Web trackers can build accurate topical user profiles(e.g., in terms of habits and personal characteristics) by monitoring a user's browsing activities across websites. This process, known as behavioral targeting, has a number of practical benefits but it also raises privacy concerns. Most existing techniques either try to block web tracking altogether or aim to endow it with privacy preserving mechanisms, but they are system-centered rather than user-centered. Nowadays, the majority of users want to have some degree of control over their privacy, while their perspectives and feelings towards web tracking maybe different, ranging from a desire to avoid being profiled at all to a willingness to trade personal information for better services. Regardless of a specific user's preference, from a technical point of view there is is no simple way for him/her to monitor, let alone to influence, the behavior of web trackers. In this paper, we describe an approach which makes users aware of their likely tracking profile and gives them the possibility to bias the profile towards both ends of the web tracking spectrum, either by improving its accuracy beyond the tracker capabilities (thus emphasizing behavioral targeting) or by filling in false interests(thus increasing privacy). This goal is achieved by simulating the process of learning a user profile on the part of the tracker and then by retrofitting a web traffic suitable for producing the desired profile. Our approach has been implemented as a web browser extension called ManTra (Management of Tracking). The system has been evaluated in several dimensions, including its ability to learn an accurate ad-oriented user profile and to influence the behavior of a commercial tool for web tracking personalization, i.e., Google's Ads Settings.
{"title":"Enhancing User Awareness and Control of Web Tracking with ManTra","authors":"D. Re, Claudio Carpineto","doi":"10.1109/WI.2016.0061","DOIUrl":"https://doi.org/10.1109/WI.2016.0061","url":null,"abstract":"Web trackers can build accurate topical user profiles(e.g., in terms of habits and personal characteristics) by monitoring a user's browsing activities across websites. This process, known as behavioral targeting, has a number of practical benefits but it also raises privacy concerns. Most existing techniques either try to block web tracking altogether or aim to endow it with privacy preserving mechanisms, but they are system-centered rather than user-centered. Nowadays, the majority of users want to have some degree of control over their privacy, while their perspectives and feelings towards web tracking maybe different, ranging from a desire to avoid being profiled at all to a willingness to trade personal information for better services. Regardless of a specific user's preference, from a technical point of view there is is no simple way for him/her to monitor, let alone to influence, the behavior of web trackers. In this paper, we describe an approach which makes users aware of their likely tracking profile and gives them the possibility to bias the profile towards both ends of the web tracking spectrum, either by improving its accuracy beyond the tracker capabilities (thus emphasizing behavioral targeting) or by filling in false interests(thus increasing privacy). This goal is achieved by simulating the process of learning a user profile on the part of the tracker and then by retrofitting a web traffic suitable for producing the desired profile. Our approach has been implemented as a web browser extension called ManTra (Management of Tracking). The system has been evaluated in several dimensions, including its ability to learn an accurate ad-oriented user profile and to influence the behavior of a commercial tool for web tracking personalization, i.e., Google's Ads Settings.","PeriodicalId":6513,"journal":{"name":"2016 IEEE/WIC/ACM International Conference on Web Intelligence (WI)","volume":"24 96 1","pages":"391-398"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83248775","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}
Masatoshi Suzuki, Koji Matsuda, S. Sekine, Naoaki Okazaki, Kentaro Inui
This paper addresses the task of assigning multiple labels of fine-grained named entity (NE) types to Wikipedia articles. To address the sparseness of the input feature space, which is salient particularly in fine-grained type classification, we propose to learn article vectors (i.e. entity embeddings) from hypertext structure of Wikipedia using a Skip-gram model and incorporate them into the input feature set. To conduct large-scale practical experiments, we created a new dataset containing over 22,000 manually labeled instances. The results of our experiments show that our idea gained statistically significant improvements in classification results.
{"title":"Fine-Grained Named Entity Classification with Wikipedia Article Vectors","authors":"Masatoshi Suzuki, Koji Matsuda, S. Sekine, Naoaki Okazaki, Kentaro Inui","doi":"10.1109/WI.2016.0080","DOIUrl":"https://doi.org/10.1109/WI.2016.0080","url":null,"abstract":"This paper addresses the task of assigning multiple labels of fine-grained named entity (NE) types to Wikipedia articles. To address the sparseness of the input feature space, which is salient particularly in fine-grained type classification, we propose to learn article vectors (i.e. entity embeddings) from hypertext structure of Wikipedia using a Skip-gram model and incorporate them into the input feature set. To conduct large-scale practical experiments, we created a new dataset containing over 22,000 manually labeled instances. The results of our experiments show that our idea gained statistically significant improvements in classification results.","PeriodicalId":6513,"journal":{"name":"2016 IEEE/WIC/ACM International Conference on Web Intelligence (WI)","volume":"1 1","pages":"483-486"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91054755","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}
R. Nielek, Oskar Jarczyk, Kamil Pawlak, Leszek Bukowski, Roman Bartusiak, A. Wierzbicki
GitHub is one of the most commonly used web-based code repository hosting service. Majority of projects hosted on GitHub are really small but, on the other hand, developers spend most of their time working in medium to large repositories. Developers can freely join and leave projects following their current needs and interests. Based on real data collected from GitHub we have tried to predict which developer will join which project. A mix of carefully selected list of features and machine learning techniques let us achieve a precision of 0.886, in the best case scenario, where there is quite a long history of a user and a repository in the system. Even when proposed classifier faces a cold start problem, it delivers precision equal to 0.729 which is still acceptable for automatic recommendation of noteworthy projects for developers.
{"title":"Choose a Job You Love: Predicting Choices of GitHub Developers","authors":"R. Nielek, Oskar Jarczyk, Kamil Pawlak, Leszek Bukowski, Roman Bartusiak, A. Wierzbicki","doi":"10.1109/WI.2016.0037","DOIUrl":"https://doi.org/10.1109/WI.2016.0037","url":null,"abstract":"GitHub is one of the most commonly used web-based code repository hosting service. Majority of projects hosted on GitHub are really small but, on the other hand, developers spend most of their time working in medium to large repositories. Developers can freely join and leave projects following their current needs and interests. Based on real data collected from GitHub we have tried to predict which developer will join which project. A mix of carefully selected list of features and machine learning techniques let us achieve a precision of 0.886, in the best case scenario, where there is quite a long history of a user and a repository in the system. Even when proposed classifier faces a cold start problem, it delivers precision equal to 0.729 which is still acceptable for automatic recommendation of noteworthy projects for developers.","PeriodicalId":6513,"journal":{"name":"2016 IEEE/WIC/ACM International Conference on Web Intelligence (WI)","volume":"28 1","pages":"200-207"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91338052","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}
Majority of the social network analysis studies for counter-terrorism and homeland security consider homogeneous network. However, a terrorist activity (attack) is often defined by several attributes such as terrorist organisation, time, place, attack type etc. To capture inherent dependency between the attributes, we need to adopt a network which is capable of capturing the dependency between the attributes. In this paper, we define a heterogeneous network to represent a collection of terrorist activities. Further, we propose personalised PageRank (PPR) as a method capable of performing various analytical operations over heterogeneous network just by changing model parameters without changing the underlying model. Using global terrorist data (GTD), behavioural network, and news discussion network, we show various applications of PPR for counter-terrorism over heterogeneous network just by changing the model parameter. In addition we propose heterogeneous version of four local proximity based link prediction methods, namely, Common Neighbour, Adamic-Adar, Jaccard Coefficient, and Resource Allocation.
{"title":"Personalised PageRank as a Method of Exploiting Heterogeneous Network for Counter Terrorism and Homeland Security","authors":"Akash Anil, Sanasam Ranbir Singh, R. Sarmah","doi":"10.1109/WI.2016.0053","DOIUrl":"https://doi.org/10.1109/WI.2016.0053","url":null,"abstract":"Majority of the social network analysis studies for counter-terrorism and homeland security consider homogeneous network. However, a terrorist activity (attack) is often defined by several attributes such as terrorist organisation, time, place, attack type etc. To capture inherent dependency between the attributes, we need to adopt a network which is capable of capturing the dependency between the attributes. In this paper, we define a heterogeneous network to represent a collection of terrorist activities. Further, we propose personalised PageRank (PPR) as a method capable of performing various analytical operations over heterogeneous network just by changing model parameters without changing the underlying model. Using global terrorist data (GTD), behavioural network, and news discussion network, we show various applications of PPR for counter-terrorism over heterogeneous network just by changing the model parameter. In addition we propose heterogeneous version of four local proximity based link prediction methods, namely, Common Neighbour, Adamic-Adar, Jaccard Coefficient, and Resource Allocation.","PeriodicalId":6513,"journal":{"name":"2016 IEEE/WIC/ACM International Conference on Web Intelligence (WI)","volume":"24 1","pages":"327-334"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90462547","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 majority of techniques in socio-behavioral modeling tend to consider user-generated content in a bulk, with the assumption that this sort of aggregation would not have any negative impact on overall predictability of the system, which is not necessarily the case. We propose a novel user-centric approach designed specifically to capture most predictive hidden variables that can be discovered in a context of the specific individual. The concept of social filtering closely resembles collaborative filtering with the main difference that none of the considered users intentionally participates in the recommendation process. Its objective is to determine both the subset of best expert users able to reflect a particular social trend of interest and their transformation into feature space used for modeling. We introduce three-step selection procedure that includes activity-and relevance-based filtering and ensemble of expert users, and show that proper choice of expert individuals is critical to prediction quality.
{"title":"Social Filtering: User-Centric Approach to Social Trend Prediction","authors":"Iuliia Chepurna, M. Makrehchi","doi":"10.1109/WI.2016.0115","DOIUrl":"https://doi.org/10.1109/WI.2016.0115","url":null,"abstract":"The majority of techniques in socio-behavioral modeling tend to consider user-generated content in a bulk, with the assumption that this sort of aggregation would not have any negative impact on overall predictability of the system, which is not necessarily the case. We propose a novel user-centric approach designed specifically to capture most predictive hidden variables that can be discovered in a context of the specific individual. The concept of social filtering closely resembles collaborative filtering with the main difference that none of the considered users intentionally participates in the recommendation process. Its objective is to determine both the subset of best expert users able to reflect a particular social trend of interest and their transformation into feature space used for modeling. We introduce three-step selection procedure that includes activity-and relevance-based filtering and ensemble of expert users, and show that proper choice of expert individuals is critical to prediction quality.","PeriodicalId":6513,"journal":{"name":"2016 IEEE/WIC/ACM International Conference on Web Intelligence (WI)","volume":"3 1","pages":"650-655"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84238630","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}
Recommendation Systems (RS) can offer suggestions of items to users. Due to explosive growth of internet, e-commerce, and social networks, RS research has experienced great interest in recent years. Online Analytical Processing (OLAP) and data warehousing technologies have existed for a while and have been popular in many big businesses. In this paper we proposed a new RS system called RS-OLAP which applies the functionalities of OLAP to RS. In particular we aggregate and rollup hierarchical rating data such as users' locations, items' locations and category hierarchies, and incorporate traditional RS algorithms such as Collaborative Filtering (CF) at different levels. In addition, we proposed three other RS algorithms: Top-rated Items in User's Frequent Categories (TIUFC), Pair-wise Association Recommender System (PARS), and RS for spatial items. We also give a framework and prototype for RS-OLAP.
{"title":"A Recommendation System Using OLAP Approach","authors":"Lixin Fu","doi":"10.1109/WI.2016.0109","DOIUrl":"https://doi.org/10.1109/WI.2016.0109","url":null,"abstract":"Recommendation Systems (RS) can offer suggestions of items to users. Due to explosive growth of internet, e-commerce, and social networks, RS research has experienced great interest in recent years. Online Analytical Processing (OLAP) and data warehousing technologies have existed for a while and have been popular in many big businesses. In this paper we proposed a new RS system called RS-OLAP which applies the functionalities of OLAP to RS. In particular we aggregate and rollup hierarchical rating data such as users' locations, items' locations and category hierarchies, and incorporate traditional RS algorithms such as Collaborative Filtering (CF) at different levels. In addition, we proposed three other RS algorithms: Top-rated Items in User's Frequent Categories (TIUFC), Pair-wise Association Recommender System (PARS), and RS for spatial items. We also give a framework and prototype for RS-OLAP.","PeriodicalId":6513,"journal":{"name":"2016 IEEE/WIC/ACM International Conference on Web Intelligence (WI)","volume":"22 1","pages":"622-625"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81919750","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}
3D Surfaces are widely employed to model geometric assets (e.g., mountains on a landscape), which are used in digital animations and video games. A single surface commonly needs to be created and modified by a group of collaborators, but most of the 3D content creation applications are essentially single-user. In addition, such surfaces are visualized in 2D projections, causing confusion to new users, when imagining their shape in 3D. In this paper, we propose a novel approach based on Augmented Reality (AR) to the task of collaboratively authoring surfaces on the Web using mobile devices. We rely on AR technology to help collaborators to easily understand the shape of the surface's 3D representation, and we provide them with the basic authoring tools to intuitively modify its shape. To support real time face-to-face interaction, we implement an object sharing scheme, which according to our results is enough in practice. In this way, our approach is able to create a new online collaborative setting in which a group of collocated participants, each one using a mobile device, or connected to the Web, can concurrently modify a surface, while visualizing it in their own real environment through AR.
{"title":"Collaborative Web Authoring of 3D Surfaces Using Augmented Reality on Mobile Devices","authors":"Andrés Cortés-Dávalos, S. Mendoza","doi":"10.1109/WI.2016.0113","DOIUrl":"https://doi.org/10.1109/WI.2016.0113","url":null,"abstract":"3D Surfaces are widely employed to model geometric assets (e.g., mountains on a landscape), which are used in digital animations and video games. A single surface commonly needs to be created and modified by a group of collaborators, but most of the 3D content creation applications are essentially single-user. In addition, such surfaces are visualized in 2D projections, causing confusion to new users, when imagining their shape in 3D. In this paper, we propose a novel approach based on Augmented Reality (AR) to the task of collaboratively authoring surfaces on the Web using mobile devices. We rely on AR technology to help collaborators to easily understand the shape of the surface's 3D representation, and we provide them with the basic authoring tools to intuitively modify its shape. To support real time face-to-face interaction, we implement an object sharing scheme, which according to our results is enough in practice. In this way, our approach is able to create a new online collaborative setting in which a group of collocated participants, each one using a mobile device, or connected to the Web, can concurrently modify a surface, while visualizing it in their own real environment through AR.","PeriodicalId":6513,"journal":{"name":"2016 IEEE/WIC/ACM International Conference on Web Intelligence (WI)","volume":"6 1","pages":"640-643"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89181881","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 emergence of computing power and the abundance of data have made it possible to assist human decisions, especially in the stock markets, in which the ability to predict future values would lower the risk of investing. In this paper, we present a new approach for identifying the predictive power of public emotions extracted from various sections of daily news articles on the movements of stock market indices. The approach utilizes the results of a lexicon emotion analysis conducted on crowd-annotated news to extract various types of public emotions from daily news articles. We also propose a model and an analysis method to score news articles regarding public emotions, and to identify which news sections and emotions cause movements in a stock market index. The results of an experiment conducted with 24,763 news articles show that some types of public emotions are significantly correlated with changes in the trading volume and the closing price of a stock market.
{"title":"Predictive Power of Public Emotions as Extracted from Daily News Articles on the Movements of Stock Market Indices","authors":"Chayanin Wong, In-Young Ko","doi":"10.1109/WI.2016.0126","DOIUrl":"https://doi.org/10.1109/WI.2016.0126","url":null,"abstract":"The emergence of computing power and the abundance of data have made it possible to assist human decisions, especially in the stock markets, in which the ability to predict future values would lower the risk of investing. In this paper, we present a new approach for identifying the predictive power of public emotions extracted from various sections of daily news articles on the movements of stock market indices. The approach utilizes the results of a lexicon emotion analysis conducted on crowd-annotated news to extract various types of public emotions from daily news articles. We also propose a model and an analysis method to score news articles regarding public emotions, and to identify which news sections and emotions cause movements in a stock market index. The results of an experiment conducted with 24,763 news articles show that some types of public emotions are significantly correlated with changes in the trading volume and the closing price of a stock market.","PeriodicalId":6513,"journal":{"name":"2016 IEEE/WIC/ACM International Conference on Web Intelligence (WI)","volume":"54 1","pages":"705-708"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88246688","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 multi-agent systems, norms are being used to regulate the behavior of the autonomous agents. Norms describe the actions that can be performed, must be performed, and cannot be performed in the system. One of the main challenges on developing normative systems is that norms may conflict with each other. Norms are in conflict when the fulfillment of one norm violates the other and vice-versa. In previous works, the conflict checkers consider that conflicts can be detected by simply analyzing pairs of norms. However, there may be conflicts that can only be detected when we analyze several norms together. In this paper, we present a conflict checker that is able to check direct conflicts among multiple norms and a strategy developed to minimize the complexity of such problem, since the checking of multiple norms is a NP-hard problem. The algorithms are presented, a discussion about its complexity is provided and the validation of the conflict checker is described.
{"title":"An Approach to Verify Conflicts among Multiple Norms in Multi-agent Systems","authors":"E. Silvestre, V. Silva","doi":"10.1109/WI.2016.0058","DOIUrl":"https://doi.org/10.1109/WI.2016.0058","url":null,"abstract":"In multi-agent systems, norms are being used to regulate the behavior of the autonomous agents. Norms describe the actions that can be performed, must be performed, and cannot be performed in the system. One of the main challenges on developing normative systems is that norms may conflict with each other. Norms are in conflict when the fulfillment of one norm violates the other and vice-versa. In previous works, the conflict checkers consider that conflicts can be detected by simply analyzing pairs of norms. However, there may be conflicts that can only be detected when we analyze several norms together. In this paper, we present a conflict checker that is able to check direct conflicts among multiple norms and a strategy developed to minimize the complexity of such problem, since the checking of multiple norms is a NP-hard problem. The algorithms are presented, a discussion about its complexity is provided and the validation of the conflict checker is described.","PeriodicalId":6513,"journal":{"name":"2016 IEEE/WIC/ACM International Conference on Web Intelligence (WI)","volume":"21 1","pages":"367-374"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87851714","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}
Rui Meng, Shuguang Han, Yun Huang, Daqing He, Peter Brusilovsky
Although the volume of online educational resources has dramatically increased in recent years, many of these resources are isolated and distributed in diverse websites and databases. This hinders the discovery and overall usage of online educational resources. By using linking between related subsections of online textbooks as a testbed, this paper explores multiple knowledge-based content linking algorithms for connecting online educational resources. We focus on examining semantic-based methods for identifying important knowledge components in textbooks and their usefulness in linking book subsections. To overcome the data sparsity in representing textbook content, we evaluated the utility of external corpuses, such as more textbooks or other online educational resources in the same domain. Our results show that semantic modeling can be integrated with a term-based approach for additional performance improvement, and that using extra textbooks significantly benefits semantic modeling. Similar results are obtained when we applied the same approach to other domains.
{"title":"Knowledge-Based Content Linking for Online Textbooks","authors":"Rui Meng, Shuguang Han, Yun Huang, Daqing He, Peter Brusilovsky","doi":"10.1109/WI.2016.0014","DOIUrl":"https://doi.org/10.1109/WI.2016.0014","url":null,"abstract":"Although the volume of online educational resources has dramatically increased in recent years, many of these resources are isolated and distributed in diverse websites and databases. This hinders the discovery and overall usage of online educational resources. By using linking between related subsections of online textbooks as a testbed, this paper explores multiple knowledge-based content linking algorithms for connecting online educational resources. We focus on examining semantic-based methods for identifying important knowledge components in textbooks and their usefulness in linking book subsections. To overcome the data sparsity in representing textbook content, we evaluated the utility of external corpuses, such as more textbooks or other online educational resources in the same domain. Our results show that semantic modeling can be integrated with a term-based approach for additional performance improvement, and that using extra textbooks significantly benefits semantic modeling. Similar results are obtained when we applied the same approach to other domains.","PeriodicalId":6513,"journal":{"name":"2016 IEEE/WIC/ACM International Conference on Web Intelligence (WI)","volume":"111 3S 1","pages":"18-25"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75655625","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：