F. Abel, A. Benczúr, Daniel Kohlsdorf, M. Larson, Róbert Pálovics
The 2016 ACM Recommender Systems Challenge focused on the problem of job recommendations. Given a large dataset from XING that consisted of anonymized user profiles, job postings, and interactions between them, the participating teams had to predict postings that a user will interact with. The challenge ran for four months with 366 registered teams. 119 of those teams actively participated and submitted together 4,232 solutions yielding in an impressive neck-and-neck race that was decided within the last days of the challenge.
{"title":"RecSys Challenge 2016: Job Recommendations","authors":"F. Abel, A. Benczúr, Daniel Kohlsdorf, M. Larson, Róbert Pálovics","doi":"10.1145/2959100.2959207","DOIUrl":"https://doi.org/10.1145/2959100.2959207","url":null,"abstract":"The 2016 ACM Recommender Systems Challenge focused on the problem of job recommendations. Given a large dataset from XING that consisted of anonymized user profiles, job postings, and interactions between them, the participating teams had to predict postings that a user will interact with. The challenge ran for four months with 366 registered teams. 119 of those teams actively participated and submitted together 4,232 solutions yielding in an impressive neck-and-neck race that was decided within the last days of the challenge.","PeriodicalId":315651,"journal":{"name":"Proceedings of the 10th ACM Conference on Recommender Systems","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121235556","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}
Pairwise learning-to-rank algorithms have been shown to allow recommender systems to leverage unary user feedback. We propose Multi-feedback Bayesian Personalized Ranking (MF-BPR), a pairwise method that exploits different types of feedback with an extended sampling method. The feedback types are drawn from different "channels", in which users interact with items (e.g., clicks, likes, listens, follows, and purchases). We build on the insight that different kinds of feedback, e.g., a click versus a like, reflect different levels of commitment or preference. Our approach differs from previous work in that it exploits multiple sources of feedback simultaneously during the training process. The novelty of MF-BPR is an extended sampling method that equates feedback sources with "levels" that reflect the expected contribution of the signal. We demonstrate the effectiveness of our approach with a series of experiments carried out on three datasets containing multiple types of feedback. Our experimental results demonstrate that with a right sampling method, MF-BPR outperforms BPR in terms of accuracy. We find that the advantage of MF-BPR lies in its ability to leverage level information when sampling negative items.
{"title":"Bayesian Personalized Ranking with Multi-Channel User Feedback","authors":"B. Loni, Roberto Pagano, M. Larson, A. Hanjalic","doi":"10.1145/2959100.2959163","DOIUrl":"https://doi.org/10.1145/2959100.2959163","url":null,"abstract":"Pairwise learning-to-rank algorithms have been shown to allow recommender systems to leverage unary user feedback. We propose Multi-feedback Bayesian Personalized Ranking (MF-BPR), a pairwise method that exploits different types of feedback with an extended sampling method. The feedback types are drawn from different \"channels\", in which users interact with items (e.g., clicks, likes, listens, follows, and purchases). We build on the insight that different kinds of feedback, e.g., a click versus a like, reflect different levels of commitment or preference. Our approach differs from previous work in that it exploits multiple sources of feedback simultaneously during the training process. The novelty of MF-BPR is an extended sampling method that equates feedback sources with \"levels\" that reflect the expected contribution of the signal. We demonstrate the effectiveness of our approach with a series of experiments carried out on three datasets containing multiple types of feedback. Our experimental results demonstrate that with a right sampling method, MF-BPR outperforms BPR in terms of accuracy. We find that the advantage of MF-BPR lies in its ability to leverage level information when sampling negative items.","PeriodicalId":315651,"journal":{"name":"Proceedings of the 10th ACM Conference on Recommender Systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128966792","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}
Chih-Ming Chen, Ming-Feng Tsai, Yu-Ching Lin, Yi-Hsuan Yang
A common scenario considered in recommender systems is to predict a user's preferences on unseen items based on his/her preferences on observed items. A major limitation of this scenario is that a user might be interested in different things each time when using the system, but there is no way to allow the user to actively alter or adjust the recommended results. To address this issue, we propose the idea of "query-based recommendation" that allows a user to specify his/her search intention while exploring new items, thereby incorporating the concept of information retrieval into recommendation systems. Moreover, the idea is more desirable when the user intention can be expressed in different ways. Take music recommendation as an example: the proposed system allows a user to explore new song tracks by specifying either a track, an album, or an artist. To enable such heterogeneous queries in a recommender system, we present a novel technique called "Heterogeneous Preference Embedding" to encode user preference and query intention into low-dimensional vector spaces. Then, with simple search methods or similarity calculations, we can use the encoded representation of queries to generate recommendations. This method is fairly flexible and it is easy to add other types of information when available. Evaluations on three music listening datasets confirm the effectiveness of the proposed method over the state-of-the-art matrix factorization and network embedding methods.
{"title":"Query-based Music Recommendations via Preference Embedding","authors":"Chih-Ming Chen, Ming-Feng Tsai, Yu-Ching Lin, Yi-Hsuan Yang","doi":"10.1145/2959100.2959169","DOIUrl":"https://doi.org/10.1145/2959100.2959169","url":null,"abstract":"A common scenario considered in recommender systems is to predict a user's preferences on unseen items based on his/her preferences on observed items. A major limitation of this scenario is that a user might be interested in different things each time when using the system, but there is no way to allow the user to actively alter or adjust the recommended results. To address this issue, we propose the idea of \"query-based recommendation\" that allows a user to specify his/her search intention while exploring new items, thereby incorporating the concept of information retrieval into recommendation systems. Moreover, the idea is more desirable when the user intention can be expressed in different ways. Take music recommendation as an example: the proposed system allows a user to explore new song tracks by specifying either a track, an album, or an artist. To enable such heterogeneous queries in a recommender system, we present a novel technique called \"Heterogeneous Preference Embedding\" to encode user preference and query intention into low-dimensional vector spaces. Then, with simple search methods or similarity calculations, we can use the encoded representation of queries to generate recommendations. This method is fairly flexible and it is easy to add other types of information when available. Evaluations on three music listening datasets confirm the effectiveness of the proposed method over the state-of-the-art matrix factorization and network embedding methods.","PeriodicalId":315651,"journal":{"name":"Proceedings of the 10th ACM Conference on Recommender Systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129003081","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}
Short-length random walks on the bipartite user-item graph have recently been shown to provide accurate and diverse recommendations. Nonetheless, these approaches suffer from severe time and space requirements, which can be alleviated via random walk sampling, at the cost of reduced recommendation quality. In addition, these approaches ignore users' ratings, which further limits their expressiveness. In this paper, we introduce a computationally efficient graph-based approach for collaborative filtering based on short-path enumeration. Moreover, we propose three scoring functions based on the Bayesian paradigm that effectively exploit distributional aspects of the users' ratings. We experiment with seven publicly available datasets against state-of-the-art graph-based and matrix factorization approaches. Our empirical results demonstrate the effectiveness of the proposed approach, with significant improvements in most settings. Furthermore, analytical results demonstrate its efficiency compared to other graph-based approaches.
{"title":"Efficient Bayesian Methods for Graph-based Recommendation","authors":"Ramon Lopes, R. Assunção, Rodrygo L. T. Santos","doi":"10.1145/2959100.2959132","DOIUrl":"https://doi.org/10.1145/2959100.2959132","url":null,"abstract":"Short-length random walks on the bipartite user-item graph have recently been shown to provide accurate and diverse recommendations. Nonetheless, these approaches suffer from severe time and space requirements, which can be alleviated via random walk sampling, at the cost of reduced recommendation quality. In addition, these approaches ignore users' ratings, which further limits their expressiveness. In this paper, we introduce a computationally efficient graph-based approach for collaborative filtering based on short-path enumeration. Moreover, we propose three scoring functions based on the Bayesian paradigm that effectively exploit distributional aspects of the users' ratings. We experiment with seven publicly available datasets against state-of-the-art graph-based and matrix factorization approaches. Our empirical results demonstrate the effectiveness of the proposed approach, with significant improvements in most settings. Furthermore, analytical results demonstrate its efficiency compared to other graph-based approaches.","PeriodicalId":315651,"journal":{"name":"Proceedings of the 10th ACM Conference on Recommender Systems","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127619960","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}
Google Play is a seamless approach to digital entertainment on all of your devices. It gives you one place to find, enjoy and share your favorite entertainment, from apps to movies, music, books and more, on the web or any device. With more than 1 billion active users in 190+ countries around the world, Play is an important distribution platform for developers to build a global audience. More than 50 billion apps-have been downloaded from Google Play. However, generating personalized recommendations for different kind of content is a complex technical and product problem. Each of Play verticals (apps, games, books, movies, music) has different business goals, metrics to optimize, and user behavior. In this talk, we'll present an overview of how Play recommendations work across these verticals, how we evaluate our results, and the impact of deep neural networks in improving recommendations.
Google Play是你在所有设备上进行数字娱乐的无缝途径。它给你一个地方找到，享受和分享你最喜欢的娱乐，从应用程序到电影，音乐，书籍和更多，在网络或任何设备上。Play在全球190多个国家拥有超过10亿活跃用户，是开发者打造全球用户的重要分销平台。Google Play的应用程序下载量已经超过500亿次。然而，为不同类型的内容生成个性化推荐是一个复杂的技术和产品问题。每个Play垂直领域(游戏邦注:包括应用、游戏、书籍、电影和音乐)都有不同的商业目标、需要优化的指标和用户行为。在本次演讲中，我们将概述Play推荐如何在这些垂直领域中发挥作用，我们如何评估我们的结果，以及深度神经网络在改进推荐方面的影响。
{"title":"Multi-corpus Personalized Recommendations on Google Play","authors":"L. Koc, C. Master","doi":"10.1145/2959100.2959129","DOIUrl":"https://doi.org/10.1145/2959100.2959129","url":null,"abstract":"Google Play is a seamless approach to digital entertainment on all of your devices. It gives you one place to find, enjoy and share your favorite entertainment, from apps to movies, music, books and more, on the web or any device. With more than 1 billion active users in 190+ countries around the world, Play is an important distribution platform for developers to build a global audience. More than 50 billion apps-have been downloaded from Google Play. However, generating personalized recommendations for different kind of content is a complex technical and product problem. Each of Play verticals (apps, games, books, movies, music) has different business goals, metrics to optimize, and user behavior. In this talk, we'll present an overview of how Play recommendations work across these verticals, how we evaluate our results, and the impact of deep neural networks in improving recommendations.","PeriodicalId":315651,"journal":{"name":"Proceedings of the 10th ACM Conference on Recommender Systems","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125260415","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}
Foursquare recently launched Marsbot, an SMS-based app for local recommendations. Marsbot is an intelligent friend that lives in your pocket and learns about you through the places you go in the real world. While this product is aligned with Foursquare's long-standing mission to find the best places, it represents a new era in the way people interact with recommendation engines. The promise of the latest crop of personal assistants is get us information more quickly and seamlessly, but building them comes with many challenges. In this talk, we discuss why we built Marsbot and some of the many lessons learned along the way.
{"title":"Marsbot: Building a Personal Assistant","authors":"Max Sklar","doi":"10.1145/2959100.2959119","DOIUrl":"https://doi.org/10.1145/2959100.2959119","url":null,"abstract":"Foursquare recently launched Marsbot, an SMS-based app for local recommendations. Marsbot is an intelligent friend that lives in your pocket and learns about you through the places you go in the real world. While this product is aligned with Foursquare's long-standing mission to find the best places, it represents a new era in the way people interact with recommendation engines. The promise of the latest crop of personal assistants is get us information more quickly and seamlessly, but building them comes with many challenges. In this talk, we discuss why we built Marsbot and some of the many lessons learned along the way.","PeriodicalId":315651,"journal":{"name":"Proceedings of the 10th ACM Conference on Recommender Systems","volume":"83 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124629312","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}
Yu-Chin Juan, Yong Zhuang, Wei-Sheng Chin, Chih-Jen Lin
Click-through rate (CTR) prediction plays an important role in computational advertising. Models based on degree-2 polynomial mappings and factorization machines (FMs) are widely used for this task. Recently, a variant of FMs, field-aware factorization machines (FFMs), outperforms existing models in some world-wide CTR-prediction competitions. Based on our experiences in winning two of them, in this paper we establish FFMs as an effective method for classifying large sparse data including those from CTR prediction. First, we propose efficient implementations for training FFMs. Then we comprehensively analyze FFMs and compare this approach with competing models. Experiments show that FFMs are very useful for certain classification problems. Finally, we have released a package of FFMs for public use.
{"title":"Field-aware Factorization Machines for CTR Prediction","authors":"Yu-Chin Juan, Yong Zhuang, Wei-Sheng Chin, Chih-Jen Lin","doi":"10.1145/2959100.2959134","DOIUrl":"https://doi.org/10.1145/2959100.2959134","url":null,"abstract":"Click-through rate (CTR) prediction plays an important role in computational advertising. Models based on degree-2 polynomial mappings and factorization machines (FMs) are widely used for this task. Recently, a variant of FMs, field-aware factorization machines (FFMs), outperforms existing models in some world-wide CTR-prediction competitions. Based on our experiences in winning two of them, in this paper we establish FFMs as an effective method for classifying large sparse data including those from CTR prediction. First, we propose efficient implementations for training FFMs. Then we comprehensively analyze FFMs and compare this approach with competing models. Experiments show that FFMs are very useful for certain classification problems. Finally, we have released a package of FFMs for public use.","PeriodicalId":315651,"journal":{"name":"Proceedings of the 10th ACM Conference on Recommender Systems","volume":"273 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123419920","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}
Roberto Pagano, P. Cremonesi, M. Larson, Balázs Hidasi, D. Tikk, Alexandros Karatzoglou, Massimo Quadrana
A critical change has occurred in the status of context in recommender systems. In the past, context has been considered 'additional evidence'. This past picture is at odds with many present application domains, where user and item information is scarce. Such domains face continuous cold start conditions and must exploit session rather than user information. In this paper, we describe the `Contextual Turn?: the move towards context-driven recommendation algorithms for which context is critical, rather than additional. We cover application domains, algorithms that promise to address the challenges of context-driven recommendation, and the steps that the community has taken to tackle context-driven problems. Our goal is to point out the commonalities of context-driven problems, and urge the community to address the overarching challenges that context-driven recommendation poses.
{"title":"The Contextual Turn: from Context-Aware to Context-Driven Recommender Systems","authors":"Roberto Pagano, P. Cremonesi, M. Larson, Balázs Hidasi, D. Tikk, Alexandros Karatzoglou, Massimo Quadrana","doi":"10.1145/2959100.2959136","DOIUrl":"https://doi.org/10.1145/2959100.2959136","url":null,"abstract":"A critical change has occurred in the status of context in recommender systems. In the past, context has been considered 'additional evidence'. This past picture is at odds with many present application domains, where user and item information is scarce. Such domains face continuous cold start conditions and must exploit session rather than user information. In this paper, we describe the `Contextual Turn?: the move towards context-driven recommendation algorithms for which context is critical, rather than additional. We cover application domains, algorithms that promise to address the challenges of context-driven recommendation, and the steps that the community has taken to tackle context-driven problems. Our goal is to point out the commonalities of context-driven problems, and urge the community to address the overarching challenges that context-driven recommendation poses.","PeriodicalId":315651,"journal":{"name":"Proceedings of the 10th ACM Conference on Recommender Systems","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121641530","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}
Most evaluations of novel algorithmic contributions assess their accuracy in predicting what was withheld in an offline evaluation scenario. However, several doubts have been raised that standard offline evaluation practices are not appropriate to select the best algorithm for field deployment. The goal of this work is therefore to compare the offline and the online evaluation methodology with the same study participants, i.e. a within users experimental design. This paper presents empirical evidence that the ranking of algorithms based on offline accuracy measurements clearly contradicts the results from the online study with the same set of users. Thus the external validity of the most commonly applied evaluation methodology is not guaranteed.
{"title":"Contrasting Offline and Online Results when Evaluating Recommendation Algorithms","authors":"Marco Rossetti, Fabio Stella, M. Zanker","doi":"10.1145/2959100.2959176","DOIUrl":"https://doi.org/10.1145/2959100.2959176","url":null,"abstract":"Most evaluations of novel algorithmic contributions assess their accuracy in predicting what was withheld in an offline evaluation scenario. However, several doubts have been raised that standard offline evaluation practices are not appropriate to select the best algorithm for field deployment. The goal of this work is therefore to compare the offline and the online evaluation methodology with the same study participants, i.e. a within users experimental design. This paper presents empirical evidence that the ranking of algorithms based on offline accuracy measurements clearly contradicts the results from the online study with the same set of users. Thus the external validity of the most commonly applied evaluation methodology is not guaranteed.","PeriodicalId":315651,"journal":{"name":"Proceedings of the 10th ACM Conference on Recommender Systems","volume":"146 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114642369","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}
The intent-aware diversification framework was introduced initially in information retrieval and adopted to the context of recommender systems in the work of Vargas et al. The framework considers a set of aspects associated with items to be recommended. For instance, aspects may correspond to genres in movie recommendations. The framework depends on input aspect model consisting of item selection or relevance probabilities, given an aspect, and user intents, in the form of probabilities that the user is interested in each aspect. In this paper, we examine a number of input aspect models and evaluate the impact that different models have on the framework. In particular, we propose a constrained PLSA model that allows for interpretable output, in terms of known aspects, while achieving greater performance that the explicit co-occurrence counting method used in previous work. We evaluate the proposed models using a well-known MovieLens dataset for which item genres are available.
{"title":"Intent-Aware Diversification Using a Constrained PLSA","authors":"Jacek Wasilewski, N. Hurley","doi":"10.1145/2959100.2959177","DOIUrl":"https://doi.org/10.1145/2959100.2959177","url":null,"abstract":"The intent-aware diversification framework was introduced initially in information retrieval and adopted to the context of recommender systems in the work of Vargas et al. The framework considers a set of aspects associated with items to be recommended. For instance, aspects may correspond to genres in movie recommendations. The framework depends on input aspect model consisting of item selection or relevance probabilities, given an aspect, and user intents, in the form of probabilities that the user is interested in each aspect. In this paper, we examine a number of input aspect models and evaluate the impact that different models have on the framework. In particular, we propose a constrained PLSA model that allows for interpretable output, in terms of known aspects, while achieving greater performance that the explicit co-occurrence counting method used in previous work. We evaluate the proposed models using a well-known MovieLens dataset for which item genres are available.","PeriodicalId":315651,"journal":{"name":"Proceedings of the 10th ACM Conference on Recommender Systems","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114771622","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：