leads to challenging scenarios, where improving the recommendations in one component can lead to performance degradation of another, i.e., “cannibalization”. To address this problem we introduce an efficient two-layer bandit approach which is contextualized to user cohorts of similar taste. We mitigate cannibalization at runtime within a single multi-intent content surfacing platform by formalizing relevant offline evaluation metrics, and by involving the cross-component interactions in the bandit rewards. The user engagement in our proposed system has more than doubled as measured by online A/B testings.
{"title":"Two-Layer Bandit Optimization for Recommendations","authors":"Siyong Ma, Puja Das, S. M. Nikolakaki, Qifeng Chen, Humeyra Topcu Altintas","doi":"10.1145/3523227.3547396","DOIUrl":"https://doi.org/10.1145/3523227.3547396","url":null,"abstract":"Online commercial app marketplaces serve millions of apps to billions of users in an efficient manner. Bandit optimization algorithms are used to ensure that the recommendations are relevant, and converge to the best performing content over time. However, directly applying bandits to real-world systems, where the catalog of items is dynamic and continuously refreshed, is not straightforward. One of the challenges we face is the existence of several competing content surfacing components, a phenomenon not unusual in large-scale recommender systems. This often leads to challenging scenarios, where improving the recommendations in one component can lead to performance degradation of another, i.e., “cannibalization”. To address this problem we introduce an efficient two-layer bandit approach which is contextualized to user cohorts of similar taste. We mitigate cannibalization at runtime within a single multi-intent content surfacing platform by formalizing relevant offline evaluation metrics, and by involving the cross-component interactions in the bandit rewards. The user engagement in our proposed system has more than doubled as measured by online A/B testings.","PeriodicalId":443279,"journal":{"name":"Proceedings of the 16th ACM Conference on Recommender Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132742521","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
At Peloton, we are challenged to not just surface relevant recommendations of fitness classes to our members, but also timely ones. As our fitness content library expands, we continually produce classes on certain themes which are most timely during a narrow time window. To address this challenge, we provide some control over our recommendations to external stakeholders, such as production and marketing teams. They enter timed boosts of certain classes during the windows they are relevant in. We have built out algorithms which take these desired classes and elevate the number of impressions for them, while preserving members’ engagement with our recommendations. In this paper, we discuss the system, the algorithms and some results from a few A/B tests showing how boosting works in practice.
{"title":"Timely Personalization at Peloton: A System and Algorithm for Boosting Time-Relevant Content","authors":"Shayak Banerjee, Vijay Pappu, N. Talukder, Shoya Yoshida, Arnab Bhadury, Allison Schloss, Jasmine Paulino","doi":"10.1145/3523227.3547391","DOIUrl":"https://doi.org/10.1145/3523227.3547391","url":null,"abstract":"At Peloton, we are challenged to not just surface relevant recommendations of fitness classes to our members, but also timely ones. As our fitness content library expands, we continually produce classes on certain themes which are most timely during a narrow time window. To address this challenge, we provide some control over our recommendations to external stakeholders, such as production and marketing teams. They enter timed boosts of certain classes during the windows they are relevant in. We have built out algorithms which take these desired classes and elevate the number of impressions for them, while preserving members’ engagement with our recommendations. In this paper, we discuss the system, the algorithms and some results from a few A/B tests showing how boosting works in practice.","PeriodicalId":443279,"journal":{"name":"Proceedings of the 16th ACM Conference on Recommender Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124515409","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Riccardo Nembrini, Costantino Carugno, Maurizio Ferrari Dacrema, P. Cremonesi
After decades of being mainly confined to theoretical research, Quantum Computing is now becoming a useful tool for solving realistic problems. This work aims to experimentally explore the feasibility of using currently available quantum computers, based on the Quantum Annealing paradigm, to build a recommender system exploiting community detection. Community detection, by partitioning users and items into densely connected clusters, can boost the accuracy of non-personalized recommendation by assuming that users within each community share similar tastes. However, community detection is a computationally expensive process. The recent availability of Quantum Annealers as cloud-based devices, constitutes a new and promising direction to explore community detection, although effectively leveraging this new technology is a long-term path that still requires advancements in both hardware and algorithms. This work aims to begin this path by assessing the quality of community detection formulated as a Quadratic Unconstrained Binary Optimization problem on a real recommendation scenario. Results on several datasets show that the quantum solver is able to detect communities of comparable quality with respect to classical solvers, but with better speedup, and the non-personalized recommendation models built on top of these communities exhibit improved recommendation quality. The takeaway is that quantum computing, although in its early stages of maturity and applicability, shows promise in its ability to support new recommendation models and to bring improved scalability as technology evolves.
{"title":"Towards Recommender Systems with Community Detection and Quantum Computing","authors":"Riccardo Nembrini, Costantino Carugno, Maurizio Ferrari Dacrema, P. Cremonesi","doi":"10.1145/3523227.3551478","DOIUrl":"https://doi.org/10.1145/3523227.3551478","url":null,"abstract":"After decades of being mainly confined to theoretical research, Quantum Computing is now becoming a useful tool for solving realistic problems. This work aims to experimentally explore the feasibility of using currently available quantum computers, based on the Quantum Annealing paradigm, to build a recommender system exploiting community detection. Community detection, by partitioning users and items into densely connected clusters, can boost the accuracy of non-personalized recommendation by assuming that users within each community share similar tastes. However, community detection is a computationally expensive process. The recent availability of Quantum Annealers as cloud-based devices, constitutes a new and promising direction to explore community detection, although effectively leveraging this new technology is a long-term path that still requires advancements in both hardware and algorithms. This work aims to begin this path by assessing the quality of community detection formulated as a Quadratic Unconstrained Binary Optimization problem on a real recommendation scenario. Results on several datasets show that the quantum solver is able to detect communities of comparable quality with respect to classical solvers, but with better speedup, and the non-personalized recommendation models built on top of these communities exhibit improved recommendation quality. The takeaway is that quantum computing, although in its early stages of maturity and applicability, shows promise in its ability to support new recommendation models and to bring improved scalability as technology evolves.","PeriodicalId":443279,"journal":{"name":"Proceedings of the 16th ACM Conference on Recommender Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128780989","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Traditional research in Recommender Systems (RSs) often solely focuses on accuracy and a limited number of beyond-accuracy dimensions. Nonetheless, real-world RSs need to consider several other aspects, such as customer satisfaction or stakeholders’ interests. Consequently, the evaluation criteria must comprehend other dimensions, like click rate, or revenue, to cite a few of them. However, what objective should the system optimize, and what objective should it sacrifice? An emerging approach to tackle the problem and aim to blend different (sometimes conflicting) objectives is Multi-Objective Recommender Systems (MORSs). This proposal sketches a strategy to exploit the Pareto optimality to introduce a new optimal solution selection approach and investigate how existing RSs perform with multi-objective tasks. The goals are twofold: (i) discovering how to rank the solutions lying on the Pareto frontier to find the best trade-off solution and (ii) comparing the Pareto frontiers of different recommendation approaches to assess whether one performs better for the considered objectives. These measures could lead to a new class of MORSs that train an RS on multiple objectives to reach the best trade-off solution directly.
{"title":"Pursuing Optimal Trade-Off Solutions in Multi-Objective Recommender Systems","authors":"Vincenzo Paparella","doi":"10.1145/3523227.3547425","DOIUrl":"https://doi.org/10.1145/3523227.3547425","url":null,"abstract":"Traditional research in Recommender Systems (RSs) often solely focuses on accuracy and a limited number of beyond-accuracy dimensions. Nonetheless, real-world RSs need to consider several other aspects, such as customer satisfaction or stakeholders’ interests. Consequently, the evaluation criteria must comprehend other dimensions, like click rate, or revenue, to cite a few of them. However, what objective should the system optimize, and what objective should it sacrifice? An emerging approach to tackle the problem and aim to blend different (sometimes conflicting) objectives is Multi-Objective Recommender Systems (MORSs). This proposal sketches a strategy to exploit the Pareto optimality to introduce a new optimal solution selection approach and investigate how existing RSs perform with multi-objective tasks. The goals are twofold: (i) discovering how to rank the solutions lying on the Pareto frontier to find the best trade-off solution and (ii) comparing the Pareto frontiers of different recommendation approaches to assess whether one performs better for the considered objectives. These measures could lead to a new class of MORSs that train an RS on multiple objectives to reach the best trade-off solution directly.","PeriodicalId":443279,"journal":{"name":"Proceedings of the 16th ACM Conference on Recommender Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115913389","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}