Search systems are often used to support learning-oriented goals. This trend has given rise to the “search-as-learning” movement, which proposes that search systems should be designed to support learning. To this end, an important research question is: How does a searcher’s type of learning objective (LO) influence their trajectory (or pathway) toward that objective? We report on a lab study (N = 36) in which participants gathered information to meet a specific type of LO. To characterize LOs and pathways, we leveraged Anderson and Krathwohl’s (A&K’s) taxonomy [3]. A&K’s taxonomy situates LOs at the intersection of two orthogonal dimensions: (1) cognitive process (CP) (remember, understand, apply, analyze, evaluate, and create) and (2) knowledge type (factual, conceptual, procedural, and metacognitive knowledge). Participants completed learning-oriented search tasks that varied along three CPs (apply, evaluate, and create) and three knowledge types (factual, conceptual, and procedural knowledge). A pathway is defined as a sequence of learning instances (e.g., subgoals) that were also each classified into cells from A&K’s taxonomy. Our study used a think-aloud protocol, and pathways were generated through a qualitative analysis of participants’ think-aloud comments and recorded screen activities. We investigate three research questions. First, in RQ1, we study the impact of the LO on pathway characteristics (e.g., pathway length). Second, in RQ2, we study the impact of the LO on the types of A&K cells traversed along the pathway. Third, in RQ3, we study common and uncommon transitions between A&K cells along pathways conditioned on the knowledge type of the objective. We discuss implications of our results for designing search systems to support learning.
{"title":"Understanding the “Pathway” Towards a Searcher’s Learning Objective","authors":"Kelsey Urgo, Jaime Arguello","doi":"10.1145/3495222","DOIUrl":"https://doi.org/10.1145/3495222","url":null,"abstract":"Search systems are often used to support learning-oriented goals. This trend has given rise to the “search-as-learning” movement, which proposes that search systems should be designed to support learning. To this end, an important research question is: How does a searcher’s type of learning objective (LO) influence their trajectory (or pathway) toward that objective? We report on a lab study (N = 36) in which participants gathered information to meet a specific type of LO. To characterize LOs and pathways, we leveraged Anderson and Krathwohl’s (A&K’s) taxonomy [3]. A&K’s taxonomy situates LOs at the intersection of two orthogonal dimensions: (1) cognitive process (CP) (remember, understand, apply, analyze, evaluate, and create) and (2) knowledge type (factual, conceptual, procedural, and metacognitive knowledge). Participants completed learning-oriented search tasks that varied along three CPs (apply, evaluate, and create) and three knowledge types (factual, conceptual, and procedural knowledge). A pathway is defined as a sequence of learning instances (e.g., subgoals) that were also each classified into cells from A&K’s taxonomy. Our study used a think-aloud protocol, and pathways were generated through a qualitative analysis of participants’ think-aloud comments and recorded screen activities. We investigate three research questions. First, in RQ1, we study the impact of the LO on pathway characteristics (e.g., pathway length). Second, in RQ2, we study the impact of the LO on the types of A&K cells traversed along the pathway. Third, in RQ3, we study common and uncommon transitions between A&K cells along pathways conditioned on the knowledge type of the objective. We discuss implications of our results for designing search systems to support learning.","PeriodicalId":6934,"journal":{"name":"ACM Transactions on Information Systems (TOIS)","volume":"1 1","pages":"1 - 43"},"PeriodicalIF":0.0,"publicationDate":"2022-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84006104","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}
Relation extraction (RE), an important information extraction task, faced the great challenge brought by limited annotation data. To this end, distant supervision was proposed to automatically label RE data, and thus largely increased the number of annotated instances. Unfortunately, lots of noise relation annotations brought by automatic labeling become a new obstacle. Some recent studies have shown that the teacher-student framework of knowledge distillation can alleviate the interference of noise relation annotations via label softening. Nevertheless, we find that they still suffer from two problems: propagation of inaccurate dark knowledge and constraint of a unified distillation temperature. In this article, we propose a simple and effective Multi-instance Dynamic Temperature Distillation (MiDTD) framework, which is model-agnostic and mainly involves two modules: multi-instance target fusion (MiTF) and dynamic temperature regulation (DTR). MiTF combines the teacher’s predictions for multiple sentences with the same entity pair to amend the inaccurate dark knowledge in each student’s target. DTR allocates alterable distillation temperatures to different training instances to enable the softness of most student’s targets to be regulated to a moderate range. In experiments, we construct three concrete MiDTD instantiations with BERT, PCNN, and BiLSTM-based RE models, and the distilled students significantly outperform their teachers and the state-of-the-art (SOTA) methods.
{"title":"MiDTD: A Simple and Effective Distillation Framework for Distantly Supervised Relation Extraction","authors":"Rui Li, Cheng Yang, Tingwei Li, Sen Su","doi":"10.1145/3503917","DOIUrl":"https://doi.org/10.1145/3503917","url":null,"abstract":"Relation extraction (RE), an important information extraction task, faced the great challenge brought by limited annotation data. To this end, distant supervision was proposed to automatically label RE data, and thus largely increased the number of annotated instances. Unfortunately, lots of noise relation annotations brought by automatic labeling become a new obstacle. Some recent studies have shown that the teacher-student framework of knowledge distillation can alleviate the interference of noise relation annotations via label softening. Nevertheless, we find that they still suffer from two problems: propagation of inaccurate dark knowledge and constraint of a unified distillation temperature. In this article, we propose a simple and effective Multi-instance Dynamic Temperature Distillation (MiDTD) framework, which is model-agnostic and mainly involves two modules: multi-instance target fusion (MiTF) and dynamic temperature regulation (DTR). MiTF combines the teacher’s predictions for multiple sentences with the same entity pair to amend the inaccurate dark knowledge in each student’s target. DTR allocates alterable distillation temperatures to different training instances to enable the softness of most student’s targets to be regulated to a moderate range. In experiments, we construct three concrete MiDTD instantiations with BERT, PCNN, and BiLSTM-based RE models, and the distilled students significantly outperform their teachers and the state-of-the-art (SOTA) methods.","PeriodicalId":6934,"journal":{"name":"ACM Transactions on Information Systems (TOIS)","volume":"8 1","pages":"1 - 32"},"PeriodicalIF":0.0,"publicationDate":"2022-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77071427","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}
Peng Zhang, Baoxi Liu, T. Lu, X. Ding, Hansu Gu, Ning Gu
User-generated contents (UGC) in social media are the direct expression of users’ interests, preferences, and opinions. User behavior prediction based on UGC has increasingly been investigated in recent years. Compared to learning a person’s behavioral patterns in each social media site separately, jointly predicting user behavior in multiple social media sites and complementing each other (cross-site user behavior prediction) can be more accurate. However, cross-site user behavior prediction based on UGC is a challenging task due to the difficulty of cross-site data sampling, the complexity of UGC modeling, and uncertainty of knowledge sharing among different sites. For these problems, we propose a Cross-Site Multi-Task (CSMT) learning method to jointly predict user behavior in multiple social media sites. CSMT mainly derives from the hierarchical attention network and multi-task learning. Using this method, the UGC in each social media site can obtain fine-grained representations in terms of words, topics, posts, hashtags, and time slices as well as the relevances among them, and prediction tasks in different social media sites can be jointly implemented and complement each other. By utilizing two cross-site datasets sampled from Weibo, Douban, Facebook, and Twitter, we validate our method’s superiority on several classification metrics compared with existing related methods.
{"title":"Jointly Predicting Future Content in Multiple Social Media Sites Based on Multi-task Learning","authors":"Peng Zhang, Baoxi Liu, T. Lu, X. Ding, Hansu Gu, Ning Gu","doi":"10.1145/3495530","DOIUrl":"https://doi.org/10.1145/3495530","url":null,"abstract":"User-generated contents (UGC) in social media are the direct expression of users’ interests, preferences, and opinions. User behavior prediction based on UGC has increasingly been investigated in recent years. Compared to learning a person’s behavioral patterns in each social media site separately, jointly predicting user behavior in multiple social media sites and complementing each other (cross-site user behavior prediction) can be more accurate. However, cross-site user behavior prediction based on UGC is a challenging task due to the difficulty of cross-site data sampling, the complexity of UGC modeling, and uncertainty of knowledge sharing among different sites. For these problems, we propose a Cross-Site Multi-Task (CSMT) learning method to jointly predict user behavior in multiple social media sites. CSMT mainly derives from the hierarchical attention network and multi-task learning. Using this method, the UGC in each social media site can obtain fine-grained representations in terms of words, topics, posts, hashtags, and time slices as well as the relevances among them, and prediction tasks in different social media sites can be jointly implemented and complement each other. By utilizing two cross-site datasets sampled from Weibo, Douban, Facebook, and Twitter, we validate our method’s superiority on several classification metrics compared with existing related methods.","PeriodicalId":6934,"journal":{"name":"ACM Transactions on Information Systems (TOIS)","volume":"43 1","pages":"1 - 28"},"PeriodicalIF":0.0,"publicationDate":"2022-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90804755","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}
Weiren Yu, J. Mccann, Chengyuan Zhang, H. Ferhatosmanoğlu
SimRank is an attractive link-based similarity measure used in fertile fields of Web search and sociometry. However, the existing deterministic method by Kusumoto et al. [24] for retrieving SimRank does not always produce high-quality similarity results, as it fails to accurately obtain diagonal correction matrix D. Moreover, SimRank has a “connectivity trait” problem: increasing the number of paths between a pair of nodes would decrease its similarity score. The best-known remedy, SimRank++ [1], cannot completely fix this problem, since its score would still be zero if there are no common in-neighbors between two nodes. In this article, we study fast high-quality link-based similarity search on billion-scale graphs. (1) We first devise a “varied-D” method to accurately compute SimRank in linear memory. We also aggregate duplicate computations, which reduces the time of [24] from quadratic to linear in the number of iterations. (2) We propose a novel “cosine-based” SimRank model to circumvent the “connectivity trait” problem. (3) To substantially speed up the partial-pairs “cosine-based” SimRank search on large graphs, we devise an efficient dimensionality reduction algorithm, PSR#, with guaranteed accuracy. (4) We give mathematical insights to the semantic difference between SimRank and its variant, and correct an argument in [24] that “if D is replaced by a scaled identity matrix (1-Ɣ)I, their top-K rankings will not be affected much”. (5) We propose a novel method that can accurately convert from Li et al. SimRank ~{S} to Jeh and Widom’s SimRank S. (6) We propose GSR#, a generalisation of our “cosine-based” SimRank model, to quantify pairwise similarities across two distinct graphs, unlike SimRank that would assess nodes across two graphs as completely dissimilar. Extensive experiments on various datasets demonstrate the superiority of our proposed approaches in terms of high search quality, computational efficiency, accuracy, and scalability on billion-edge graphs.
{"title":"Scaling High-Quality Pairwise Link-Based Similarity Retrieval on Billion-Edge Graphs","authors":"Weiren Yu, J. Mccann, Chengyuan Zhang, H. Ferhatosmanoğlu","doi":"10.1145/3495209","DOIUrl":"https://doi.org/10.1145/3495209","url":null,"abstract":"SimRank is an attractive link-based similarity measure used in fertile fields of Web search and sociometry. However, the existing deterministic method by Kusumoto et al. [24] for retrieving SimRank does not always produce high-quality similarity results, as it fails to accurately obtain diagonal correction matrix D. Moreover, SimRank has a “connectivity trait” problem: increasing the number of paths between a pair of nodes would decrease its similarity score. The best-known remedy, SimRank++ [1], cannot completely fix this problem, since its score would still be zero if there are no common in-neighbors between two nodes. In this article, we study fast high-quality link-based similarity search on billion-scale graphs. (1) We first devise a “varied-D” method to accurately compute SimRank in linear memory. We also aggregate duplicate computations, which reduces the time of [24] from quadratic to linear in the number of iterations. (2) We propose a novel “cosine-based” SimRank model to circumvent the “connectivity trait” problem. (3) To substantially speed up the partial-pairs “cosine-based” SimRank search on large graphs, we devise an efficient dimensionality reduction algorithm, PSR#, with guaranteed accuracy. (4) We give mathematical insights to the semantic difference between SimRank and its variant, and correct an argument in [24] that “if D is replaced by a scaled identity matrix (1-Ɣ)I, their top-K rankings will not be affected much”. (5) We propose a novel method that can accurately convert from Li et al. SimRank ~{S} to Jeh and Widom’s SimRank S. (6) We propose GSR#, a generalisation of our “cosine-based” SimRank model, to quantify pairwise similarities across two distinct graphs, unlike SimRank that would assess nodes across two graphs as completely dissimilar. Extensive experiments on various datasets demonstrate the superiority of our proposed approaches in terms of high search quality, computational efficiency, accuracy, and scalability on billion-edge graphs.","PeriodicalId":6934,"journal":{"name":"ACM Transactions on Information Systems (TOIS)","volume":"34 1","pages":"1 - 45"},"PeriodicalIF":0.0,"publicationDate":"2022-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77502118","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}
In conversational search, agents can interact with users by asking clarifying questions to increase their chance of finding better results. Many recent works and shared tasks in both natural language processing and information retrieval communities have focused on identifying the need to ask clarifying questions and methodologies of generating them. These works assume that asking a clarifying question is a safe alternative to retrieving results. As existing conversational search models are far from perfect, it is possible and common that they could retrieve/generate bad clarifying questions. Asking too many clarifying questions can also drain a user’s patience when the user prefers searching efficiency over correctness. Hence, these models can backfire and harm a user’s search experience due to these risks from asking clarifying questions. In this work, we propose a simulation framework to simulate the risk of asking questions in conversational search and further revise a risk-aware conversational search model to control the risk. We show the model’s robustness and effectiveness through extensive experiments on three conversational datasets — MSDialog, Ubuntu Dialog Corpus, and Opendialkg — in which we compare it with multiple baselines. We show that the risk-control module can work with two different re-ranker models and outperform all of the baselines in most of our experiments.
{"title":"Simulating and Modeling the Risk of Conversational Search","authors":"Zhenduo Wang, Qingyao Ai","doi":"10.1145/3507357","DOIUrl":"https://doi.org/10.1145/3507357","url":null,"abstract":"In conversational search, agents can interact with users by asking clarifying questions to increase their chance of finding better results. Many recent works and shared tasks in both natural language processing and information retrieval communities have focused on identifying the need to ask clarifying questions and methodologies of generating them. These works assume that asking a clarifying question is a safe alternative to retrieving results. As existing conversational search models are far from perfect, it is possible and common that they could retrieve/generate bad clarifying questions. Asking too many clarifying questions can also drain a user’s patience when the user prefers searching efficiency over correctness. Hence, these models can backfire and harm a user’s search experience due to these risks from asking clarifying questions. In this work, we propose a simulation framework to simulate the risk of asking questions in conversational search and further revise a risk-aware conversational search model to control the risk. We show the model’s robustness and effectiveness through extensive experiments on three conversational datasets — MSDialog, Ubuntu Dialog Corpus, and Opendialkg — in which we compare it with multiple baselines. We show that the risk-control module can work with two different re-ranker models and outperform all of the baselines in most of our experiments.","PeriodicalId":6934,"journal":{"name":"ACM Transactions on Information Systems (TOIS)","volume":"64 1","pages":"1 - 33"},"PeriodicalIF":0.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90345383","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}
Jiashu Zhao, J. Huang, Hongbo Deng, Yi Chang, Long Xia
In this article, we propose a Latent Dirichlet Allocation– (LDA) based topic-graph probabilistic personalization model for Web search. This model represents a user graph in a latent topic graph and simultaneously estimates the probabilities that the user is interested in the topics, as well as the probabilities that the user is not interested in the topics. For a given query issued by the user, the webpages that have higher relevancy to the interested topics are promoted, and the webpages more relevant to the non-interesting topics are penalized. In particular, we simulate a user’s search intent by building two profiles: A positive user profile for the probabilities of the user is interested in the topics and a corresponding negative user profile for the probabilities of being not interested in the the topics. The profiles are estimated based on the user’s search logs. A clicked webpage is assumed to include interesting topics. A skipped (viewed but not clicked) webpage is assumed to cover some non-interesting topics to the user. Such estimations are performed in the latent topic space generated by LDA. Moreover, a new approach is proposed to estimate the correlation between a given query and the user’s search history so as to determine how much personalization should be considered for the query. We compare our proposed models with several strong baselines including state-of-the-art personalization approaches. Experiments conducted on a large-scale real user search log collection illustrate the effectiveness of the proposed models.
{"title":"Are Topics Interesting or Not? An LDA-based Topic-graph Probabilistic Model for Web Search Personalization","authors":"Jiashu Zhao, J. Huang, Hongbo Deng, Yi Chang, Long Xia","doi":"10.1145/3476106","DOIUrl":"https://doi.org/10.1145/3476106","url":null,"abstract":"In this article, we propose a Latent Dirichlet Allocation– (LDA) based topic-graph probabilistic personalization model for Web search. This model represents a user graph in a latent topic graph and simultaneously estimates the probabilities that the user is interested in the topics, as well as the probabilities that the user is not interested in the topics. For a given query issued by the user, the webpages that have higher relevancy to the interested topics are promoted, and the webpages more relevant to the non-interesting topics are penalized. In particular, we simulate a user’s search intent by building two profiles: A positive user profile for the probabilities of the user is interested in the topics and a corresponding negative user profile for the probabilities of being not interested in the the topics. The profiles are estimated based on the user’s search logs. A clicked webpage is assumed to include interesting topics. A skipped (viewed but not clicked) webpage is assumed to cover some non-interesting topics to the user. Such estimations are performed in the latent topic space generated by LDA. Moreover, a new approach is proposed to estimate the correlation between a given query and the user’s search history so as to determine how much personalization should be considered for the query. We compare our proposed models with several strong baselines including state-of-the-art personalization approaches. Experiments conducted on a large-scale real user search log collection illustrate the effectiveness of the proposed models.","PeriodicalId":6934,"journal":{"name":"ACM Transactions on Information Systems (TOIS)","volume":"14 1","pages":"1 - 24"},"PeriodicalIF":0.0,"publicationDate":"2021-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83646035","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}
Yang Deng, Yaliang Li, Wenxuan Zhang, Bolin Ding, W. Lam
Recently, Product Question Answering (PQA) on E-Commerce platforms has attracted increasing attention as it can act as an intelligent online shopping assistant and improve the customer shopping experience. Its key function, automatic answer generation for product-related questions, has been studied by aiming to generate content-preserving while question-related answers. However, an important characteristic of PQA, i.e., personalization, is neglected by existing methods. It is insufficient to provide the same “completely summarized” answer to all customers, since many customers are more willing to see personalized answers with customized information only for themselves, by taking into consideration their own preferences toward product aspects or information needs. To tackle this challenge, we propose a novel Personalized Answer GEneration method with multi-perspective preference modeling, which explores historical user-generated contents to model user preference for generating personalized answers in PQA. Specifically, we first retrieve question-related user history as external knowledge to model knowledge-level user preference. Then, we leverage the Gaussian Softmax distribution model to capture latent aspect-level user preference. Finally, we develop a persona-aware pointer network to generate personalized answers in terms of both content and style by utilizing personal user preference and dynamic user vocabulary. Experimental results on real-world E-Commerce QA datasets demonstrate that the proposed method outperforms existing methods by generating informative and customized answers and show that answer generation in E-Commerce can benefit from personalization.
{"title":"Toward Personalized Answer Generation in E-Commerce via Multi-perspective Preference Modeling","authors":"Yang Deng, Yaliang Li, Wenxuan Zhang, Bolin Ding, W. Lam","doi":"10.1145/3507782","DOIUrl":"https://doi.org/10.1145/3507782","url":null,"abstract":"Recently, Product Question Answering (PQA) on E-Commerce platforms has attracted increasing attention as it can act as an intelligent online shopping assistant and improve the customer shopping experience. Its key function, automatic answer generation for product-related questions, has been studied by aiming to generate content-preserving while question-related answers. However, an important characteristic of PQA, i.e., personalization, is neglected by existing methods. It is insufficient to provide the same “completely summarized” answer to all customers, since many customers are more willing to see personalized answers with customized information only for themselves, by taking into consideration their own preferences toward product aspects or information needs. To tackle this challenge, we propose a novel Personalized Answer GEneration method with multi-perspective preference modeling, which explores historical user-generated contents to model user preference for generating personalized answers in PQA. Specifically, we first retrieve question-related user history as external knowledge to model knowledge-level user preference. Then, we leverage the Gaussian Softmax distribution model to capture latent aspect-level user preference. Finally, we develop a persona-aware pointer network to generate personalized answers in terms of both content and style by utilizing personal user preference and dynamic user vocabulary. Experimental results on real-world E-Commerce QA datasets demonstrate that the proposed method outperforms existing methods by generating informative and customized answers and show that answer generation in E-Commerce can benefit from personalization.","PeriodicalId":6934,"journal":{"name":"ACM Transactions on Information Systems (TOIS)","volume":"54 1","pages":"1 - 28"},"PeriodicalIF":0.0,"publicationDate":"2021-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80427754","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}
Surong Yan, Kwei-Jay Lin, Xiaolin Zheng, Haosen Wang
Explicit and implicit knowledge about users and items have been used to describe complex and heterogeneous side information for recommender systems (RSs). Many existing methods use knowledge graph embedding (KGE) to learn the representation of a user-item knowledge graph (KG) in low-dimensional space. In this article, we propose a lightweight end-to-end joint learning framework for fusing the tasks of KGE and RSs at the model level. Our method proposes a lightweight KG embedding method by using bidirectional bijection relation-type modeling to enable scalability for large graphs while using self-adaptive negative sampling to optimize negative sample generating. Our method further generates the integrated views for users and items based on relation-types to explicitly model users’ preferences and items’ features, respectively. Finally, we add virtual “recommendation” relations between the integrated views of users and items to model the preferences of users on items, seamlessly integrating RS with user-item KG over a unified graph. Experimental results on multiple datasets and benchmarks show that our method can achieve a better accuracy of recommendation compared with existing state-of-the-art methods. Complexity and runtime analysis suggests that our method can gain a lower time and space complexity than most of existing methods and improve scalability.
{"title":"LkeRec: Toward Lightweight End-to-End Joint Representation Learning for Building Accurate and Effective Recommendation","authors":"Surong Yan, Kwei-Jay Lin, Xiaolin Zheng, Haosen Wang","doi":"10.1145/3486673","DOIUrl":"https://doi.org/10.1145/3486673","url":null,"abstract":"Explicit and implicit knowledge about users and items have been used to describe complex and heterogeneous side information for recommender systems (RSs). Many existing methods use knowledge graph embedding (KGE) to learn the representation of a user-item knowledge graph (KG) in low-dimensional space. In this article, we propose a lightweight end-to-end joint learning framework for fusing the tasks of KGE and RSs at the model level. Our method proposes a lightweight KG embedding method by using bidirectional bijection relation-type modeling to enable scalability for large graphs while using self-adaptive negative sampling to optimize negative sample generating. Our method further generates the integrated views for users and items based on relation-types to explicitly model users’ preferences and items’ features, respectively. Finally, we add virtual “recommendation” relations between the integrated views of users and items to model the preferences of users on items, seamlessly integrating RS with user-item KG over a unified graph. Experimental results on multiple datasets and benchmarks show that our method can achieve a better accuracy of recommendation compared with existing state-of-the-art methods. Complexity and runtime analysis suggests that our method can gain a lower time and space complexity than most of existing methods and improve scalability.","PeriodicalId":6934,"journal":{"name":"ACM Transactions on Information Systems (TOIS)","volume":"12 1","pages":"1 - 28"},"PeriodicalIF":0.0,"publicationDate":"2021-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75139229","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}
Xiangnan He, Z. Ren, Emine Yilmaz, Marc Najork, Tat-seng Chua
As a powerful data structure that represents the relationships among data objects, graph-structure data is ubiquitous in real-world applications. Learning on graph-structure data has become a hot spot in machine learning and data mining. Since most data in user-oriented services can be naturally organized as graphs, graph technologies have attracted increasing attention from IR community and achieved immense success, especially in two major research topics—user modeling and recommendation. In the recent decade, the IR and related communities have witnessed a number of major contributions to the field of graph learning. They include but not limited to collaborative filtering (e.g., He et al. [2020], Wang et al. [2019b], Wu et al. [2021], and Ying et al. [2018]), knowledge-aware recommendation (e.g., Cao et al. [2019] andWang et al. [2018, 2019a]), user profiling and demographic inference (e.g., Chen et al. [2019] and Rahimi et al. [2018]), social and sequential recommendation (e.g., Wang et al. [2020b] and Wu et al. [2019a, b]), bias and fairness (e.g., Rahman et al. [2019], Zhang et al. [2021a], and Zheng et al. [2021]). The growing body of work in this area has been supplemented by an increasing number of recent workshops (e.g., Cui et al. [2021], Ding et al. [2020], Jannach et al. [2020], and Yin et al. [2021]) and tutorials (e.g., Chen et al. [2020], Mehrotra et al. [2020], Tang and Dong [2019], Wang et al. [2020a], and Xu et al. [2018]). Despite such great
作为表示数据对象之间关系的强大数据结构，图结构数据在实际应用程序中无处不在。图结构数据的学习已成为机器学习和数据挖掘领域的研究热点。由于面向用户的服务中的大多数数据都可以自然地组织成图形，因此图形技术越来越受到IR社区的关注，并取得了巨大的成功，特别是在用户建模和推荐这两个主要研究课题上。在最近的十年中，IR和相关社区见证了对图学习领域的许多重大贡献。包括但不限于协同过滤(例如,他et al。[2020],王et al。(2019 b),吴et al。[2021],并应et al . [2018]), knowledge-aware建议(例如,曹et al。[2019]andWang et al .(2018, 2019)),用户分析和统计推断(例如,Chen等人[2019]和拉希米et al .[2018]),社会和顺序推荐王(例如,et al。(2020 b)和吴et al . (2019 a, b)),偏见和公平(例如,拉赫曼et al。[2019],Zhang et al。(2021),郑等[2021])。最近越来越多的研讨会(例如，Cui等人[2021]、Ding等人[2020]、Jannach等人[2020]和Yin等人[2021])和教程(例如，Chen等人[2020]、Mehrotra等人[2020]、Tang和Dong[2019]、Wang等人[2020a]和Xu等人[2018])补充了这一领域不断增长的工作。尽管如此伟大
{"title":"Introduction to the Special Section on Graph Technologies for User Modeling and Recommendation, Part 2","authors":"Xiangnan He, Z. Ren, Emine Yilmaz, Marc Najork, Tat-seng Chua","doi":"10.1145/3490180","DOIUrl":"https://doi.org/10.1145/3490180","url":null,"abstract":"As a powerful data structure that represents the relationships among data objects, graph-structure data is ubiquitous in real-world applications. Learning on graph-structure data has become a hot spot in machine learning and data mining. Since most data in user-oriented services can be naturally organized as graphs, graph technologies have attracted increasing attention from IR community and achieved immense success, especially in two major research topics—user modeling and recommendation. In the recent decade, the IR and related communities have witnessed a number of major contributions to the field of graph learning. They include but not limited to collaborative filtering (e.g., He et al. [2020], Wang et al. [2019b], Wu et al. [2021], and Ying et al. [2018]), knowledge-aware recommendation (e.g., Cao et al. [2019] andWang et al. [2018, 2019a]), user profiling and demographic inference (e.g., Chen et al. [2019] and Rahimi et al. [2018]), social and sequential recommendation (e.g., Wang et al. [2020b] and Wu et al. [2019a, b]), bias and fairness (e.g., Rahman et al. [2019], Zhang et al. [2021a], and Zheng et al. [2021]). The growing body of work in this area has been supplemented by an increasing number of recent workshops (e.g., Cui et al. [2021], Ding et al. [2020], Jannach et al. [2020], and Yin et al. [2021]) and tutorials (e.g., Chen et al. [2020], Mehrotra et al. [2020], Tang and Dong [2019], Wang et al. [2020a], and Xu et al. [2018]). Despite such great","PeriodicalId":6934,"journal":{"name":"ACM Transactions on Information Systems (TOIS)","volume":"266 1","pages":"1 - 5"},"PeriodicalIF":0.0,"publicationDate":"2021-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77493648","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}
As conversational search becomes more pervasive, it becomes increasingly important to understand the users’ underlying information needs when they converse with such systems in diverse domains. We conduct an in situ study to understand information needs arising in a home cooking context as well as how they are verbally communicated to an assistant. A human experimenter plays this role in our study. Based on the transcriptions of utterances, we derive a detailed hierarchical taxonomy of diverse information needs occurring in this context, which require different levels of assistance to be solved. The taxonomy shows that needs can be communicated through different linguistic means and require different amounts of context to be understood. In a second contribution, we perform classification experiments to determine the feasibility of predicting the type of information need a user has during a dialogue using the turn provided. For this multi-label classification problem, we achieve average F1 measures of 40% using BERT-based models. We demonstrate with examples which types of needs are difficult to predict and show why, concluding that models need to include more context information in order to improve both information need classification and assistance to make such systems usable.
{"title":"“What Can I Cook with these Ingredients?” - Understanding Cooking-Related Information Needs in Conversational Search","authors":"Alexander Frummet, David Elsweiler, Bernd Ludwig","doi":"10.1145/3498330","DOIUrl":"https://doi.org/10.1145/3498330","url":null,"abstract":"As conversational search becomes more pervasive, it becomes increasingly important to understand the users’ underlying information needs when they converse with such systems in diverse domains. We conduct an in situ study to understand information needs arising in a home cooking context as well as how they are verbally communicated to an assistant. A human experimenter plays this role in our study. Based on the transcriptions of utterances, we derive a detailed hierarchical taxonomy of diverse information needs occurring in this context, which require different levels of assistance to be solved. The taxonomy shows that needs can be communicated through different linguistic means and require different amounts of context to be understood. In a second contribution, we perform classification experiments to determine the feasibility of predicting the type of information need a user has during a dialogue using the turn provided. For this multi-label classification problem, we achieve average F1 measures of 40% using BERT-based models. We demonstrate with examples which types of needs are difficult to predict and show why, concluding that models need to include more context information in order to improve both information need classification and assistance to make such systems usable.","PeriodicalId":6934,"journal":{"name":"ACM Transactions on Information Systems (TOIS)","volume":"38 1","pages":"1 - 32"},"PeriodicalIF":0.0,"publicationDate":"2021-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73263535","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：