Jia Chen, Jiaxin Mao, Yiqun Liu, Fan Zhang, M. Zhang, Shaoping Ma
As queries submitted by users directly affect search experiences, how to organize queries has always been a research focus in Web search studies. While search request becomes complex and exploratory, many search sessions contain more than a single query thus reformulation becomes a necessity. To help users better formulate their queries in these complex search tasks, modern search engines usually provide a series of reformulation entries on search engine result pages (SERPs), i.e., query suggestions and related entities. However, few existing work have thoroughly studied why and how users perform query reformulations in these heterogeneous interfaces. Therefore, whether search engines provide sufficient assistance for users in reformulating queries remains under-investigated. To shed light on this research question, we conducted a field study to analyze fine-grained user reformulation behaviors including reformulation type, entry, reason, and the inspiration source with various search intents. Different from existing efforts that rely on external assessors to make judgments, in the field study we collect both implicit behavior signals and explicit user feedback information. Analysis results demonstrate that query reformulation behavior in Web search varies with the type of search tasks. We also found that the current query suggestion/related query recommendations provided by search engines do not offer enough help for users in complex search tasks. Based on the findings in our field study, we design a supervised learning framework to predict: 1) the reason behind each query reformulation, and 2) how users organize the reformulated query, both of which are novel challenges in this domain. This work provides insight into complex query reformulation behavior in Web search as well as the guidance for designing better query suggestion techniques in search engines.
{"title":"Towards a Better Understanding of Query Reformulation Behavior in Web Search","authors":"Jia Chen, Jiaxin Mao, Yiqun Liu, Fan Zhang, M. Zhang, Shaoping Ma","doi":"10.1145/3442381.3450127","DOIUrl":"https://doi.org/10.1145/3442381.3450127","url":null,"abstract":"As queries submitted by users directly affect search experiences, how to organize queries has always been a research focus in Web search studies. While search request becomes complex and exploratory, many search sessions contain more than a single query thus reformulation becomes a necessity. To help users better formulate their queries in these complex search tasks, modern search engines usually provide a series of reformulation entries on search engine result pages (SERPs), i.e., query suggestions and related entities. However, few existing work have thoroughly studied why and how users perform query reformulations in these heterogeneous interfaces. Therefore, whether search engines provide sufficient assistance for users in reformulating queries remains under-investigated. To shed light on this research question, we conducted a field study to analyze fine-grained user reformulation behaviors including reformulation type, entry, reason, and the inspiration source with various search intents. Different from existing efforts that rely on external assessors to make judgments, in the field study we collect both implicit behavior signals and explicit user feedback information. Analysis results demonstrate that query reformulation behavior in Web search varies with the type of search tasks. We also found that the current query suggestion/related query recommendations provided by search engines do not offer enough help for users in complex search tasks. Based on the findings in our field study, we design a supervised learning framework to predict: 1) the reason behind each query reformulation, and 2) how users organize the reformulated query, both of which are novel challenges in this domain. This work provides insight into complex query reformulation behavior in Web search as well as the guidance for designing better query suggestion techniques in search engines.","PeriodicalId":106672,"journal":{"name":"Proceedings of the Web Conference 2021","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128410844","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}
J. Piao, Guozheng Zhang, Fengli Xu, Zhilong Chen, Yong Li
Customer value is essential for successful customer relationship management. Although growing evidence suggests that customers’ purchase decisions can be influenced by social relationships, social influence is largely overlooked in previous research. In this work, we fill this gap with a novel framework — Motif-based Multi-view Graph Attention Networks with Gated Fusion (MAG), which jointly considers customer demographics, past behaviors, and social network structures. Specifically, (1) to make the best use of higher-order information in complex social networks, we design a motif-based multi-view graph attention module, which explicitly captures different higher-order structures, along with the attention mechanism auto-assigning high weights to informative ones. (2) To model the complex effects of customer attributes and social influence, we propose a gated fusion module with two gates: one depicts the susceptibility to social influence and the other depicts the dependency of the two factors. Extensive experiments on two large-scale datasets show superior performance of our model over the state-of-the-art baselines. Further, we discover that the increase of motifs does not guarantee better performances and identify how motifs play different roles. These findings shed light on how to understand socio-economic relationships among customers and find high-value customers.
{"title":"Predicting Customer Value with Social Relationships via Motif-based Graph Attention Networks","authors":"J. Piao, Guozheng Zhang, Fengli Xu, Zhilong Chen, Yong Li","doi":"10.1145/3442381.3449849","DOIUrl":"https://doi.org/10.1145/3442381.3449849","url":null,"abstract":"Customer value is essential for successful customer relationship management. Although growing evidence suggests that customers’ purchase decisions can be influenced by social relationships, social influence is largely overlooked in previous research. In this work, we fill this gap with a novel framework — Motif-based Multi-view Graph Attention Networks with Gated Fusion (MAG), which jointly considers customer demographics, past behaviors, and social network structures. Specifically, (1) to make the best use of higher-order information in complex social networks, we design a motif-based multi-view graph attention module, which explicitly captures different higher-order structures, along with the attention mechanism auto-assigning high weights to informative ones. (2) To model the complex effects of customer attributes and social influence, we propose a gated fusion module with two gates: one depicts the susceptibility to social influence and the other depicts the dependency of the two factors. Extensive experiments on two large-scale datasets show superior performance of our model over the state-of-the-art baselines. Further, we discover that the increase of motifs does not guarantee better performances and identify how motifs play different roles. These findings shed light on how to understand socio-economic relationships among customers and find high-value customers.","PeriodicalId":106672,"journal":{"name":"Proceedings of the Web Conference 2021","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127033676","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 Cross, Ahmed Mourad, G. Zuccon, B. Koopman
Increasingly, people go online to seek health advice. They commonly use the symptoms they are experiencing to identify the health conditions they may have (self-diagnosis task) as well as to determine an appropriate action to take (triaging task); e.g., should they seek emergent medical attention or attempt to treat themselves at home? This paper investigates the effectiveness of two of the most common methods people use for self-diagnosis and triaging: online symptom checkers and traditional web search engines. To this end, we conducted a user study with 64 real-world users performing 8 simulated self-diagnosis tasks. Participants were exposed to both a representative symptom checker and a search engine. The results of our study provides empirical evidence for whether using a search engine for health information improves people’s understanding of their health condition and their ability to act on them, compared to interacting with a symptom checker, which bases its interaction model on a question-answering process. Additionally, recorded answers to qualitative questionnaires from study participants provide insights into which style of interaction and system they prefer to use for obtaining medical information, and how helpful they thought each system was. These findings can help inform the development of better search engines and symptom checkers that support people seeking health advice online.
{"title":"Search Engines vs. Symptom Checkers: A Comparison of their Effectiveness for Online Health Advice","authors":"Sebastian Cross, Ahmed Mourad, G. Zuccon, B. Koopman","doi":"10.1145/3442381.3450140","DOIUrl":"https://doi.org/10.1145/3442381.3450140","url":null,"abstract":"Increasingly, people go online to seek health advice. They commonly use the symptoms they are experiencing to identify the health conditions they may have (self-diagnosis task) as well as to determine an appropriate action to take (triaging task); e.g., should they seek emergent medical attention or attempt to treat themselves at home? This paper investigates the effectiveness of two of the most common methods people use for self-diagnosis and triaging: online symptom checkers and traditional web search engines. To this end, we conducted a user study with 64 real-world users performing 8 simulated self-diagnosis tasks. Participants were exposed to both a representative symptom checker and a search engine. The results of our study provides empirical evidence for whether using a search engine for health information improves people’s understanding of their health condition and their ability to act on them, compared to interacting with a symptom checker, which bases its interaction model on a question-answering process. Additionally, recorded answers to qualitative questionnaires from study participants provide insights into which style of interaction and system they prefer to use for obtaining medical information, and how helpful they thought each system was. These findings can help inform the development of better search engines and symptom checkers that support people seeking health advice online.","PeriodicalId":106672,"journal":{"name":"Proceedings of the Web Conference 2021","volume":"81 6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123221444","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}
Existing research on cross-lingual retrieval cannot take good advantage of large-scale pretrained language models such as multilingual BERT and XLM. We hypothesize that the absence of cross-lingual passage-level relevance data for finetuning and the lack of query-document style pretraining are key factors of this issue. In this paper, we introduce two novel retrieval-oriented pretraining tasks to further pretrain cross-lingual language models for downstream retrieval tasks such as cross-lingual ad-hoc retrieval (CLIR) and cross-lingual question answering (CLQA). We construct distant supervision data from multilingual Wikipedia using section alignment to support retrieval-oriented language model pretraining. We also propose to directly finetune language models on part of the evaluation collection by making Transformers capable of accepting longer sequences. Experiments on multiple benchmark datasets show that our proposed model can significantly improve upon general multilingual language models in both the cross-lingual retrieval setting and the cross-lingual transfer setting.
{"title":"Cross-lingual Language Model Pretraining for Retrieval","authors":"Puxuan Yu, Hongliang Fei, P. Li","doi":"10.1145/3442381.3449830","DOIUrl":"https://doi.org/10.1145/3442381.3449830","url":null,"abstract":"Existing research on cross-lingual retrieval cannot take good advantage of large-scale pretrained language models such as multilingual BERT and XLM. We hypothesize that the absence of cross-lingual passage-level relevance data for finetuning and the lack of query-document style pretraining are key factors of this issue. In this paper, we introduce two novel retrieval-oriented pretraining tasks to further pretrain cross-lingual language models for downstream retrieval tasks such as cross-lingual ad-hoc retrieval (CLIR) and cross-lingual question answering (CLQA). We construct distant supervision data from multilingual Wikipedia using section alignment to support retrieval-oriented language model pretraining. We also propose to directly finetune language models on part of the evaluation collection by making Transformers capable of accepting longer sequences. Experiments on multiple benchmark datasets show that our proposed model can significantly improve upon general multilingual language models in both the cross-lingual retrieval setting and the cross-lingual transfer setting.","PeriodicalId":106672,"journal":{"name":"Proceedings of the Web Conference 2021","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123531503","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 Bowen, Zhenyu Zhang, Jiawei Sheng, Tingwen Liu, Yubin Wang, Yu-Chih Wang, Bin Wang
Open Information Extraction (OIE), the task aimed at discovering all textual facts organized in the form of (subject, predicate, object) found within a sentence, has gained much attention recently. However, in some knowledge-driven applications such as question answering, we often have a target entity and hope to obtain its structured factual knowledge for better understanding, instead of extracting all possible facts aimlessly from the corpus. In this paper, we define a new task, namely Semi-Open Information Extraction (SOIE), to address this need. The goal of SOIE is to discover domain-independent facts towards a particular entity from general and diverse web text. To facilitate research on this new task, we propose a large-scale human-annotated benchmark called SOIED, consisting of 61,984 facts for 8,013 subject entities annotated on 24,000 Chinese sentences collected from the web search engine. In addition, we propose a novel unified model called USE for this task. First, we introduce subject-guided sequence as input to a pre-trained language model and normalize the hidden representations conditioned on the subject embedding to encode the sentence in a subject-aware manner. Second, we decompose SOIE into three uncoupled subtasks: predicate extraction, object extraction, and boundary alignment. They can all be formulated as the problem of table filling by forming a two-dimensional tag table based on a task-specific tagging scheme. Third, we introduce a collaborative learning strategy that enables the interactive relations among subtasks to be better exploited by explicitly exchanging informative clues. Finally, we evaluate USE and several strong baselines on our new dataset. Experimental results demonstrate the advantages of the proposed method and reveal insight for future improvement.
{"title":"Semi-Open Information Extraction","authors":"Yu Bowen, Zhenyu Zhang, Jiawei Sheng, Tingwen Liu, Yubin Wang, Yu-Chih Wang, Bin Wang","doi":"10.1145/3442381.3450029","DOIUrl":"https://doi.org/10.1145/3442381.3450029","url":null,"abstract":"Open Information Extraction (OIE), the task aimed at discovering all textual facts organized in the form of (subject, predicate, object) found within a sentence, has gained much attention recently. However, in some knowledge-driven applications such as question answering, we often have a target entity and hope to obtain its structured factual knowledge for better understanding, instead of extracting all possible facts aimlessly from the corpus. In this paper, we define a new task, namely Semi-Open Information Extraction (SOIE), to address this need. The goal of SOIE is to discover domain-independent facts towards a particular entity from general and diverse web text. To facilitate research on this new task, we propose a large-scale human-annotated benchmark called SOIED, consisting of 61,984 facts for 8,013 subject entities annotated on 24,000 Chinese sentences collected from the web search engine. In addition, we propose a novel unified model called USE for this task. First, we introduce subject-guided sequence as input to a pre-trained language model and normalize the hidden representations conditioned on the subject embedding to encode the sentence in a subject-aware manner. Second, we decompose SOIE into three uncoupled subtasks: predicate extraction, object extraction, and boundary alignment. They can all be formulated as the problem of table filling by forming a two-dimensional tag table based on a task-specific tagging scheme. Third, we introduce a collaborative learning strategy that enables the interactive relations among subtasks to be better exploited by explicitly exchanging informative clues. Finally, we evaluate USE and several strong baselines on our new dataset. Experimental results demonstrate the advantages of the proposed method and reveal insight for future improvement.","PeriodicalId":106672,"journal":{"name":"Proceedings of the Web Conference 2021","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123680568","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}
Zachary Weinberg, Diogo Barradas, Nicolas Christin
The Great Firewall of China (GFW) prevents Chinese citizens from accessing online content deemed objectionable by the Chinese government. One way it does this is to search for forbidden keywords in unencrypted packet streams. When it detects them, it terminates the offending stream by injecting TCP RST packets, and blocks further traffic between the same two hosts for a few minutes. We report on a detailed investigation of the GFW’s application-layer understanding of HTTP. Forbidden keywords are only detected in certain locations within an HTTP request. Requests that contain the English word “search” are inspected for a longer list of forbidden keywords than requests without this word. The firewall can be evaded by bending the rules of the HTTP specification. We observe censorship based on the cleartext TLS Server Name Indication (SNI), but we find no evidence for bulk decryption of HTTPS. We also report on changes since 2014 in the contents of the forbidden keyword list. Over 85% of the forbidden keywords have been replaced since 2014, with the surviving terms referring to perennially sensitive topics. The new keywords refer to recent events and controversies. The GFW’s keyword list is not kept in sync with the blocklists used by Chinese chat clients.
{"title":"Chinese Wall or Swiss Cheese? Keyword filtering in the Great Firewall of China","authors":"Zachary Weinberg, Diogo Barradas, Nicolas Christin","doi":"10.1145/3442381.3450076","DOIUrl":"https://doi.org/10.1145/3442381.3450076","url":null,"abstract":"The Great Firewall of China (GFW) prevents Chinese citizens from accessing online content deemed objectionable by the Chinese government. One way it does this is to search for forbidden keywords in unencrypted packet streams. When it detects them, it terminates the offending stream by injecting TCP RST packets, and blocks further traffic between the same two hosts for a few minutes. We report on a detailed investigation of the GFW’s application-layer understanding of HTTP. Forbidden keywords are only detected in certain locations within an HTTP request. Requests that contain the English word “search” are inspected for a longer list of forbidden keywords than requests without this word. The firewall can be evaded by bending the rules of the HTTP specification. We observe censorship based on the cleartext TLS Server Name Indication (SNI), but we find no evidence for bulk decryption of HTTPS. We also report on changes since 2014 in the contents of the forbidden keyword list. Over 85% of the forbidden keywords have been replaced since 2014, with the surviving terms referring to perennially sensitive topics. The new keywords refer to recent events and controversies. The GFW’s keyword list is not kept in sync with the blocklists used by Chinese chat clients.","PeriodicalId":106672,"journal":{"name":"Proceedings of the Web Conference 2021","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121312623","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}
Matthew Johnson, Jenny T Liang, Michelle Lin, S. Singanamalla, Kurtis Heimerl
While only generating a minuscule percentage of global traffic, largely lost in the noise of large-scale analyses, remote rural networks are the physical frontier of the Internet today. Through tight integration with a local operator’s infrastructure, we gather a unique dataset to characterize and report a year of interaction between finances, utilization, and performance of a novel, remote, data-only Community LTE Network in Bokondini, Indonesia. With visibility to drill down to individual users, we find use highly unbalanced and the network supported by only a handful of relatively heavy consumers. 45% of users are offline more days than online, and the median user consumes only 77 MB per day online and 36 MB per day on average, limiting consumption by frequently “topping up” in small amounts. Outside video and social media, messaging and IP calling provided by over-the-top services like Facebook Messenger, QQ, and WhatsApp comprise a relatively large percentage of traffic consistently across both heavy and light users. Our analysis shows that Internet-only Community Cellular Networks can be profitable despite most users spending less than $1 USD/day, and offers insights into the unique properties of these networks.
{"title":"Whale Watching in Inland Indonesia: Analyzing a Small, Remote, Internet-Based Community Cellular Network","authors":"Matthew Johnson, Jenny T Liang, Michelle Lin, S. Singanamalla, Kurtis Heimerl","doi":"10.1145/3442381.3449996","DOIUrl":"https://doi.org/10.1145/3442381.3449996","url":null,"abstract":"While only generating a minuscule percentage of global traffic, largely lost in the noise of large-scale analyses, remote rural networks are the physical frontier of the Internet today. Through tight integration with a local operator’s infrastructure, we gather a unique dataset to characterize and report a year of interaction between finances, utilization, and performance of a novel, remote, data-only Community LTE Network in Bokondini, Indonesia. With visibility to drill down to individual users, we find use highly unbalanced and the network supported by only a handful of relatively heavy consumers. 45% of users are offline more days than online, and the median user consumes only 77 MB per day online and 36 MB per day on average, limiting consumption by frequently “topping up” in small amounts. Outside video and social media, messaging and IP calling provided by over-the-top services like Facebook Messenger, QQ, and WhatsApp comprise a relatively large percentage of traffic consistently across both heavy and light users. Our analysis shows that Internet-only Community Cellular Networks can be profitable despite most users spending less than $1 USD/day, and offers insights into the unique properties of these networks.","PeriodicalId":106672,"journal":{"name":"Proceedings of the Web Conference 2021","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121478234","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 initiate the study of information elicitation mechanisms for a crowd containing both self-interested agents, who respond to incentives, and adversarial agents, who may collude to disrupt the system. Our mechanisms work in the peer prediction setting where ground truth need not be accessible to the mechanism or even exist. We provide a meta-mechanism that reduces the design of peer prediction mechanisms to a related robust learning problem. The resulting mechanisms are ϵ-informed truthful, which means truth-telling is the highest paid ϵ-Bayesian Nash equilibrium (up to ϵ-error) and pays strictly more than uninformative equilibria. The value of ϵ depends on the properties of robust learning algorithm, and typically limits to 0 as the number of tasks and agents increase. We show how to use our meta-mechanism to design mechanisms with provable guarantees in two important crowdsourcing settings even when some agents are self-interested and others are adversarial.
{"title":"Information Elicitation from Rowdy Crowds","authors":"G. Schoenebeck, Fang-Yi Yu, Yichi Zhang","doi":"10.1145/3442381.3449840","DOIUrl":"https://doi.org/10.1145/3442381.3449840","url":null,"abstract":"We initiate the study of information elicitation mechanisms for a crowd containing both self-interested agents, who respond to incentives, and adversarial agents, who may collude to disrupt the system. Our mechanisms work in the peer prediction setting where ground truth need not be accessible to the mechanism or even exist. We provide a meta-mechanism that reduces the design of peer prediction mechanisms to a related robust learning problem. The resulting mechanisms are ϵ-informed truthful, which means truth-telling is the highest paid ϵ-Bayesian Nash equilibrium (up to ϵ-error) and pays strictly more than uninformative equilibria. The value of ϵ depends on the properties of robust learning algorithm, and typically limits to 0 as the number of tasks and agents increase. We show how to use our meta-mechanism to design mechanisms with provable guarantees in two important crowdsourcing settings even when some agents are self-interested and others are adversarial.","PeriodicalId":106672,"journal":{"name":"Proceedings of the Web Conference 2021","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121512494","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 Liu, Xiang Ao, Zidi Qin, Jianfeng Chi, Jinghua Feng, Hao Yang, Qing He
Graph-based fraud detection approaches have escalated lots of attention recently due to the abundant relational information of graph-structured data, which may be beneficial for the detection of fraudsters. However, the GNN-based algorithms could fare poorly when the label distribution of nodes is heavily skewed, and it is common in sensitive areas such as financial fraud, etc. To remedy the class imbalance problem of graph-based fraud detection, we propose a Pick and Choose Graph Neural Network (PC-GNN for short) for imbalanced supervised learning on graphs. First, nodes and edges are picked with a devised label-balanced sampler to construct sub-graphs for mini-batch training. Next, for each node in the sub-graph, the neighbor candidates are chosen by a proposed neighborhood sampler. Finally, information from the selected neighbors and different relations are aggregated to obtain the final representation of a target node. Experiments on both benchmark and real-world graph-based fraud detection tasks demonstrate that PC-GNN apparently outperforms state-of-the-art baselines.
基于图的欺诈检测方法近年来受到越来越多的关注，因为图结构数据中含有丰富的关系信息，这可能有利于欺诈者的检测。然而，当节点的标签分布严重偏斜时，基于gnn的算法可能会表现不佳，并且在金融欺诈等敏感领域很常见。为了解决基于图的欺诈检测中的类不平衡问题，我们提出了一种用于图上不平衡监督学习的Pick and Choose图神经网络(PC-GNN)。首先，使用设计的标签平衡采样器选择节点和边，构建用于小批量训练的子图。接下来，对于子图中的每个节点，由提议的邻域采样器选择邻居候选节点。最后，聚合来自所选邻居和不同关系的信息，以获得目标节点的最终表示。在基准测试和现实世界基于图形的欺诈检测任务上的实验表明，PC-GNN明显优于最先进的基线。
{"title":"Pick and Choose: A GNN-based Imbalanced Learning Approach for Fraud Detection","authors":"Yang Liu, Xiang Ao, Zidi Qin, Jianfeng Chi, Jinghua Feng, Hao Yang, Qing He","doi":"10.1145/3442381.3449989","DOIUrl":"https://doi.org/10.1145/3442381.3449989","url":null,"abstract":"Graph-based fraud detection approaches have escalated lots of attention recently due to the abundant relational information of graph-structured data, which may be beneficial for the detection of fraudsters. However, the GNN-based algorithms could fare poorly when the label distribution of nodes is heavily skewed, and it is common in sensitive areas such as financial fraud, etc. To remedy the class imbalance problem of graph-based fraud detection, we propose a Pick and Choose Graph Neural Network (PC-GNN for short) for imbalanced supervised learning on graphs. First, nodes and edges are picked with a devised label-balanced sampler to construct sub-graphs for mini-batch training. Next, for each node in the sub-graph, the neighbor candidates are chosen by a proposed neighborhood sampler. Finally, information from the selected neighbors and different relations are aggregated to obtain the final representation of a target node. Experiments on both benchmark and real-world graph-based fraud detection tasks demonstrate that PC-GNN apparently outperforms state-of-the-art baselines.","PeriodicalId":106672,"journal":{"name":"Proceedings of the Web Conference 2021","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132110565","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}
Networks are ubiquitous in the real world. Link prediction, as one of the key problems for network-structured data, aims to predict whether there exists a link between two nodes. The traditional approaches are based on the explicit similarity computation between the compact node representation by embedding each node into a low-dimensional space. In order to efficiently handle the intensive similarity computation in link prediction, the hashing technique has been successfully used to produce the node representation in the Hamming space. However, the hashing-based link prediction algorithms face accuracy loss from the randomized hashing techniques or inefficiency from the learning to hash techniques in the embedding process. Currently, the Graph Neural Network (GNN) framework has been widely applied to the graph-related tasks in an end-to-end manner, but it commonly requires substantial computational resources and memory costs due to massive parameter learning, which makes the GNN-based algorithms impractical without the help of a powerful workhorse. In this paper, we propose a simple and effective model called #GNN, which balances the trade-off between accuracy and efficiency. #GNN is able to efficiently acquire node representation in the Hamming space for link prediction by exploiting the randomized hashing technique to implement message passing and capture high-order proximity in the GNN framework. Furthermore, we characterize the discriminative power of #GNN in probability. The extensive experimental results demonstrate that the proposed #GNN algorithm achieves accuracy comparable to the learning-based algorithms and outperforms the randomized algorithm, while running significantly faster than the learning-based algorithms. Also, the proposed algorithm shows excellent scalability on a large-scale network with the limited resources.
{"title":"Hashing-Accelerated Graph Neural Networks for Link Prediction","authors":"Wei Wu, Bin Li, Chuan Luo, W. Nejdl","doi":"10.1145/3442381.3449884","DOIUrl":"https://doi.org/10.1145/3442381.3449884","url":null,"abstract":"Networks are ubiquitous in the real world. Link prediction, as one of the key problems for network-structured data, aims to predict whether there exists a link between two nodes. The traditional approaches are based on the explicit similarity computation between the compact node representation by embedding each node into a low-dimensional space. In order to efficiently handle the intensive similarity computation in link prediction, the hashing technique has been successfully used to produce the node representation in the Hamming space. However, the hashing-based link prediction algorithms face accuracy loss from the randomized hashing techniques or inefficiency from the learning to hash techniques in the embedding process. Currently, the Graph Neural Network (GNN) framework has been widely applied to the graph-related tasks in an end-to-end manner, but it commonly requires substantial computational resources and memory costs due to massive parameter learning, which makes the GNN-based algorithms impractical without the help of a powerful workhorse. In this paper, we propose a simple and effective model called #GNN, which balances the trade-off between accuracy and efficiency. #GNN is able to efficiently acquire node representation in the Hamming space for link prediction by exploiting the randomized hashing technique to implement message passing and capture high-order proximity in the GNN framework. Furthermore, we characterize the discriminative power of #GNN in probability. The extensive experimental results demonstrate that the proposed #GNN algorithm achieves accuracy comparable to the learning-based algorithms and outperforms the randomized algorithm, while running significantly faster than the learning-based algorithms. Also, the proposed algorithm shows excellent scalability on a large-scale network with the limited resources.","PeriodicalId":106672,"journal":{"name":"Proceedings of the Web Conference 2021","volume":"148 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133029945","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：