A fundamental role of news websites is to recommend articles that are interesting to read. The key challenge of news recommendation is to recommend newly published articles. Unlike other domains, outdated items are considered to be irrelevant in the news recommendation task. Another challenge is that the recommendation candidates are not seen in the training phase. In this paper, we introduce deep neural network models to overcome these challenges. we propose a modified session-based Recurrent Neural Network (RNN) model tailored to news recommendation as well as a history-based RNN model that spans the whole user's past histories. Finally, we propose a Convolutional Neural Network (CNN) model to capture user preferences and to personalize recommendation results. Experimental results on real-world news dataset shows that our model outperforms competitive baselines.
{"title":"Deep Neural Networks for News Recommendations","authors":"Keunchan Park, Jisoo Lee, Jaeho Choi","doi":"10.1145/3132847.3133154","DOIUrl":"https://doi.org/10.1145/3132847.3133154","url":null,"abstract":"A fundamental role of news websites is to recommend articles that are interesting to read. The key challenge of news recommendation is to recommend newly published articles. Unlike other domains, outdated items are considered to be irrelevant in the news recommendation task. Another challenge is that the recommendation candidates are not seen in the training phase. In this paper, we introduce deep neural network models to overcome these challenges. we propose a modified session-based Recurrent Neural Network (RNN) model tailored to news recommendation as well as a history-based RNN model that spans the whole user's past histories. Finally, we propose a Convolutional Neural Network (CNN) model to capture user preferences and to personalize recommendation results. Experimental results on real-world news dataset shows that our model outperforms competitive baselines.","PeriodicalId":20449,"journal":{"name":"Proceedings of the 2017 ACM on Conference on Information and Knowledge Management","volume":"17 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2017-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75679456","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}
Time series are ubiquitous in the world since they are used to measure various phenomena (e.g., temperature, spread of a virus, sales, etc.). Forecasting of time series is highly beneficial (and necessary) for optimizing decisions, yet is a very challenging problem; using only the historical values of the time series is often insufficient. In this paper, we study how to construct effective additional features based on related text data for time series forecasting. Besides the commonly used n-gram features, we propose a general strategy for constructing multiple topical features based on the topics discovered by a topic model. We evaluate feature effectiveness using a data set for predicting stock price changes where we constructed additional features from news text articles for stock market prediction. We found that: 1) Text-based features outperform time series-based features, suggesting the great promise of leveraging text data for improving time series forecasting. 2) Topic-based features are not very effective stand-alone, but they can further improve performance when added on top of n-gram features. 3) The best topic-based feature appears to be a long-term aggregation of topics over time with high weights on recent topics.
{"title":"A Study of Feature Construction for Text-based Forecasting of Time Series Variables","authors":"Yiren Wang, Dominic Seyler, Shubhra (Santu) Karmaker, ChengXiang Zhai","doi":"10.1145/3132847.3133109","DOIUrl":"https://doi.org/10.1145/3132847.3133109","url":null,"abstract":"Time series are ubiquitous in the world since they are used to measure various phenomena (e.g., temperature, spread of a virus, sales, etc.). Forecasting of time series is highly beneficial (and necessary) for optimizing decisions, yet is a very challenging problem; using only the historical values of the time series is often insufficient. In this paper, we study how to construct effective additional features based on related text data for time series forecasting. Besides the commonly used n-gram features, we propose a general strategy for constructing multiple topical features based on the topics discovered by a topic model. We evaluate feature effectiveness using a data set for predicting stock price changes where we constructed additional features from news text articles for stock market prediction. We found that: 1) Text-based features outperform time series-based features, suggesting the great promise of leveraging text data for improving time series forecasting. 2) Topic-based features are not very effective stand-alone, but they can further improve performance when added on top of n-gram features. 3) The best topic-based feature appears to be a long-term aggregation of topics over time with high weights on recent topics.","PeriodicalId":20449,"journal":{"name":"Proceedings of the 2017 ACM on Conference on Information and Knowledge Management","volume":"36 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2017-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74789308","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}
Conventional spatial keyword queries confront the difficulty of returning desired objects that are synonyms but morphologically different to query keywords. To overcome this flaw, this paper investigates the interactive spatial keyword querying with semantics. It aims to enhance the conventional queries by not only making sense of the query keywords, but also refining the understanding of query semantics through interactions. On top of the probabilistic topic model, a novel interactive strategy is proposed to precisely infer the latent query semantics by learning from user feedbacks. In each interaction, the returned objects are carefully selected to ensure effective inference of user intended query semantics. Query processing is carried out on a small candidate object set at each round of interaction, and the whole querying process terminates when the latent query semantics learned from user feedback becomes explicit enough. The experimental results on real check-in dataset demonstrates that the quality of results has been significantly improved through limited number of interactions.
{"title":"Interactive Spatial Keyword Querying with Semantics","authors":"Jiabao Sun, Jiajie Xu, Kai Zheng, Chengfei Liu","doi":"10.1145/3132847.3132969","DOIUrl":"https://doi.org/10.1145/3132847.3132969","url":null,"abstract":"Conventional spatial keyword queries confront the difficulty of returning desired objects that are synonyms but morphologically different to query keywords. To overcome this flaw, this paper investigates the interactive spatial keyword querying with semantics. It aims to enhance the conventional queries by not only making sense of the query keywords, but also refining the understanding of query semantics through interactions. On top of the probabilistic topic model, a novel interactive strategy is proposed to precisely infer the latent query semantics by learning from user feedbacks. In each interaction, the returned objects are carefully selected to ensure effective inference of user intended query semantics. Query processing is carried out on a small candidate object set at each round of interaction, and the whole querying process terminates when the latent query semantics learned from user feedback becomes explicit enough. The experimental results on real check-in dataset demonstrates that the quality of results has been significantly improved through limited number of interactions.","PeriodicalId":20449,"journal":{"name":"Proceedings of the 2017 ACM on Conference on Information and Knowledge Management","volume":"14 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2017-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74343076","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}
CNN (Convolution Neural Network) is widely used in visual analysis and achieves exceptionally high performances in image classification, face detection, object recognition, image recoloring, and other learning jobs. Using deep learning frameworks, such as Torch and Tensorflow, CNN can be efficiently computed by leveraging the power of GPU. However, one drawback of GPU is its limited memory which prohibits us from handling large images. Passing a 4K resolution image to the VGG network will result in an exception of out-of-memory for Titan-X GPU. In this paper, we propose a new approach that adopts the BSP (bulk synchronization parallel) model to compute CNNs for images of any size. Before fed to a specific CNN layer, the image is split into smaller pieces which go through the neural network separately. Then, a specific padding and normalization technique is adopted to merge sub-images back into one image. Our approach can be easily extended to support distributed multi-GPUs. In this paper, we use neural style network as our example to illustrate the effectiveness of our approach. We show that using one Titan-X GPU, we can transfer the style of an image with 10,000×10,000 pixels within 1 minute.
{"title":"A New Approach to Compute CNNs for Extremely Large Images","authors":"Sai Wu, Mengdan Zhang, Gang Chen, Ke Chen","doi":"10.1145/3132847.3132872","DOIUrl":"https://doi.org/10.1145/3132847.3132872","url":null,"abstract":"CNN (Convolution Neural Network) is widely used in visual analysis and achieves exceptionally high performances in image classification, face detection, object recognition, image recoloring, and other learning jobs. Using deep learning frameworks, such as Torch and Tensorflow, CNN can be efficiently computed by leveraging the power of GPU. However, one drawback of GPU is its limited memory which prohibits us from handling large images. Passing a 4K resolution image to the VGG network will result in an exception of out-of-memory for Titan-X GPU. In this paper, we propose a new approach that adopts the BSP (bulk synchronization parallel) model to compute CNNs for images of any size. Before fed to a specific CNN layer, the image is split into smaller pieces which go through the neural network separately. Then, a specific padding and normalization technique is adopted to merge sub-images back into one image. Our approach can be easily extended to support distributed multi-GPUs. In this paper, we use neural style network as our example to illustrate the effectiveness of our approach. We show that using one Titan-X GPU, we can transfer the style of an image with 10,000×10,000 pixels within 1 minute.","PeriodicalId":20449,"journal":{"name":"Proceedings of the 2017 ACM on Conference on Information and Knowledge Management","volume":"7 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2017-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73334645","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}
Predicting the popularity of user-created recipes has great potential to be adopted in several applications on recipe-sharing websites. To ensure timely prediction when a recipe is uploaded, a prediction model needs to be trained based on the recipe's content features (i.e., its visual and semantic features). This paper presents a novel approach to predicting recipe popularity using deep visual-semantic fusion. We first pre-train a deep model that predicts the popularity of recipes based on each single modality. We insert additional layers to the two models and concatenate their activations. Finally, we train a network comprising fully connected (FC) layers on the fused features to learn more powerful features, which are used for training a regressor. Based on experiments conducted on more than 150K recipes collected from the Cookpad website, we present a comprehensive comparison with several baselines to verify the effectiveness of our method. The best practice for the proposed method is also described.
{"title":"Recipe Popularity Prediction with Deep Visual-Semantic Fusion","authors":"Satoshi Sanjo, Marie Katsurai","doi":"10.1145/3132847.3133137","DOIUrl":"https://doi.org/10.1145/3132847.3133137","url":null,"abstract":"Predicting the popularity of user-created recipes has great potential to be adopted in several applications on recipe-sharing websites. To ensure timely prediction when a recipe is uploaded, a prediction model needs to be trained based on the recipe's content features (i.e., its visual and semantic features). This paper presents a novel approach to predicting recipe popularity using deep visual-semantic fusion. We first pre-train a deep model that predicts the popularity of recipes based on each single modality. We insert additional layers to the two models and concatenate their activations. Finally, we train a network comprising fully connected (FC) layers on the fused features to learn more powerful features, which are used for training a regressor. Based on experiments conducted on more than 150K recipes collected from the Cookpad website, we present a comprehensive comparison with several baselines to verify the effectiveness of our method. The best practice for the proposed method is also described.","PeriodicalId":20449,"journal":{"name":"Proceedings of the 2017 ACM on Conference on Information and Knowledge Management","volume":"94 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2017-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79279835","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 online advertising, Click-Through Rate (CTR) prediction is a crucial task, as it may benefit the ranking and pricing of online ads. To the best of our knowledge, most of the existing CTR prediction methods are shallow layer models (e.g., Logistic Regression and Factorization Machines) or deep layer models (e.g., Neural Networks). Unfortunately, the shallow layer models cannot capture or utilize high-order nonlinear features in ad data. On the other side, the deep layer models cannot satisfy the necessity of updating CTR models online efficiently due to their high computational complexity. To address the shortcomings above, in this paper, we propose a novel hybrid method based on feature learning of both Deep and Shallow Layers (DSL). In DSL, we utilize Deep Neural Network as a deep layer model trained offline to learn high-order nonlinear features and use Factorization Machines as a shallow layer model for CTR prediction. Furthermore, we also develop an online learning implementation based on DSL, i.e., onlineDSL. Extensive experiments on large-scale real-world datasets clearly validate the effectiveness of our DSL method and onlineDSL algorithm compared with several state-of-the-art baselines.
{"title":"An Ad CTR Prediction Method Based on Feature Learning of Deep and Shallow Layers","authors":"Zai Huang, Zhen Pan, Qi Liu, Bai Long, Haiping Ma, Enhong Chen","doi":"10.1145/3132847.3133072","DOIUrl":"https://doi.org/10.1145/3132847.3133072","url":null,"abstract":"In online advertising, Click-Through Rate (CTR) prediction is a crucial task, as it may benefit the ranking and pricing of online ads. To the best of our knowledge, most of the existing CTR prediction methods are shallow layer models (e.g., Logistic Regression and Factorization Machines) or deep layer models (e.g., Neural Networks). Unfortunately, the shallow layer models cannot capture or utilize high-order nonlinear features in ad data. On the other side, the deep layer models cannot satisfy the necessity of updating CTR models online efficiently due to their high computational complexity. To address the shortcomings above, in this paper, we propose a novel hybrid method based on feature learning of both Deep and Shallow Layers (DSL). In DSL, we utilize Deep Neural Network as a deep layer model trained offline to learn high-order nonlinear features and use Factorization Machines as a shallow layer model for CTR prediction. Furthermore, we also develop an online learning implementation based on DSL, i.e., onlineDSL. Extensive experiments on large-scale real-world datasets clearly validate the effectiveness of our DSL method and onlineDSL algorithm compared with several state-of-the-art baselines.","PeriodicalId":20449,"journal":{"name":"Proceedings of the 2017 ACM on Conference on Information and Knowledge Management","volume":"6 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2017-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79343953","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}
If search engines are ever to index all of the spoken content in the world, they will need to handle hundreds of languages for which no automatic speech recognition systems exist. Zero-resource spoken term discovery, in which repeated content is detected in some acoustic representation, offers a potentially useful source of indexing features. This paper describes a text-based simulation of a zero-resource spoken term discovery system that allows any information retrieval test collection to be used as a basis for early development of information retrieval techniques. It is proposed that these techniques can be later applied to actual zero-resource spoken term discovery results.
{"title":"Simulating Zero-Resource Spoken Term Discovery","authors":"Jerome White, Douglas W. Oard","doi":"10.1145/3132847.3133160","DOIUrl":"https://doi.org/10.1145/3132847.3133160","url":null,"abstract":"If search engines are ever to index all of the spoken content in the world, they will need to handle hundreds of languages for which no automatic speech recognition systems exist. Zero-resource spoken term discovery, in which repeated content is detected in some acoustic representation, offers a potentially useful source of indexing features. This paper describes a text-based simulation of a zero-resource spoken term discovery system that allows any information retrieval test collection to be used as a basis for early development of information retrieval techniques. It is proposed that these techniques can be later applied to actual zero-resource spoken term discovery results.","PeriodicalId":20449,"journal":{"name":"Proceedings of the 2017 ACM on Conference on Information and Knowledge Management","volume":"2 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2017-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84210329","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}
Daizong Ding, Mi Zhang, Shao-Yuan Li, Jie Tang, Xiaotie Chen, Zhi-Hua Zhou
Friendship is the cornerstone to build a social network. In online social networks, statistics show that the leading reason for user to create a new friendship is due to recommendation. Thus the accuracy of recommendation matters. In this paper, we propose a Bayesian Personalized Ranking Deep Neural Network (BayDNN) model for friend recommendation in social networks. With BayDNN, we achieve significant improvement on two public datasets: Epinions and Slashdot. For example, on Epinions dataset, BayDNN significantly outperforms the state-of-the-art algorithms, with a 5% improvement on NDCG over the best baseline. The advantages of the proposed BayDNN mainly come from its underlying convolutional neural network (CNN), which offers a mechanism to extract latent deep structural feature representations of the complicated network data, and a novel Bayesian personalized ranking idea, which precisely captures the users' personal bias based on the extracted deep features. To get good parameter estimation for the neural network, we present a fine-tuned pre-training strategy for the proposed BayDNN model based on Poisson and Bernoulli probabilistic models.
{"title":"BayDNN","authors":"Daizong Ding, Mi Zhang, Shao-Yuan Li, Jie Tang, Xiaotie Chen, Zhi-Hua Zhou","doi":"10.1145/3132847.3132941","DOIUrl":"https://doi.org/10.1145/3132847.3132941","url":null,"abstract":"Friendship is the cornerstone to build a social network. In online social networks, statistics show that the leading reason for user to create a new friendship is due to recommendation. Thus the accuracy of recommendation matters. In this paper, we propose a Bayesian Personalized Ranking Deep Neural Network (BayDNN) model for friend recommendation in social networks. With BayDNN, we achieve significant improvement on two public datasets: Epinions and Slashdot. For example, on Epinions dataset, BayDNN significantly outperforms the state-of-the-art algorithms, with a 5% improvement on NDCG over the best baseline. The advantages of the proposed BayDNN mainly come from its underlying convolutional neural network (CNN), which offers a mechanism to extract latent deep structural feature representations of the complicated network data, and a novel Bayesian personalized ranking idea, which precisely captures the users' personal bias based on the extracted deep features. To get good parameter estimation for the neural network, we present a fine-tuned pre-training strategy for the proposed BayDNN model based on Poisson and Bernoulli probabilistic models.","PeriodicalId":20449,"journal":{"name":"Proceedings of the 2017 ACM on Conference on Information and Knowledge Management","volume":"5 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2017-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81971266","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}
Recommending a ranked list of interesting venues to users based on their preferences has become a key functionality in Location-Based Social Networks (LBSNs) such as Yelp and Gowalla. Bayesian Personalised Ranking (BPR) is a popular pairwise recommendation technique that is used to generate the ranked list of venues of interest to a user, by leveraging the user's implicit feedback such as their check-ins as instances of positive feedback, while randomly sampling other venues as negative instances. To alleviate the sparsity that affects the usefulness of recommendations by BPR for users with few check-ins, various approaches have been proposed in the literature to incorporate additional sources of information such as the social links between users, the textual content of comments, as well as the geographical location of the venues. However, such approaches can only readily leverage one source of additional information for negative sampling. Instead, we propose a novel Personalised Ranking Framework with Multiple sampling Criteria (PRFMC) that leverages both geographical influence and social correlation to enhance the effectiveness of BPR. In particular, we apply a multi-centre Gaussian model and a power-law distribution method, to capture geographical influence and social correlation when sampling negative venues, respectively. Finally, we conduct comprehensive experiments using three large-scale datasets from the Yelp, Gowalla and Brightkite LBSNs. The experimental results demonstrate the effectiveness of fusing both geographical influence and social correlation in our proposed PRFMC framework and its superiority in comparison to BPR-based and other similar ranking approaches. Indeed, our PRFMC approach attains a 37% improvement in MRR over a recently proposed approach that identifies negative venues only from social links.
{"title":"A Personalised Ranking Framework with Multiple Sampling Criteria for Venue Recommendation","authors":"Jarana Manotumruksa, C. Macdonald, I. Ounis","doi":"10.1145/3132847.3132985","DOIUrl":"https://doi.org/10.1145/3132847.3132985","url":null,"abstract":"Recommending a ranked list of interesting venues to users based on their preferences has become a key functionality in Location-Based Social Networks (LBSNs) such as Yelp and Gowalla. Bayesian Personalised Ranking (BPR) is a popular pairwise recommendation technique that is used to generate the ranked list of venues of interest to a user, by leveraging the user's implicit feedback such as their check-ins as instances of positive feedback, while randomly sampling other venues as negative instances. To alleviate the sparsity that affects the usefulness of recommendations by BPR for users with few check-ins, various approaches have been proposed in the literature to incorporate additional sources of information such as the social links between users, the textual content of comments, as well as the geographical location of the venues. However, such approaches can only readily leverage one source of additional information for negative sampling. Instead, we propose a novel Personalised Ranking Framework with Multiple sampling Criteria (PRFMC) that leverages both geographical influence and social correlation to enhance the effectiveness of BPR. In particular, we apply a multi-centre Gaussian model and a power-law distribution method, to capture geographical influence and social correlation when sampling negative venues, respectively. Finally, we conduct comprehensive experiments using three large-scale datasets from the Yelp, Gowalla and Brightkite LBSNs. The experimental results demonstrate the effectiveness of fusing both geographical influence and social correlation in our proposed PRFMC framework and its superiority in comparison to BPR-based and other similar ranking approaches. Indeed, our PRFMC approach attains a 37% improvement in MRR over a recently proposed approach that identifies negative venues only from social links.","PeriodicalId":20449,"journal":{"name":"Proceedings of the 2017 ACM on Conference on Information and Knowledge Management","volume":"32 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2017-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82022287","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}
Weijie Bian, Si Li, Zhao Yang, Guang Chen, Zhiqing Lin
Answer selection for question answering is a challenging task, since it requires effective capture of the complex semantic relations between questions and answers. Previous remarkable approaches mainly adopt general Compare-Aggregate framework that performs word-level comparison and aggregation. In this paper, unlike previous Compare-Aggregate models which utilize the traditional attention mechanism to generate corresponding word-level vector before comparison, we propose a novel attention mechanism named Dynamic-Clip Attention which is directly integrated into the Compare-Aggregate framework. Dynamic-Clip Attention focuses on filtering out noise in attention matrix, in order to better mine the semantic relevance of word-level vectors. At the same time, different from previous Compare-Aggregate works which treat answer selection task as a pointwise classification problem, we propose a listwise ranking approach to model this task to learn the relative order of candidate answers. Experiments on TrecQA and WikiQA datasets show that our proposed model achieves the state-of-the-art performance.
{"title":"A Compare-Aggregate Model with Dynamic-Clip Attention for Answer Selection","authors":"Weijie Bian, Si Li, Zhao Yang, Guang Chen, Zhiqing Lin","doi":"10.1145/3132847.3133089","DOIUrl":"https://doi.org/10.1145/3132847.3133089","url":null,"abstract":"Answer selection for question answering is a challenging task, since it requires effective capture of the complex semantic relations between questions and answers. Previous remarkable approaches mainly adopt general Compare-Aggregate framework that performs word-level comparison and aggregation. In this paper, unlike previous Compare-Aggregate models which utilize the traditional attention mechanism to generate corresponding word-level vector before comparison, we propose a novel attention mechanism named Dynamic-Clip Attention which is directly integrated into the Compare-Aggregate framework. Dynamic-Clip Attention focuses on filtering out noise in attention matrix, in order to better mine the semantic relevance of word-level vectors. At the same time, different from previous Compare-Aggregate works which treat answer selection task as a pointwise classification problem, we propose a listwise ranking approach to model this task to learn the relative order of candidate answers. Experiments on TrecQA and WikiQA datasets show that our proposed model achieves the state-of-the-art performance.","PeriodicalId":20449,"journal":{"name":"Proceedings of the 2017 ACM on Conference on Information and Knowledge Management","volume":"26 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2017-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84699126","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：