Mercan Topkara, Justin D. Weisz, Shimei Pan, Jie Lu, J. Lai
Expertise and skill assessments are a common aspect of working in an enterprise, but manual assessments are onerous and quickly outdated. Automated assessments can alleviate these problems, albeit at the risk of being inaccurate. In this short paper, we focus on the problem of how to design an engaging learning system in the presence of potentially inaccurate automated expertise assessments; especially when the users are in their early stage of using the system. We explore two dimensions associated with reporting automated expertise assessments to users: i) the inclusion of a social comparison, and ii) the precision of how expertise scores are presented. In a controlled experiment (N=60), we examined the impact of these dimensions on the perceived accuracy of the assessments, the perceived utility of the system, and peoples' willingness to share expertise scores within the enterprise.
{"title":"Dare to Compare: Motivating Expertise Building in the Enterprise through Intelligent User Modeling Interfaces","authors":"Mercan Topkara, Justin D. Weisz, Shimei Pan, Jie Lu, J. Lai","doi":"10.1145/2665994.2665998","DOIUrl":"https://doi.org/10.1145/2665994.2665998","url":null,"abstract":"Expertise and skill assessments are a common aspect of working in an enterprise, but manual assessments are onerous and quickly outdated. Automated assessments can alleviate these problems, albeit at the risk of being inaccurate. In this short paper, we focus on the problem of how to design an engaging learning system in the presence of potentially inaccurate automated expertise assessments; especially when the users are in their early stage of using the system. We explore two dimensions associated with reporting automated expertise assessments to users: i) the inclusion of a social comparison, and ii) the precision of how expertise scores are presented. In a controlled experiment (N=60), we examined the impact of these dimensions on the perceived accuracy of the assessments, the perceived utility of the system, and peoples' willingness to share expertise scores within the enterprise.","PeriodicalId":346862,"journal":{"name":"DUBMOD '14","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123780058","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}
Collaborative filtering (CF), which plays an important role in making personalized recommendation, is one of the most traditional and effective recommendation algorithms. However, there are several factors that impact its recommendation accuracy, e.g., the sparse matrix problem. In the past studies, most researchers merely focused on user ratings to model user profile but ignored the implying patterns. In this paper, we utilize user activity to discriminate user rating patterns and propose a new method of user-based collaborative filtering based on user activity level. Experimental results on movie-lens data-set has proved that the algorithm we proposed improves recommendation accuracy significantly compared with traditional user-based CF algorithm with respect to various evaluation metrics.
{"title":"An Improved Collaborative Filtering Algorithm Combining User Activity Level","authors":"Jiaqi Fan, Lisi Jiang, Weimin Pan","doi":"10.1145/2665994.2665995","DOIUrl":"https://doi.org/10.1145/2665994.2665995","url":null,"abstract":"Collaborative filtering (CF), which plays an important role in making personalized recommendation, is one of the most traditional and effective recommendation algorithms. However, there are several factors that impact its recommendation accuracy, e.g., the sparse matrix problem. In the past studies, most researchers merely focused on user ratings to model user profile but ignored the implying patterns. In this paper, we utilize user activity to discriminate user rating patterns and propose a new method of user-based collaborative filtering based on user activity level. Experimental results on movie-lens data-set has proved that the algorithm we proposed improves recommendation accuracy significantly compared with traditional user-based CF algorithm with respect to various evaluation metrics.","PeriodicalId":346862,"journal":{"name":"DUBMOD '14","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126239376","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}
Session search retrieves documents for a sequence of queries in a session. Prior research demonstrated that query aggregation is an effective technique for session search. This paper proposes a novel query aggregation scheme based on the discount factor in reinforcement learning. Moreover, we compare various query aggregation schemes and investigate the best scheme for aggregating queries in session search. Evaluation conducted over TREC 2011 and 2012 shows that the proposed scheme works the best and outperforms the TREC best system as well as learned weights by learning to rank.
{"title":"Query Aggregation in Session Search","authors":"Dongyi Guan, G. Yang","doi":"10.1145/2665994.2666001","DOIUrl":"https://doi.org/10.1145/2665994.2666001","url":null,"abstract":"Session search retrieves documents for a sequence of queries in a session. Prior research demonstrated that query aggregation is an effective technique for session search. This paper proposes a novel query aggregation scheme based on the discount factor in reinforcement learning. Moreover, we compare various query aggregation schemes and investigate the best scheme for aggregating queries in session search. Evaluation conducted over TREC 2011 and 2012 shows that the proposed scheme works the best and outperforms the TREC best system as well as learned weights by learning to rank.","PeriodicalId":346862,"journal":{"name":"DUBMOD '14","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131651828","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}
Personalization plays a significant role in persuasion. In this paper, we present a framework called PPLUM which combines social media-based large-scale user modeling with automated personal persuasion generation. This work has many applications such as advertising consumer products, promoting political candidates, or encouraging good health and investment behaviors.
{"title":"PPLUM: A Framework for Large-Scale Personal Persuasion","authors":"Shimei Pan, Michelle X. Zhou","doi":"10.1145/2665994.2665999","DOIUrl":"https://doi.org/10.1145/2665994.2665999","url":null,"abstract":"Personalization plays a significant role in persuasion. In this paper, we present a framework called PPLUM which combines social media-based large-scale user modeling with automated personal persuasion generation. This work has many applications such as advertising consumer products, promoting political candidates, or encouraging good health and investment behaviors.","PeriodicalId":346862,"journal":{"name":"DUBMOD '14","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127664020","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}
There were 6.8 billion estimates for mobile subscriptions worldwide by end of 2013 [11]. As the mobile market gets saturated, it becomes harder for telecom providers to acquire new customers, and makes it essential for them to retain their own. Due to the high competition between different telecom providers and the ability of customers to move from one provider to another, all telecom service providers suffer from customer churn. As a result, churn prediction has become one of the main telecom challenges. The primary goal of churn prediction is to predict a list of potential churners, so that telecom providers can start targeting them by retention campaigns. This work describes work in progress in which we model churn as a dyadic social behavior, where customer churn propagates in the telecom network over strong social ties. We propose a novel method for measuring social tie strength between telecom customers. We then, incorporate strong social ties in an influence propagation model, and apply a machine-learning based prediction model that combines both churn social influence and other traditional churn factors. The goals of our proposed model is to enhance churn prediction by modeling churn as a dyadic phenomena, provide an enhanced evaluation for the social tie strength based on customers social interactions, and to study the effect of strong social ties on churn propagation over mobile telecom networks.
{"title":"Enhanced Customer Churn Prediction using Social Network Analysis","authors":"Marwa N. Abd-Allah, A. Salah, S. El-Beltagy","doi":"10.1145/2665994.2665997","DOIUrl":"https://doi.org/10.1145/2665994.2665997","url":null,"abstract":"There were 6.8 billion estimates for mobile subscriptions worldwide by end of 2013 [11]. As the mobile market gets saturated, it becomes harder for telecom providers to acquire new customers, and makes it essential for them to retain their own. Due to the high competition between different telecom providers and the ability of customers to move from one provider to another, all telecom service providers suffer from customer churn. As a result, churn prediction has become one of the main telecom challenges. The primary goal of churn prediction is to predict a list of potential churners, so that telecom providers can start targeting them by retention campaigns. This work describes work in progress in which we model churn as a dyadic social behavior, where customer churn propagates in the telecom network over strong social ties. We propose a novel method for measuring social tie strength between telecom customers. We then, incorporate strong social ties in an influence propagation model, and apply a machine-learning based prediction model that combines both churn social influence and other traditional churn factors. The goals of our proposed model is to enhance churn prediction by modeling churn as a dyadic phenomena, provide an enhanced evaluation for the social tie strength based on customers social interactions, and to study the effect of strong social ties on churn propagation over mobile telecom networks.","PeriodicalId":346862,"journal":{"name":"DUBMOD '14","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123945557","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 rise of online social media has led to an explosion in user-generated content. However, user-generated content is difficult to analyze in isolation from its context. Accordingly, context detection and tracking its evolution is essential to understanding social media. This paper presents a statistical model that can detect interpretable topics along with their contexts. A topic is represented by a cluster of words that frequently occur together, and a context is represented by a cluster of hashtags that frequently occur with a topic. The model combines a context with a related topic by jointly modeling words with hashtags and time. Experiments on real datasets demonstrate that the proposed model successfully discovers both meaningful topics and contexts, and tracks their evolution.
{"title":"Context over Time: Modeling Context Evolution in Social Media","authors":"Md. Hijbul Alam, Woo-Jong Ryu, SangKeun Lee","doi":"10.1145/2665994.2665996","DOIUrl":"https://doi.org/10.1145/2665994.2665996","url":null,"abstract":"The rise of online social media has led to an explosion in user-generated content. However, user-generated content is difficult to analyze in isolation from its context. Accordingly, context detection and tracking its evolution is essential to understanding social media. This paper presents a statistical model that can detect interpretable topics along with their contexts. A topic is represented by a cluster of words that frequently occur together, and a context is represented by a cluster of hashtags that frequently occur with a topic. The model combines a context with a related topic by jointly modeling words with hashtags and time. Experiments on real datasets demonstrate that the proposed model successfully discovers both meaningful topics and contexts, and tracks their evolution.","PeriodicalId":346862,"journal":{"name":"DUBMOD '14","volume":"106 1-2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132814594","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}
Pinterest is a website and mobile application that allows users to discover, save, and share content ('pins') across a wide range of interest areas. As the user base grows more diverse both demographically and psychographically, we wish to understand emerging patterns of behavior that reflect underlying differences in users' intent and satisfaction with the service. In this paper, we propose a methodology for generating a meaningful segmentation of Pinterest users based on three types of behavior: (1) engagement with various categories of content, (2) frequencies of various types of actions, and (3) sequences of actions.
{"title":"Behavioral Segmentation of Pinterest Users","authors":"Jolie M. Martin","doi":"10.1145/2665994.2666000","DOIUrl":"https://doi.org/10.1145/2665994.2666000","url":null,"abstract":"Pinterest is a website and mobile application that allows users to discover, save, and share content ('pins') across a wide range of interest areas. As the user base grows more diverse both demographically and psychographically, we wish to understand emerging patterns of behavior that reflect underlying differences in users' intent and satisfaction with the service. In this paper, we propose a methodology for generating a meaningful segmentation of Pinterest users based on three types of behavior: (1) engagement with various categories of content, (2) frequencies of various types of actions, and (3) sequences of actions.","PeriodicalId":346862,"journal":{"name":"DUBMOD '14","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120962108","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：