Pub Date : 2020-01-20DOI: 10.1093/benz/9780199773787.article.b00104587
Changfeng Sun, Han Liu, Meng Liu, Z. Ren, Tian Gan, Liqiang Nie
Paraphrase the content of the speaker’s words. This step is especially helpful in confirming that you and the speaker are on the same page. If you can put what the speaker says into your own words, it demonstrates you’ve listened attentively and allows the speaker to correct or clarify any misunderstanding. Express a connection between what the speaker said and what you heard. It could be a feeling, an experience, or a common principle shared with the other person.
{"title":"LARA","authors":"Changfeng Sun, Han Liu, Meng Liu, Z. Ren, Tian Gan, Liqiang Nie","doi":"10.1093/benz/9780199773787.article.b00104587","DOIUrl":"https://doi.org/10.1093/benz/9780199773787.article.b00104587","url":null,"abstract":"Paraphrase the content of the speaker’s words. This step is especially helpful in confirming that you and the speaker are on the same page. If you can put what the speaker says into your own words, it demonstrates you’ve listened attentively and allows the speaker to correct or clarify any misunderstanding. Express a connection between what the speaker said and what you heard. It could be a feeling, an experience, or a common principle shared with the other person.","PeriodicalId":319008,"journal":{"name":"Proceedings of the 13th International Conference on Web Search and Data Mining","volume":"479 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122784039","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}
We present HSDM, a full-day workshop on Health Search and Data Mining co-located with WSDM 2020's Health Day. This event builds on recent biomedical workshops in the NLP and ML communities but puts a clear emphasis on search and data mining (and their intersection) that is lacking in other venues. The program will include two keynote addresses by key opinion leaders in the clinical, search, and data mining domains. The technical program consists of 6 original research presentations. Finally, we will close with a panel discussion with keynote speakers, PC members, and the audience. This workshop aims to help consolidate the growing interest in biomedical applications of data-driven methods that becomes apparent all over the search and data mining spectrum, in WSDM's spirit of collaboration between industry and academia.
{"title":"Overview of the Health Search and Data Mining (HSDM 2020) Workshop","authors":"Carsten Eickhoff, Yubin Kim, Ryen W. White","doi":"10.1145/3336191.3371879","DOIUrl":"https://doi.org/10.1145/3336191.3371879","url":null,"abstract":"We present HSDM, a full-day workshop on Health Search and Data Mining co-located with WSDM 2020's Health Day. This event builds on recent biomedical workshops in the NLP and ML communities but puts a clear emphasis on search and data mining (and their intersection) that is lacking in other venues. The program will include two keynote addresses by key opinion leaders in the clinical, search, and data mining domains. The technical program consists of 6 original research presentations. Finally, we will close with a panel discussion with keynote speakers, PC members, and the audience. This workshop aims to help consolidate the growing interest in biomedical applications of data-driven methods that becomes apparent all over the search and data mining spectrum, in WSDM's spirit of collaboration between industry and academia.","PeriodicalId":319008,"journal":{"name":"Proceedings of the 13th International Conference on Web Search and Data Mining","volume":"199 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128680222","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}
Andrew Yates, Siddhant Arora, Xinyu Crystina Zhang, Wei Yang, Kevin Martin Jose, Jimmy J. Lin
We present Capreolus, a toolkit designed to facilitate end-to-end it ad hoc retrieval experiments with neural networks by providing implementations of prominent neural ranking models within a common framework. Our toolkit adopts a standard reranking architecture via tight integration with the Anserini toolkit for candidate document generation using standard bag-of-words approaches. Using Capreolus, we are able to reproduce Yang et al.'s recent SIGIR 2019 finding that, in a reranking scenario on the test collection from the TREC 2004 Robust Track, many neural retrieval models do not significantly outperform a strong query expansion baseline. Furthermore, we find that this holds true for five additional models implemented in Capreolus. We describe the architecture and design of our toolkit, which includes a Web interface to facilitate comparisons between rankings returned by different models.
{"title":"Capreolus: A Toolkit for End-to-End Neural Ad Hoc Retrieval","authors":"Andrew Yates, Siddhant Arora, Xinyu Crystina Zhang, Wei Yang, Kevin Martin Jose, Jimmy J. Lin","doi":"10.1145/3336191.3371868","DOIUrl":"https://doi.org/10.1145/3336191.3371868","url":null,"abstract":"We present Capreolus, a toolkit designed to facilitate end-to-end it ad hoc retrieval experiments with neural networks by providing implementations of prominent neural ranking models within a common framework. Our toolkit adopts a standard reranking architecture via tight integration with the Anserini toolkit for candidate document generation using standard bag-of-words approaches. Using Capreolus, we are able to reproduce Yang et al.'s recent SIGIR 2019 finding that, in a reranking scenario on the test collection from the TREC 2004 Robust Track, many neural retrieval models do not significantly outperform a strong query expansion baseline. Furthermore, we find that this holds true for five additional models implemented in Capreolus. We describe the architecture and design of our toolkit, which includes a Web interface to facilitate comparisons between rankings returned by different models.","PeriodicalId":319008,"journal":{"name":"Proceedings of the 13th International Conference on Web Search and Data Mining","volume":"214 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123762103","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}
A. Beheshti, V. Hashemi, S. Yakhchi, H. M. Nezhad, S. Ghafari, Jian Yang
Enabling the analysis of behavioral disorders over time in social networks, can help in suicide prevention, (school) bullying detection and extremist/criminal activity prediction. In this paper, we present a novel data analytics pipeline to enable the analysis of patterns of behavioral disorders on social networks. We present a Social Behavior Graph (sbGraph) model, to enable the analysis of factors that are driving behavior disorders over time. We use the golden standards in personality, behavior and attitude to build a domain specific Knowledge Base (KB). We use this domain knowledge to design cognitive services to automatically contextualize the raw social data and to prepare them for behavioral analytics. Then we introduce a pattern-based word embedding technique, namely personality2vec, on each feature extracted to build the sbGraph. The goal is to use mathematical embedding from a space with a dimension per feature to a continuous vector space which can be mapped to classes of behavioral disorders (such as cyber-bullying and radicalization) in the domain specific KB. We implement an interactive dashboard to enable social network analysts to analyze and understand the patterns of behavioral disorders over time. We focus on a motivating scenario in Australian government's office of the e-Safety commissioner, where the goal is to empowering all citizens to have safer, more positive experiences online.
{"title":"personality2vec: Enabling the Analysis of Behavioral Disorders in Social Networks","authors":"A. Beheshti, V. Hashemi, S. Yakhchi, H. M. Nezhad, S. Ghafari, Jian Yang","doi":"10.1145/3336191.3371865","DOIUrl":"https://doi.org/10.1145/3336191.3371865","url":null,"abstract":"Enabling the analysis of behavioral disorders over time in social networks, can help in suicide prevention, (school) bullying detection and extremist/criminal activity prediction. In this paper, we present a novel data analytics pipeline to enable the analysis of patterns of behavioral disorders on social networks. We present a Social Behavior Graph (sbGraph) model, to enable the analysis of factors that are driving behavior disorders over time. We use the golden standards in personality, behavior and attitude to build a domain specific Knowledge Base (KB). We use this domain knowledge to design cognitive services to automatically contextualize the raw social data and to prepare them for behavioral analytics. Then we introduce a pattern-based word embedding technique, namely personality2vec, on each feature extracted to build the sbGraph. The goal is to use mathematical embedding from a space with a dimension per feature to a continuous vector space which can be mapped to classes of behavioral disorders (such as cyber-bullying and radicalization) in the domain specific KB. We implement an interactive dashboard to enable social network analysts to analyze and understand the patterns of behavioral disorders over time. We focus on a motivating scenario in Australian government's office of the e-Safety commissioner, where the goal is to empowering all citizens to have safer, more positive experiences online.","PeriodicalId":319008,"journal":{"name":"Proceedings of the 13th International Conference on Web Search and Data Mining","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133809494","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}
Lixin Su, Jiafeng Guo, Yixing Fan, Yanyan Lan, Xueqi Cheng
Reading comprehension (RC) aims to locate a text span from a context passage to answer the given question. Despite the effectiveness of modern neural RC models, most existing work relies on maximum likelihood estimation (MLE) and ignores the structure of the output space. That is during training, one treats all the text spans do not match the ground truth as equally poor, leading to overconfident predictions on ground truth labels and reduced generalization ability in test. One way to bridge the gap between training and test is to take into account the task reward of alternative outputs using the reinforcement learning (RL) algorithms, which is often deficient in optimization as compared with MLE. In this paper, we propose a new learning criterion for the RC task which combines the merits of both MLE and RL-based methods. Specifically, we show that we are able to derive the distribution of the outputs, i.e., label distribution, using their corresponding task rewards based on the decomposition property of the RC problem. We then optimize the RC model by directly learning towards the auxiliary label distribution, instead of the ground truth label, using the MLE framework. In this way, we can make use of the structure of the output space for better generalization (as RL) via efficient optimization (as MLE). We name our approach as Label Distribution augmented MLE (LD-MLE), which is a general learning criterion that could be adopted by almost all the existing RC models. Experiments on three representative benchmark datasets demonstrate that RC models learned with the LD-MLE criterion can achieve consistently improved results over those based on the traditional MLE and RL-based criteria.
{"title":"Label Distribution Augmented Maximum Likelihood Estimation for Reading Comprehension","authors":"Lixin Su, Jiafeng Guo, Yixing Fan, Yanyan Lan, Xueqi Cheng","doi":"10.1145/3336191.3371835","DOIUrl":"https://doi.org/10.1145/3336191.3371835","url":null,"abstract":"Reading comprehension (RC) aims to locate a text span from a context passage to answer the given question. Despite the effectiveness of modern neural RC models, most existing work relies on maximum likelihood estimation (MLE) and ignores the structure of the output space. That is during training, one treats all the text spans do not match the ground truth as equally poor, leading to overconfident predictions on ground truth labels and reduced generalization ability in test. One way to bridge the gap between training and test is to take into account the task reward of alternative outputs using the reinforcement learning (RL) algorithms, which is often deficient in optimization as compared with MLE. In this paper, we propose a new learning criterion for the RC task which combines the merits of both MLE and RL-based methods. Specifically, we show that we are able to derive the distribution of the outputs, i.e., label distribution, using their corresponding task rewards based on the decomposition property of the RC problem. We then optimize the RC model by directly learning towards the auxiliary label distribution, instead of the ground truth label, using the MLE framework. In this way, we can make use of the structure of the output space for better generalization (as RL) via efficient optimization (as MLE). We name our approach as Label Distribution augmented MLE (LD-MLE), which is a general learning criterion that could be adopted by almost all the existing RC models. Experiments on three representative benchmark datasets demonstrate that RC models learned with the LD-MLE criterion can achieve consistently improved results over those based on the traditional MLE and RL-based criteria.","PeriodicalId":319008,"journal":{"name":"Proceedings of the 13th International Conference on Web Search and Data Mining","volume":"473 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131071315","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}
Social advertising exploits the interconnectivity of users in social networks to spread advertisement and generate user engagements. A lot of research has focused on how to select the best subset of users in a social network to maximize the number of engagements or the generated revenue of the advertisement. However, there is a lack of studies that consider the advertiser's value-per-engagement, i.e., how much an advertiser is maximally willing to pay for each engagement. Prior work on social advertising is based on the classical framework of influence maximization. In this paper, we propose a model where advertisers compete in an auction mechanism for the influential users within a social network. The auction mechanism can dynamically determine payments for advertisers based on their reported values. The main problem is to find auctions which incentivize advertisers to truthfully reveal their values, and also respect each advertiser's budget constraint. To tackle this problem, we propose a new truthful auction mechanism called TSA. Compared with existing approaches on real and synthetic datasets, TSA performs significantly better in terms of generated revenue.
{"title":"TSA: A Truthful Mechanism for Social Advertising","authors":"Tobias Grubenmann, Reynold Cheng, L. Lakshmanan","doi":"10.1145/3336191.3371809","DOIUrl":"https://doi.org/10.1145/3336191.3371809","url":null,"abstract":"Social advertising exploits the interconnectivity of users in social networks to spread advertisement and generate user engagements. A lot of research has focused on how to select the best subset of users in a social network to maximize the number of engagements or the generated revenue of the advertisement. However, there is a lack of studies that consider the advertiser's value-per-engagement, i.e., how much an advertiser is maximally willing to pay for each engagement. Prior work on social advertising is based on the classical framework of influence maximization. In this paper, we propose a model where advertisers compete in an auction mechanism for the influential users within a social network. The auction mechanism can dynamically determine payments for advertisers based on their reported values. The main problem is to find auctions which incentivize advertisers to truthfully reveal their values, and also respect each advertiser's budget constraint. To tackle this problem, we propose a new truthful auction mechanism called TSA. Compared with existing approaches on real and synthetic datasets, TSA performs significantly better in terms of generated revenue.","PeriodicalId":319008,"journal":{"name":"Proceedings of the 13th International Conference on Web Search and Data Mining","volume":"404 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115318905","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}
Pub Date : 2020-01-20DOI: 10.1007/978-3-319-67199-4_103993
Tobias Grubenmann, Reynold Cheng, L. Lakshmanan
{"title":"TSA","authors":"Tobias Grubenmann, Reynold Cheng, L. Lakshmanan","doi":"10.1007/978-3-319-67199-4_103993","DOIUrl":"https://doi.org/10.1007/978-3-319-67199-4_103993","url":null,"abstract":"","PeriodicalId":319008,"journal":{"name":"Proceedings of the 13th International Conference on Web Search and Data Mining","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115544515","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}
Sebastian Bruch, Shuguang Han, Michael Bendersky, Marc Najork
Learning to Rank, a central problem in information retrieval, is a class of machine learning algorithms that formulate ranking as an optimization task. The objective is to learn a function that produces an ordering of a set of documents in such a way that the utility of the entire ordered list is maximized. Learning-to-rank methods do so by learning a function that computes a score for each document in the set. A ranked list is then compiled by sorting documents according to their scores. While such a deterministic mapping of scores to permutations makes sense during inference where stability of ranked lists is required, we argue that its greedy nature during training leads to less robust models. This is particularly problematic when the loss function under optimization---in agreement with ranking metrics---largely penalizes incorrect rankings and does not take into account the distribution of raw scores. In this work, we present a stochastic framework where, instead of a deterministic derivation of permutations from raw scores, permutations are sampled from a distribution defined by raw scores. Our proposed sampling method is differentiable and works well with gradient descent optimizers. We analytically study our proposed method and demonstrate when and why it leads to model robustness. We also show empirically, through experiments on publicly available learning-to-rank datasets, that the application of our proposed method to a class of ranking loss functions leads to significant model quality improvements.
{"title":"A Stochastic Treatment of Learning to Rank Scoring Functions","authors":"Sebastian Bruch, Shuguang Han, Michael Bendersky, Marc Najork","doi":"10.1145/3336191.3371844","DOIUrl":"https://doi.org/10.1145/3336191.3371844","url":null,"abstract":"Learning to Rank, a central problem in information retrieval, is a class of machine learning algorithms that formulate ranking as an optimization task. The objective is to learn a function that produces an ordering of a set of documents in such a way that the utility of the entire ordered list is maximized. Learning-to-rank methods do so by learning a function that computes a score for each document in the set. A ranked list is then compiled by sorting documents according to their scores. While such a deterministic mapping of scores to permutations makes sense during inference where stability of ranked lists is required, we argue that its greedy nature during training leads to less robust models. This is particularly problematic when the loss function under optimization---in agreement with ranking metrics---largely penalizes incorrect rankings and does not take into account the distribution of raw scores. In this work, we present a stochastic framework where, instead of a deterministic derivation of permutations from raw scores, permutations are sampled from a distribution defined by raw scores. Our proposed sampling method is differentiable and works well with gradient descent optimizers. We analytically study our proposed method and demonstrate when and why it leads to model robustness. We also show empirically, through experiments on publicly available learning-to-rank datasets, that the application of our proposed method to a class of ranking loss functions leads to significant model quality improvements.","PeriodicalId":319008,"journal":{"name":"Proceedings of the 13th International Conference on Web Search and Data Mining","volume":"146 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122444888","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}
Large scale knowledge graph (KG) has attracted wide attentions in both academia and industry recently. However, due to the complexity of SPARQL syntax and massive volume of real KG, it remains difficult for ordinary users to access KG. In this demo, we present VISION-KG, a topic-centric visualization system to help users navigate KG easily via entity summarization and entity clustering. Given a query entity v0, VISION-KG summarizes the induced subgraph of v0's neighbor nodes via our proposed facts ranking method that measures importance, relatedness and diversity. Moreover, to achieve conciseness, we split the summarized graph into several topic-centric summarized subgraph according to semantic and structural similarities among entities. We will demonstrate how VISION-KG provides a user-friendly visualization interface for navigating KG.
{"title":"VISION-KG: Topic-centric Visualization System for Summarizing Knowledge Graph","authors":"Jiaqi Wei, Shuo Han, Lei Zou","doi":"10.1145/3336191.3371863","DOIUrl":"https://doi.org/10.1145/3336191.3371863","url":null,"abstract":"Large scale knowledge graph (KG) has attracted wide attentions in both academia and industry recently. However, due to the complexity of SPARQL syntax and massive volume of real KG, it remains difficult for ordinary users to access KG. In this demo, we present VISION-KG, a topic-centric visualization system to help users navigate KG easily via entity summarization and entity clustering. Given a query entity v0, VISION-KG summarizes the induced subgraph of v0's neighbor nodes via our proposed facts ranking method that measures importance, relatedness and diversity. Moreover, to achieve conciseness, we split the summarized graph into several topic-centric summarized subgraph according to semantic and structural similarities among entities. We will demonstrate how VISION-KG provides a user-friendly visualization interface for navigating KG.","PeriodicalId":319008,"journal":{"name":"Proceedings of the 13th International Conference on Web Search and Data Mining","volume":"85 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122594832","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}
We propose a personalized user recommendation framework for content curation platforms that models preferences for both users and the items they engage with simultaneously. In this way, user preferences for specific item types (e.g., fantasy novels) can be balanced with user specialties (e.g., reviewing novels with strong female protagonists). In particular, the proposed model has three unique characteristics: (i) it simultaneously learns both user-item and user-user preferences through a multi-aspect autoencoder model; (ii) it fuses the latent representations of user preferences on users and items to construct shared factors through an adversarial framework; and (iii) it incorporates an attention layer to produce weighted aggregations of different latent representations, leading to improved personalized recommendation of users and items. Through experiments against state-of-the-art models, we find the proposed framework leads to a 18.43% (Goodreads) and 6.14% (Spotify) improvement in top-k user recommendation.
{"title":"User Recommendation in Content Curation Platforms","authors":"Jianling Wang, Ziwei Zhu, James Caverlee","doi":"10.1145/3336191.3371822","DOIUrl":"https://doi.org/10.1145/3336191.3371822","url":null,"abstract":"We propose a personalized user recommendation framework for content curation platforms that models preferences for both users and the items they engage with simultaneously. In this way, user preferences for specific item types (e.g., fantasy novels) can be balanced with user specialties (e.g., reviewing novels with strong female protagonists). In particular, the proposed model has three unique characteristics: (i) it simultaneously learns both user-item and user-user preferences through a multi-aspect autoencoder model; (ii) it fuses the latent representations of user preferences on users and items to construct shared factors through an adversarial framework; and (iii) it incorporates an attention layer to produce weighted aggregations of different latent representations, leading to improved personalized recommendation of users and items. Through experiments against state-of-the-art models, we find the proposed framework leads to a 18.43% (Goodreads) and 6.14% (Spotify) improvement in top-k user recommendation.","PeriodicalId":319008,"journal":{"name":"Proceedings of the 13th International Conference on Web Search and Data Mining","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127509683","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：