Pub Date : 2017-07-01DOI: 10.1109/BESC.2017.8256383
Charles Chu, J. Brownlow, Qinxue Meng, Bin Fu, Ben Culbert, Min Zhu, Guandong Xu, Xue-zhong He
Time series prediction is a challenging task in reality, and various methods have been proposed for it. However, only the historical series of values are exploited in most of existing methods. Therefore, the predictive models might be not effective in some cases, due to: (1) the historical series of values is not sufficient usually, and (2) features from heterogeneous sources such as the intrinsic features of data samples themselves, which could be very useful, are not take into consideration. To address these issues, we proposed a novel method in this paper which learns the predictive model based on the combination of dynamic features extracted from series of historical values and static features of data samples. To evaluate the performance of our proposed method, we compare it with linear regression and boosted trees, and the experimental results validate our method's superiority.
{"title":"Combining heterogeneous features for time series prediction","authors":"Charles Chu, J. Brownlow, Qinxue Meng, Bin Fu, Ben Culbert, Min Zhu, Guandong Xu, Xue-zhong He","doi":"10.1109/BESC.2017.8256383","DOIUrl":"https://doi.org/10.1109/BESC.2017.8256383","url":null,"abstract":"Time series prediction is a challenging task in reality, and various methods have been proposed for it. However, only the historical series of values are exploited in most of existing methods. Therefore, the predictive models might be not effective in some cases, due to: (1) the historical series of values is not sufficient usually, and (2) features from heterogeneous sources such as the intrinsic features of data samples themselves, which could be very useful, are not take into consideration. To address these issues, we proposed a novel method in this paper which learns the predictive model based on the combination of dynamic features extracted from series of historical values and static features of data samples. To evaluate the performance of our proposed method, we compare it with linear regression and boosted trees, and the experimental results validate our method's superiority.","PeriodicalId":142098,"journal":{"name":"2017 International Conference on Behavioral, Economic, Socio-cultural Computing (BESC)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122803185","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}
Pub Date : 2017-07-01DOI: 10.1109/BESC.2017.8256362
S. Liu, Guandong Xu, Xiao Zhu, Zili Zhou
Life insurance application requires in-person meetings with underwriters, tedious paperwork, and an average waiting period of six weeks before an offer can be made. This outdated process has become a barrier for broader consumer adoption, resulting large coverage gap. In this work, we aim to closing this gap by leveraging data mining techniques to optimize the insurance questionnaire form. Our experiment on 10 years of insurance application data has identified that only ∼2% of all questions have shown high relevancy to determining the risks of applicants, resulting a significantly simplified questionnaire.
{"title":"Towards simplified insurance application via sparse questionnaire optimization","authors":"S. Liu, Guandong Xu, Xiao Zhu, Zili Zhou","doi":"10.1109/BESC.2017.8256362","DOIUrl":"https://doi.org/10.1109/BESC.2017.8256362","url":null,"abstract":"Life insurance application requires in-person meetings with underwriters, tedious paperwork, and an average waiting period of six weeks before an offer can be made. This outdated process has become a barrier for broader consumer adoption, resulting large coverage gap. In this work, we aim to closing this gap by leveraging data mining techniques to optimize the insurance questionnaire form. Our experiment on 10 years of insurance application data has identified that only ∼2% of all questions have shown high relevancy to determining the risks of applicants, resulting a significantly simplified questionnaire.","PeriodicalId":142098,"journal":{"name":"2017 International Conference on Behavioral, Economic, Socio-cultural Computing (BESC)","volume":"106 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115542803","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}
Pub Date : 2017-07-01DOI: 10.1109/BESC.2017.8256380
Zili Zhou, Guandong Xu, Xiao Zhu, S. Liu
User preference prediction aims to predict a users future preferences on a large number of items according to his/her preference history. To achieve this goal, many models have been proposed, but mainly for explicit preference data, such as 5-star ratings. Nevertheless, real-world data are often in implicit format, such as purchase action, and the number of items is not always large. In this paper, we demonstrate the use of latent factor models for solving the task of predicting user preferences on implicit and low-dimensional dataset.
{"title":"Latent factor analysis for low-dimensional implicit preference prediction","authors":"Zili Zhou, Guandong Xu, Xiao Zhu, S. Liu","doi":"10.1109/BESC.2017.8256380","DOIUrl":"https://doi.org/10.1109/BESC.2017.8256380","url":null,"abstract":"User preference prediction aims to predict a users future preferences on a large number of items according to his/her preference history. To achieve this goal, many models have been proposed, but mainly for explicit preference data, such as 5-star ratings. Nevertheless, real-world data are often in implicit format, such as purchase action, and the number of items is not always large. In this paper, we demonstrate the use of latent factor models for solving the task of predicting user preferences on implicit and low-dimensional dataset.","PeriodicalId":142098,"journal":{"name":"2017 International Conference on Behavioral, Economic, Socio-cultural Computing (BESC)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132626566","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}
Pub Date : 2017-07-01DOI: 10.1109/BESC.2017.8256405
Wenjie Zheng, Zenan Xu, Yanghui Rao, Haoran Xie, Fu Lee Wang, Reggie Kwan
Sentiment analysis has important applications in many areas, including marketing, recommendation, and financial analysis. Since topic modeling can discover hidden semantic structures, researchers put forward sentiment analysis models based on topic models. These models have been successfully applied on long texts, but analysis for short text is a challenging task because of the sparsity of features in short texts. We observe that the textual context has been widely considered on text analysis task, but on sentiment analysis area, most sentiment analysis models still lack of consideration and integration of sentimental context. Thus, by taking the speciality of sentiment analysis task and short text into consideration, we propose the sentimental context to enrich the characteristics and improve the performance of sentiment classification over short text. We first put forward the concept of sentimental context, which is extracted from the text body and sentiment lexicon, and then we integrate the sentimental context and propose two sentiment classification models based on word-level and topic-level respectively. We present results on real-world datasets from various sources, validating the effectiveness of the proposed models.
{"title":"Sentiment classification of short text using sentimental context","authors":"Wenjie Zheng, Zenan Xu, Yanghui Rao, Haoran Xie, Fu Lee Wang, Reggie Kwan","doi":"10.1109/BESC.2017.8256405","DOIUrl":"https://doi.org/10.1109/BESC.2017.8256405","url":null,"abstract":"Sentiment analysis has important applications in many areas, including marketing, recommendation, and financial analysis. Since topic modeling can discover hidden semantic structures, researchers put forward sentiment analysis models based on topic models. These models have been successfully applied on long texts, but analysis for short text is a challenging task because of the sparsity of features in short texts. We observe that the textual context has been widely considered on text analysis task, but on sentiment analysis area, most sentiment analysis models still lack of consideration and integration of sentimental context. Thus, by taking the speciality of sentiment analysis task and short text into consideration, we propose the sentimental context to enrich the characteristics and improve the performance of sentiment classification over short text. We first put forward the concept of sentimental context, which is extracted from the text body and sentiment lexicon, and then we integrate the sentimental context and propose two sentiment classification models based on word-level and topic-level respectively. We present results on real-world datasets from various sources, validating the effectiveness of the proposed models.","PeriodicalId":142098,"journal":{"name":"2017 International Conference on Behavioral, Economic, Socio-cultural Computing (BESC)","volume":"11 22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129989386","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}
Pub Date : 1900-01-01DOI: 10.1109/BESC.2017.8256385
Jas Semrl, Alexandru Matei
In the fitness industry, rolling gym membership contracts allow customers to terminate a contract with little advanced notice. Customer churn prediction is a well known area in Machine Learning research. Many companies, however, face a data science skills gap when trying to translate this research onto their own datasets and IT infrastructure. In this paper we present a series of experiments that aim to predict customer behaviour, in order to increase gym utilisation and customer retention. We use two off-the-shelf machine learning platforms, so that we can evaluate whether these platforms, used by non ML experts, can help companies improve their services.
{"title":"Churn prediction model for effective gym customer retention","authors":"Jas Semrl, Alexandru Matei","doi":"10.1109/BESC.2017.8256385","DOIUrl":"https://doi.org/10.1109/BESC.2017.8256385","url":null,"abstract":"In the fitness industry, rolling gym membership contracts allow customers to terminate a contract with little advanced notice. Customer churn prediction is a well known area in Machine Learning research. Many companies, however, face a data science skills gap when trying to translate this research onto their own datasets and IT infrastructure. In this paper we present a series of experiments that aim to predict customer behaviour, in order to increase gym utilisation and customer retention. We use two off-the-shelf machine learning platforms, so that we can evaluate whether these platforms, used by non ML experts, can help companies improve their services.","PeriodicalId":142098,"journal":{"name":"2017 International Conference on Behavioral, Economic, Socio-cultural Computing (BESC)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126325913","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：