Jianling Wang, Raphael Louca, D. Hu, Caitlin Cellier, James Caverlee, Liangjie Hong
Currently, most sequence-based recommendation models aim to predict a user's next actions (e.g. next purchase) based on their past actions. These models either capture users' intrinsic preference (e.g. a comedy lover, or a fan of fantasy) from their long-term behavior patterns or infer their current needs by emphasizing recent actions. However, in e-commerce, intrinsic user behavior may be shifted by occasions such as birthdays, anniversaries, or gifting celebrations (Valentine's Day or Mother's Day), leading to purchases that deviate from long-term preferences and are not related to recent actions. In this work, we propose a novel next-item recommendation system which models a user's default, intrinsic preference, as well as two different kinds of occasion-based signals that may cause users to deviate from their normal behavior. More specifically, this model is novel in that it: (1) captures a personal occasion signal using an attention layer that models reoccurring occasions specific to that user (e.g. a birthday); (2) captures a global occasion signal using an attention layer that models seasonal or reoccurring occasions for many users (e.g. Christmas); (3) balances the user's intrinsic preferences with the personal and global occasion signals for different users at different timestamps with a gating layer. We explore two real-world e-commerce datasets (Amazon and Etsy) and show that the proposed model outperforms state-of-the-art models by 7.62% and 6.06% in predicting users' next purchase.
{"title":"Time to Shop for Valentine's Day: Shopping Occasions and Sequential Recommendation in E-commerce","authors":"Jianling Wang, Raphael Louca, D. Hu, Caitlin Cellier, James Caverlee, Liangjie Hong","doi":"10.1145/3336191.3371836","DOIUrl":"https://doi.org/10.1145/3336191.3371836","url":null,"abstract":"Currently, most sequence-based recommendation models aim to predict a user's next actions (e.g. next purchase) based on their past actions. These models either capture users' intrinsic preference (e.g. a comedy lover, or a fan of fantasy) from their long-term behavior patterns or infer their current needs by emphasizing recent actions. However, in e-commerce, intrinsic user behavior may be shifted by occasions such as birthdays, anniversaries, or gifting celebrations (Valentine's Day or Mother's Day), leading to purchases that deviate from long-term preferences and are not related to recent actions. In this work, we propose a novel next-item recommendation system which models a user's default, intrinsic preference, as well as two different kinds of occasion-based signals that may cause users to deviate from their normal behavior. More specifically, this model is novel in that it: (1) captures a personal occasion signal using an attention layer that models reoccurring occasions specific to that user (e.g. a birthday); (2) captures a global occasion signal using an attention layer that models seasonal or reoccurring occasions for many users (e.g. Christmas); (3) balances the user's intrinsic preferences with the personal and global occasion signals for different users at different timestamps with a gating layer. We explore two real-world e-commerce datasets (Amazon and Etsy) and show that the proposed model outperforms state-of-the-art models by 7.62% and 6.06% in predicting users' next purchase.","PeriodicalId":319008,"journal":{"name":"Proceedings of the 13th International Conference on Web Search and Data Mining","volume":"4 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":"129014895","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 retrieval models have gained much success, however, we have to admit that the models work in a rather different manner than how humans make relevance judgments. To bridge the gap between practical user behavior and retrieval model, it is essential to construct cognitive-oriented retrieval model. The cognitive process in IR includes two important aspects: searching and reading. Searching is the user behaviors interacted with retrieval system such as query formulation and click while reading is the information seeking behavior in a specific landing page or document. We plan to better understand user cognitive behaviors in these two aspects by conducting a lab-based user study. More importantly, the heuristics drawn from cognitive behaviors are then incorporated into retrieval models.
{"title":"Think like a Human: Constructing Cognitive-oriented Retrieval Model for Web Search","authors":"Xiangsheng Li","doi":"10.1145/3336191.3372180","DOIUrl":"https://doi.org/10.1145/3336191.3372180","url":null,"abstract":"Existing retrieval models have gained much success, however, we have to admit that the models work in a rather different manner than how humans make relevance judgments. To bridge the gap between practical user behavior and retrieval model, it is essential to construct cognitive-oriented retrieval model. The cognitive process in IR includes two important aspects: searching and reading. Searching is the user behaviors interacted with retrieval system such as query formulation and click while reading is the information seeking behavior in a specific landing page or document. We plan to better understand user cognitive behaviors in these two aspects by conducting a lab-based user study. More importantly, the heuristics drawn from cognitive behaviors are then incorporated into retrieval models.","PeriodicalId":319008,"journal":{"name":"Proceedings of the 13th International Conference on Web Search and Data Mining","volume":"16 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":"127920337","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}
With the growing popularity of neural approaches for ad-hoc ranking, there is a need for tools that can effectively reproduce prior results and ease continued research by supporting current state-of-the-art approaches. Although several excellent neural ranking tools exist, none offer an easy end-to-end ad-hoc neural raking pipeline. A complete pipeline is particularly important for ad-hoc ranking because there are numerous parameter settings that have a considerable effect on the ultimate performance yet often are under-reported in current work (e.g., initial ranking settings, re-ranking threshold, training sampling strategy, etc.). In this work, I present a complete ad-hoc neural ranking pipeline which addresses these shortcomings: OpenNIR. The pipeline is easy to use (a single command will download required data, train, and evaluate a model), yet highly configurable, allowing for continued work in areas that are understudied. Aside from the core pipeline, the software also includes several bells and whistles that make use of components of the pipeline, such as performance benchmarking and tuning of unsupervised ranker parameters for fair comparisons against traditional baselines. The pipeline and these capabilities are demonstrated. The code is available, and contributions are welcome.
{"title":"OpenNIR: A Complete Neural Ad-Hoc Ranking Pipeline","authors":"Sean MacAvaney","doi":"10.1145/3336191.3371864","DOIUrl":"https://doi.org/10.1145/3336191.3371864","url":null,"abstract":"With the growing popularity of neural approaches for ad-hoc ranking, there is a need for tools that can effectively reproduce prior results and ease continued research by supporting current state-of-the-art approaches. Although several excellent neural ranking tools exist, none offer an easy end-to-end ad-hoc neural raking pipeline. A complete pipeline is particularly important for ad-hoc ranking because there are numerous parameter settings that have a considerable effect on the ultimate performance yet often are under-reported in current work (e.g., initial ranking settings, re-ranking threshold, training sampling strategy, etc.). In this work, I present a complete ad-hoc neural ranking pipeline which addresses these shortcomings: OpenNIR. The pipeline is easy to use (a single command will download required data, train, and evaluate a model), yet highly configurable, allowing for continued work in areas that are understudied. Aside from the core pipeline, the software also includes several bells and whistles that make use of components of the pipeline, such as performance benchmarking and tuning of unsupervised ranker parameters for fair comparisons against traditional baselines. The pipeline and these capabilities are demonstrated. The code is available, and contributions are welcome.","PeriodicalId":319008,"journal":{"name":"Proceedings of the 13th International Conference on Web Search and Data Mining","volume":"72 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":"126528571","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}
Somit Gupta, Xiaolin Shi, Pavel A. Dmitriev, Xin Fu, Avijit Mukherjee
A/B Testing is the gold standard to estimate the causal relationship between a change in a product and its impact on key outcome measures. It is widely used in the industry to test changes ranging from simple copy change or UI change to more complex changes like using machine learning models to personalize user experience. The key aspect of A/B testing is evaluation of experiment results. Designing the right set of metrics - correct outcome measures, data quality indicators, guardrails that prevent harm to business, and a comprehensive set of supporting metrics to understand the "why" behind the key movements is the #1 challenge practitioners face when trying to scale their experimentation program [11, 14]. On the technical side, improving sensitivity of experiment metrics is a hard problem and an active research area, with large practical implications as more and more small and medium size businesses are trying to adopt A/B testing and suffer from insufficient power. In this tutorial we will discuss challenges, best practices, and pitfalls in evaluating experiment results, focusing on both lessons learned and practical guidelines as well as open research questions. A version of this tutorial was also present at KDD 2019 [23]. It was attended by around 150 participants.
{"title":"Challenges, Best Practices and Pitfalls in Evaluating Results of Online Controlled Experiments","authors":"Somit Gupta, Xiaolin Shi, Pavel A. Dmitriev, Xin Fu, Avijit Mukherjee","doi":"10.1145/3336191.3371871","DOIUrl":"https://doi.org/10.1145/3336191.3371871","url":null,"abstract":"A/B Testing is the gold standard to estimate the causal relationship between a change in a product and its impact on key outcome measures. It is widely used in the industry to test changes ranging from simple copy change or UI change to more complex changes like using machine learning models to personalize user experience. The key aspect of A/B testing is evaluation of experiment results. Designing the right set of metrics - correct outcome measures, data quality indicators, guardrails that prevent harm to business, and a comprehensive set of supporting metrics to understand the \"why\" behind the key movements is the #1 challenge practitioners face when trying to scale their experimentation program [11, 14]. On the technical side, improving sensitivity of experiment metrics is a hard problem and an active research area, with large practical implications as more and more small and medium size businesses are trying to adopt A/B testing and suffer from insufficient power. In this tutorial we will discuss challenges, best practices, and pitfalls in evaluating experiment results, focusing on both lessons learned and practical guidelines as well as open research questions. A version of this tutorial was also present at KDD 2019 [23]. It was attended by around 150 participants.","PeriodicalId":319008,"journal":{"name":"Proceedings of the 13th International Conference on Web Search and Data Mining","volume":"110 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":"134433527","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}
Twitter is currently a popular microblogging platform for spread of information by users in the form of tweet messages. Such tweets are shared with followers of the seed user who may reshare it with their own set of followers. Long chain of such retweets form cascades. For effective diffusion of information through such Twitter cascades, we identify two different objectives based on using temporal sequence of retweets. Firstly, we aim to infer the structure of influence trees of Twitter cascades, denoting the who-influenced-whom relationship among retweeting users in the cascade, that can play a significant role in identifying critical paths in the network for information dissemination. The constructed trees closely resemble ground truth influence trees of empirical cascades with high retweet count. Secondly, we propose a fast and efficient algorithm for detection of influential users by identifying anchor nodes from temporal retweet sequence. Identification of such a diverse set of influential users enable a faster diffusion of tweets to a large and diverse population, when targeted as seeds thereby maximizing the influence spread, facilitating several applications including viral marketing, disease control and news dissemination.
{"title":"Temporal Pattern of Retweet(s) Help to Maximize Information Diffusion in Twitter","authors":"Ayan Kumar Bhowmick","doi":"10.1145/3336191.3372181","DOIUrl":"https://doi.org/10.1145/3336191.3372181","url":null,"abstract":"Twitter is currently a popular microblogging platform for spread of information by users in the form of tweet messages. Such tweets are shared with followers of the seed user who may reshare it with their own set of followers. Long chain of such retweets form cascades. For effective diffusion of information through such Twitter cascades, we identify two different objectives based on using temporal sequence of retweets. Firstly, we aim to infer the structure of influence trees of Twitter cascades, denoting the who-influenced-whom relationship among retweeting users in the cascade, that can play a significant role in identifying critical paths in the network for information dissemination. The constructed trees closely resemble ground truth influence trees of empirical cascades with high retweet count. Secondly, we propose a fast and efficient algorithm for detection of influential users by identifying anchor nodes from temporal retweet sequence. Identification of such a diverse set of influential users enable a faster diffusion of tweets to a large and diverse population, when targeted as seeds thereby maximizing the influence spread, facilitating several applications including viral marketing, disease control and news dissemination.","PeriodicalId":319008,"journal":{"name":"Proceedings of the 13th International Conference on Web Search and Data Mining","volume":"18 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":"114685736","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}
User satisfaction is an important factor when evaluating search systems, and hence a good metric should give rise to scores that have a strong positive correlation with user satisfaction ratings. A metric should also correspond to a plausible user model, and hence provide a tangible manifestation of how users interact with search rankings. Recent work has focused on metrics whose user models accurately portray the behavior of search engine users. Here we investigate whether those same metrics then also correlate with user satisfaction. We carry out experiments using various classes of metrics, and confirm through the lens of the C/W/L framework that the metrics with user models that reflect typical behavior also tend to be the metrics that correlate well with user satisfaction ratings.
{"title":"Metrics, User Models, and Satisfaction","authors":"A. Wicaksono, Alistair Moffat","doi":"10.1145/3336191.3371799","DOIUrl":"https://doi.org/10.1145/3336191.3371799","url":null,"abstract":"User satisfaction is an important factor when evaluating search systems, and hence a good metric should give rise to scores that have a strong positive correlation with user satisfaction ratings. A metric should also correspond to a plausible user model, and hence provide a tangible manifestation of how users interact with search rankings. Recent work has focused on metrics whose user models accurately portray the behavior of search engine users. Here we investigate whether those same metrics then also correlate with user satisfaction. We carry out experiments using various classes of metrics, and confirm through the lens of the C/W/L framework that the metrics with user models that reflect typical behavior also tend to be the metrics that correlate well with user satisfaction ratings.","PeriodicalId":319008,"journal":{"name":"Proceedings of the 13th International Conference on Web Search and Data Mining","volume":"31 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":"114793102","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The goal of personalized search is to tailor the document ranking list to meet user's individual needs. Previous studies showed users usually look for the information that has been searched before. This is called re-finding behavior which is widely explored in existing personalized search approaches. However, most existing methods for identifying re-finding behavior focus on simple lexical similarities between queries. In this paper, we propose to construct memory networks (MN) to support the identification of more complex re-finding behavior. Specifically, incorporating semantic information, we devise two external memories to make an expansion of re-finding based on the query and the document respectively. We further design an intent memory to recognize session-based re-finding behavior. Endowed with these memory networks, we can build a fine-grained user model dynamically based on the current query and documents, and use the model to re-rank the results. Experimental results show the significant improvement of our model compared with traditional methods.
{"title":"Enhancing Re-finding Behavior with External Memories for Personalized Search","authors":"Yujia Zhou, Zhicheng Dou, Ji-rong Wen","doi":"10.1145/3336191.3371794","DOIUrl":"https://doi.org/10.1145/3336191.3371794","url":null,"abstract":"The goal of personalized search is to tailor the document ranking list to meet user's individual needs. Previous studies showed users usually look for the information that has been searched before. This is called re-finding behavior which is widely explored in existing personalized search approaches. However, most existing methods for identifying re-finding behavior focus on simple lexical similarities between queries. In this paper, we propose to construct memory networks (MN) to support the identification of more complex re-finding behavior. Specifically, incorporating semantic information, we devise two external memories to make an expansion of re-finding based on the query and the document respectively. We further design an intent memory to recognize session-based re-finding behavior. Endowed with these memory networks, we can build a fine-grained user model dynamically based on the current query and documents, and use the model to re-rank the results. Experimental results show the significant improvement of our model compared with traditional methods.","PeriodicalId":319008,"journal":{"name":"Proceedings of the 13th International Conference on Web Search and Data Mining","volume":"15 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":"114497326","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}
Lack of comprehensive information on frequently asked questions (FAQ) web pages forces users to pose their questions on community question answering forums or contact businesses over slow media like emails or phone calls. This in turn often results into sub-optimal user experience and opportunity loss for businesses. While previous work focuses on FAQ mining and answering queries from FAQ pages, there is no work on verifying completeness or augmenting FAQ pages. We present a system, called FAQAugmenter, which given an FAQ web page, (1) harnesses signals from query logs and the web corpus to identify missing topics, and (2) suggests ranked list of questions for FAQ web page augmentation. Our experiments with FAQ pages from five enterprises each across three categories (banks, hospitals and airports) show that FAQAugmenter suggests high quality relevant questions. FAQAugmenter will contribute significantly not just in improving quality of FAQ web pages but also in turn improving quality of downstream applications like Microsoft QnA Maker.
{"title":"FAQAugmenter: Suggesting Questions for Enterprise FAQ Pages","authors":"Ankush Chatterjee, Manish Gupta, Puneet Agrawal","doi":"10.1145/3336191.3371862","DOIUrl":"https://doi.org/10.1145/3336191.3371862","url":null,"abstract":"Lack of comprehensive information on frequently asked questions (FAQ) web pages forces users to pose their questions on community question answering forums or contact businesses over slow media like emails or phone calls. This in turn often results into sub-optimal user experience and opportunity loss for businesses. While previous work focuses on FAQ mining and answering queries from FAQ pages, there is no work on verifying completeness or augmenting FAQ pages. We present a system, called FAQAugmenter, which given an FAQ web page, (1) harnesses signals from query logs and the web corpus to identify missing topics, and (2) suggests ranked list of questions for FAQ web page augmentation. Our experiments with FAQ pages from five enterprises each across three categories (banks, hospitals and airports) show that FAQAugmenter suggests high quality relevant questions. FAQAugmenter will contribute significantly not just in improving quality of FAQ web pages but also in turn improving quality of downstream applications like Microsoft QnA Maker.","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":"115737710","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}
Recommender systems are embracing conversational technologies to obtain user preferences dynamically, and to overcome inherent limitations of their static models. A successful Conversational Recommender System (CRS) requires proper handling of interactions between conversation and recommendation. We argue that three fundamental problems need to be solved: 1) what questions to ask regarding item attributes, 2) when to recommend items, and 3) how to adapt to the users' online feedback. To the best of our knowledge, there lacks a unified framework that addresses these problems. In this work, we fill this missing interaction framework gap by proposing a new CRS framework named Estimation"Action" Reflection, or EAR, which consists of three stages to better converse with users. (1) Estimation, which builds predictive models to estimate user preference on both items and item attributes; (2) Action, which learns a dialogue policy to determine whether to ask attributes or recommend items, based on Estimation stage and conversation history; and (3) Reflection, which updates the recommender model when a user rejects the recommendations made by the Action stage. We present two conversation scenarios on binary and enumerated questions, and conduct extensive experiments on two datasets from Yelp and LastFM, for each scenario, respectively. Our experiments demonstrate significant improvements over the state-of-the-art method CRM [32], corresponding to fewer conversation turns and a higher level of recommendation hits.
{"title":"Estimation-Action-Reflection: Towards Deep Interaction Between Conversational and Recommender Systems","authors":"Wenqiang Lei, Xiangnan He, Yisong Miao, Qingyun Wu, Richang Hong, Min-Yen Kan, Tat-Seng Chua","doi":"10.1145/3336191.3371769","DOIUrl":"https://doi.org/10.1145/3336191.3371769","url":null,"abstract":"Recommender systems are embracing conversational technologies to obtain user preferences dynamically, and to overcome inherent limitations of their static models. A successful Conversational Recommender System (CRS) requires proper handling of interactions between conversation and recommendation. We argue that three fundamental problems need to be solved: 1) what questions to ask regarding item attributes, 2) when to recommend items, and 3) how to adapt to the users' online feedback. To the best of our knowledge, there lacks a unified framework that addresses these problems. In this work, we fill this missing interaction framework gap by proposing a new CRS framework named Estimation\"Action\" Reflection, or EAR, which consists of three stages to better converse with users. (1) Estimation, which builds predictive models to estimate user preference on both items and item attributes; (2) Action, which learns a dialogue policy to determine whether to ask attributes or recommend items, based on Estimation stage and conversation history; and (3) Reflection, which updates the recommender model when a user rejects the recommendations made by the Action stage. We present two conversation scenarios on binary and enumerated questions, and conduct extensive experiments on two datasets from Yelp and LastFM, for each scenario, respectively. Our experiments demonstrate significant improvements over the state-of-the-art method CRM [32], corresponding to fewer conversation turns and a higher level of recommendation hits.","PeriodicalId":319008,"journal":{"name":"Proceedings of the 13th International Conference on Web Search and Data Mining","volume":"17 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":"123673216","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}
Fiction and fantasy are archetypes of long-tail domains that lack comprehensive methods for automated language processing and knowledge extraction. We present ENTYFI, the first methodology for typing entities in fictional texts coming from books, fan communities or amateur writers. ENTYFI builds on 205 automatically induced high-quality type systems for popular fictional domains, and exploits the overlap and reuse of these fictional domains for fine-grained typing in previously unseen texts. ENTYFI comprises five steps: type system induction, domain relatedness ranking, mention detection, mention typing, and type consolidation. The recall-oriented typing module combines a supervised neural model, unsupervised Hearst-style and dependency patterns, and knowledge base lookups. The precision-oriented consolidation stage utilizes co-occurrence statistics in order to remove noise and to identify the most relevant types. Extensive experiments on newly seen fictional texts demonstrate the quality of ENTYFI.
{"title":"ENTYFI","authors":"C. Chu, Simon Razniewski, G. Weikum","doi":"10.1145/3336191.3371808","DOIUrl":"https://doi.org/10.1145/3336191.3371808","url":null,"abstract":"Fiction and fantasy are archetypes of long-tail domains that lack comprehensive methods for automated language processing and knowledge extraction. We present ENTYFI, the first methodology for typing entities in fictional texts coming from books, fan communities or amateur writers. ENTYFI builds on 205 automatically induced high-quality type systems for popular fictional domains, and exploits the overlap and reuse of these fictional domains for fine-grained typing in previously unseen texts. ENTYFI comprises five steps: type system induction, domain relatedness ranking, mention detection, mention typing, and type consolidation. The recall-oriented typing module combines a supervised neural model, unsupervised Hearst-style and dependency patterns, and knowledge base lookups. The precision-oriented consolidation stage utilizes co-occurrence statistics in order to remove noise and to identify the most relevant types. Extensive experiments on newly seen fictional texts demonstrate the quality of ENTYFI.","PeriodicalId":319008,"journal":{"name":"Proceedings of the 13th International Conference on Web Search and Data Mining","volume":"18 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":"125585541","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：