Raphael Monroe Marcondes, L. G. Vasconcelos, L. Baldochi
In order to fulfill the demands of Web shoppers, applications need to display the right content at the right time. However, due to the differences in preferences, needs and idiosyncrasies, these demands are not the same for different shoppers. In previous work, we built a toolkit to support the development of adaptive web applications. Our toolkit gathers and analyzes client logs in real time, allowing to understand the user behavior during navigation. By exploiting our toolkit, developers are able to build applications that adapt on-the-fly according to the user needs. This paper presents an extension to our previous work which allows adaptation code to be triggered automatically when interface problems are detected. Moreover, we provide adaptation templates that make the adaptation process straightforward.
{"title":"Introducing Adaptation Templates to Support the Implementation of Adaptive E-Commerce Applications","authors":"Raphael Monroe Marcondes, L. G. Vasconcelos, L. Baldochi","doi":"10.1109/ICEBE.2017.46","DOIUrl":"https://doi.org/10.1109/ICEBE.2017.46","url":null,"abstract":"In order to fulfill the demands of Web shoppers, applications need to display the right content at the right time. However, due to the differences in preferences, needs and idiosyncrasies, these demands are not the same for different shoppers. In previous work, we built a toolkit to support the development of adaptive web applications. Our toolkit gathers and analyzes client logs in real time, allowing to understand the user behavior during navigation. By exploiting our toolkit, developers are able to build applications that adapt on-the-fly according to the user needs. This paper presents an extension to our previous work which allows adaptation code to be triggered automatically when interface problems are detected. Moreover, we provide adaptation templates that make the adaptation process straightforward.","PeriodicalId":347774,"journal":{"name":"2017 IEEE 14th International Conference on e-Business Engineering (ICEBE)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123881344","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}
Z. Saberi, O. Hussain, Morteza Saberi, Elizabeth Chang
One of the main significant and challenging decisions for online retailers is assortment planning (AP). This decision become even more complex while considering demand and supply uncertainties in the AP planning. However, this lead to more efficient results in today's uncertain markets. Online retailers of late have access to massive amounts of internal and external data which they can leverage their power for tackling the inherent demand uncertainty and supplier uncertainty for assortment planning. This paper propose an AP framework for declaring how to use that data in different stage of decision making. Demand function in the framework is augmented using Google Trends (GT) and Google Correalte (GC) data which improve its accuracy. Using GT and GC increase the power of demand function extrapolability. Feature based modeling has been proposed to this end which allows us to use the GT data more easily. The final assortment decisions are then weighted against the supplier uncertainties to adjust for considering the variety and lead-time supplier effect. Techniques such as operations research methods and web science model will be utilised to develop the required approaches. While assortment planning is the combination of marketing and operations research techniques, in this work for the first time we incorporate web science techniques as the third edge of this important process.
{"title":"Online Retailer Assortment Planning and Managing under Customer and Supplier Uncertainty Effects Using Internal and External Data","authors":"Z. Saberi, O. Hussain, Morteza Saberi, Elizabeth Chang","doi":"10.1109/ICEBE.2017.12","DOIUrl":"https://doi.org/10.1109/ICEBE.2017.12","url":null,"abstract":"One of the main significant and challenging decisions for online retailers is assortment planning (AP). This decision become even more complex while considering demand and supply uncertainties in the AP planning. However, this lead to more efficient results in today's uncertain markets. Online retailers of late have access to massive amounts of internal and external data which they can leverage their power for tackling the inherent demand uncertainty and supplier uncertainty for assortment planning. This paper propose an AP framework for declaring how to use that data in different stage of decision making. Demand function in the framework is augmented using Google Trends (GT) and Google Correalte (GC) data which improve its accuracy. Using GT and GC increase the power of demand function extrapolability. Feature based modeling has been proposed to this end which allows us to use the GT data more easily. The final assortment decisions are then weighted against the supplier uncertainties to adjust for considering the variety and lead-time supplier effect. Techniques such as operations research methods and web science model will be utilised to develop the required approaches. While assortment planning is the combination of marketing and operations research techniques, in this work for the first time we incorporate web science techniques as the third edge of this important process.","PeriodicalId":347774,"journal":{"name":"2017 IEEE 14th International Conference on e-Business Engineering (ICEBE)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117263083","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}
Researches in the field of information propagation maximization consider that each content could have only one topic context and do not consider different propagation behaviors of people in different topics. In this paper, a new linear threshold diffusion model is presented which simulates the information diffusion among people with considering their different propagation behavior. Also a method is proposed for implementing the classic (basic) greedy algorithm to find the k-most influential users. The improved algorithm is called CBIG (content based improved greedy) algorithm which decreases the total amount of computations. Experimental results showed that the number of iterations of the algorithm were considerably decreased, while the precision of selecting most influential nodes is similar the base greedy algorithm.
{"title":"Finding K-Most Influential Users in Social Networks for Information Diffusion Based on Network Structure and Different User Behavioral Patterns","authors":"Maryam Shahsavari, S. Golpayegani","doi":"10.1109/ICEBE.2017.42","DOIUrl":"https://doi.org/10.1109/ICEBE.2017.42","url":null,"abstract":"Researches in the field of information propagation maximization consider that each content could have only one topic context and do not consider different propagation behaviors of people in different topics. In this paper, a new linear threshold diffusion model is presented which simulates the information diffusion among people with considering their different propagation behavior. Also a method is proposed for implementing the classic (basic) greedy algorithm to find the k-most influential users. The improved algorithm is called CBIG (content based improved greedy) algorithm which decreases the total amount of computations. Experimental results showed that the number of iterations of the algorithm were considerably decreased, while the precision of selecting most influential nodes is similar the base greedy algorithm.","PeriodicalId":347774,"journal":{"name":"2017 IEEE 14th International Conference on e-Business Engineering (ICEBE)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116143387","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}
Wang Dong, F. Asmi, Rongting Zhou, Fang Keren, M. A. Anwar
China, as a world's factory, is an eye-catching opportunity for the current global economic market. The trust in the service provider and the nature of service will be examined with reference to m-commerce in China among foreigners. The perceived trust in the service provider, perceived trust in the internet, influence of the society, perceived privacy and easiness while using accounted as factors affecting the foreigners perceived intentions to adopt m-commerce in China. The findings conclude the trust on internet is low and perceived privacy is high among foreigners. However, the intentions are still positive as trust in the m-commerce service provider is dominating the overall intentions of m-commerce users in P.R. China.
{"title":"Impact of Trust and Perceived Privacy in B2C Mobile Apps among Foreigners: A Case of People's Republic of China","authors":"Wang Dong, F. Asmi, Rongting Zhou, Fang Keren, M. A. Anwar","doi":"10.1109/ICEBE.2017.37","DOIUrl":"https://doi.org/10.1109/ICEBE.2017.37","url":null,"abstract":"China, as a world's factory, is an eye-catching opportunity for the current global economic market. The trust in the service provider and the nature of service will be examined with reference to m-commerce in China among foreigners. The perceived trust in the service provider, perceived trust in the internet, influence of the society, perceived privacy and easiness while using accounted as factors affecting the foreigners perceived intentions to adopt m-commerce in China. The findings conclude the trust on internet is low and perceived privacy is high among foreigners. However, the intentions are still positive as trust in the m-commerce service provider is dominating the overall intentions of m-commerce users in P.R. China.","PeriodicalId":347774,"journal":{"name":"2017 IEEE 14th International Conference on e-Business Engineering (ICEBE)","volume":"107 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131122221","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}
Virtual world has been developed rapidly in the last decade. Immersive is an important indicator of the virtual world. Moreover, with the VR technology advanced these years, the immersive become more and more important. The realistic experience is a crucial aspect of the immersive of the virtual world. We have identified two problems made the virtual world less immersive nowadays: Discontinuity of the Virtual World Change and Nonspontaneous of the Virtual World Change. To solve these two problems, we proposed a natural selection inspire Cell Based Evolutionary Virtual World (CEVW) framework which guides the change of the virtual world to be continuous and spontaneous.
{"title":"Cell Based Evolutionary Virtual World Framework","authors":"Bin Wang, J. Guo, Bingqing Shen","doi":"10.1109/ICEBE.2017.48","DOIUrl":"https://doi.org/10.1109/ICEBE.2017.48","url":null,"abstract":"Virtual world has been developed rapidly in the last decade. Immersive is an important indicator of the virtual world. Moreover, with the VR technology advanced these years, the immersive become more and more important. The realistic experience is a crucial aspect of the immersive of the virtual world. We have identified two problems made the virtual world less immersive nowadays: Discontinuity of the Virtual World Change and Nonspontaneous of the Virtual World Change. To solve these two problems, we proposed a natural selection inspire Cell Based Evolutionary Virtual World (CEVW) framework which guides the change of the virtual world to be continuous and spontaneous.","PeriodicalId":347774,"journal":{"name":"2017 IEEE 14th International Conference on e-Business Engineering (ICEBE)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131557079","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}
Investment recommendation plays a key role to online peer-to-peer lending, which help investors choose proper products. However, time value of money has not been considered, which is not in line with common sense. In this paper, we develop three models concerning about time value of money to help investor to maximize their benefits with fixed capital. Specifically, the first proposal model is Time Value of Money Prediction (TVM), aiming to calculate time value of money for each listing and taking into account of both effects of listing's financing time and flow risk. Then, investment profits of listings are assessed with Investment Profits Prediction (IP) model by predicting default risk. Finally, we apply Total Capital Value Maximization (TCVM) model to obtain the top-N recommendation by incorporating the round of time and data size. Experiment results carried on a real data collection reveal that the proposal models outperform traditional methods in terms of profits and contain effectiveness.
投资推荐在p2p网络借贷中起着关键作用，帮助投资者选择合适的产品。但是，没有考虑到货币的时间价值，这是不符合常识的。本文建立了三个关于货币时间价值的模型，以帮助投资者利用固定资本实现收益最大化。具体而言，第一个建议模型是货币时间价值预测(TVM)模型，旨在计算每个上市公司的货币时间价值，同时考虑上市公司融资时间和流量风险的影响。然后，通过预测违约风险，利用投资收益预测模型对上市公司的投资收益进行评估。最后，我们运用总资本价值最大化(Total Capital Value Maximization, TCVM)模型，结合回合时间和数据大小，得到top-N推荐。实际数据收集的实验结果表明，该建议模型在收益和有效性方面都优于传统方法。
{"title":"Investment Recommendation with Total Capital Value Maximization in Online P2P Lending","authors":"Yanchao Tan, Xiaolin Zheng, Mengying Zhu, Chaohui Wang, Z. Zhu, Lifeng Yu","doi":"10.1109/ICEBE.2017.32","DOIUrl":"https://doi.org/10.1109/ICEBE.2017.32","url":null,"abstract":"Investment recommendation plays a key role to online peer-to-peer lending, which help investors choose proper products. However, time value of money has not been considered, which is not in line with common sense. In this paper, we develop three models concerning about time value of money to help investor to maximize their benefits with fixed capital. Specifically, the first proposal model is Time Value of Money Prediction (TVM), aiming to calculate time value of money for each listing and taking into account of both effects of listing's financing time and flow risk. Then, investment profits of listings are assessed with Investment Profits Prediction (IP) model by predicting default risk. Finally, we apply Total Capital Value Maximization (TCVM) model to obtain the top-N recommendation by incorporating the round of time and data size. Experiment results carried on a real data collection reveal that the proposal models outperform traditional methods in terms of profits and contain effectiveness.","PeriodicalId":347774,"journal":{"name":"2017 IEEE 14th International Conference on e-Business Engineering (ICEBE)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130466658","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 this paper, we propose a framework to implement a self-adaptive goal-driven service composition based on context and QoS approach. According to our approach, every concrete composition is regarded as an instantiation of the goal abastract graph, which is regarded as the abstract composition. By means of incorporating, a feedback system, our approach can handle composition requests failures using the transformation of the failed composition requests into alternative ones that can be solved. Moreover, our approach integrate a self-adaptive mechanisms to adapt the service composition to the distributed, open and dynamic nature of services and the changing user's environment. Furthermore, we present self-adaptive mechanisms monitoring the service composition execution. At last, the result of a case study showed our approach effectiveness.
{"title":"Self-Adaptive Goal-Driven Web Service Composition Based on Context and QoS","authors":"Emna Khanfir, Raoudha Ben Djemaa, Ikram Amous","doi":"10.1109/ICEBE.2017.39","DOIUrl":"https://doi.org/10.1109/ICEBE.2017.39","url":null,"abstract":"In this paper, we propose a framework to implement a self-adaptive goal-driven service composition based on context and QoS approach. According to our approach, every concrete composition is regarded as an instantiation of the goal abastract graph, which is regarded as the abstract composition. By means of incorporating, a feedback system, our approach can handle composition requests failures using the transformation of the failed composition requests into alternative ones that can be solved. Moreover, our approach integrate a self-adaptive mechanisms to adapt the service composition to the distributed, open and dynamic nature of services and the changing user's environment. Furthermore, we present self-adaptive mechanisms monitoring the service composition execution. At last, the result of a case study showed our approach effectiveness.","PeriodicalId":347774,"journal":{"name":"2017 IEEE 14th International Conference on e-Business Engineering (ICEBE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130783750","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}
Online bike-sharing market has taken off in China since 2016 and in the competition two companies stand out and occupy most of the market share. This paper provides an analysis of rental pricing mechanism for this two companies in this e-market. Firstly, the thesis builds the spontaneous evolutionary game model of the pricing for online bicycle renting and obtains the evolutionary stable strategy. Then the authors propose optimization schemes and pricing mechanism from these leasing service suppliers' perspective. On this basis above, research on how to avoid malicious competition and create a healthy market will be discussed.
{"title":"Pricing Mechanism Research in Duopoly Online Bicycle-Sharing Market: A Game Theory Approach","authors":"Chao Yu, Y. Chai, Yi Liu","doi":"10.1109/ICEBE.2017.11","DOIUrl":"https://doi.org/10.1109/ICEBE.2017.11","url":null,"abstract":"Online bike-sharing market has taken off in China since 2016 and in the competition two companies stand out and occupy most of the market share. This paper provides an analysis of rental pricing mechanism for this two companies in this e-market. Firstly, the thesis builds the spontaneous evolutionary game model of the pricing for online bicycle renting and obtains the evolutionary stable strategy. Then the authors propose optimization schemes and pricing mechanism from these leasing service suppliers' perspective. On this basis above, research on how to avoid malicious competition and create a healthy market will be discussed.","PeriodicalId":347774,"journal":{"name":"2017 IEEE 14th International Conference on e-Business Engineering (ICEBE)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132493050","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 advantages including the high interaction, convenience, transparency and individualization, the Internet makes the online shopping become increasingly popular among businesses and consumers. The online transaction environment is characterized by virtuality and anonymity, so the trust plays an indispensable role. Currently the trust issue has become one of the main reasons that troubles and hinders the online shopping behavior of consumers. In this paper, the trusted third party (TTP) in Australia's business is studied and the factors influencing consumers' trust behavior are examined from the perspective of consumers online trust of online shopping. Based on the literature review and combined with the development status and background of Australia's e-commerce, underpinned by the Theory of Planned Behavior, the Institutional Theory and Signaling Theory, an integrative service framework of TTP is proposed and empirically tested in this paper. Partial Least Squares Structural Equation Modelling was conducted to assess the hypotheses. This study proposes that an integrative service framework of TTP should include the following five aspects: payment service, logistics & distribution service, guarantee & insurance service, certification service, and rating & recommendation service. The empirical results, obtained in a sample of 282 online shopping users, indicate that the above-mentioned five different types of TTPs, which have combined effect on the trust motivation and behavior of consumers. Through building an integrative service framework of TTP, the level of consumer trust can be more fully strengthened. When consumers think that more services from TTP are owned by an enterprise, they have smaller barrier in their expected future trust and stronger control of their own trust behavior, and finally enhance the consumers' trust behavior.
互联网以其交互性强、方便、透明、个性化等优点，使得网上购物越来越受到商家和消费者的欢迎。网络交易环境具有虚拟性和匿名性的特点，信任在其中起着不可或缺的作用。目前，信任问题已经成为困扰和阻碍消费者网购行为的主要原因之一。本文以澳大利亚商业中的可信第三方(trusted third party, TTP)为研究对象，从消费者网络购物信任的角度考察消费者信任行为的影响因素。本文在文献综述的基础上，结合澳大利亚电子商务的发展现状和背景，以计划行为理论、制度理论和信号理论为基础，提出了TTP的一体化服务框架，并进行了实证检验。采用偏最小二乘结构方程模型对假设进行评估。本研究提出TTP的综合服务框架应包括以下五个方面:支付服务、物流和amp;配送服务、保障;保险服务、认证服务、评级服务;推荐服务。在282名网购用户的样本中获得的实证结果表明，上述五种不同类型的TTPs对消费者的信任动机和行为有综合影响。通过构建TTP的一体化服务框架，可以更充分地增强消费者的信任水平。当消费者认为企业拥有更多的TTP服务时，他们对未来信任的预期障碍更小，对自身信任行为的控制力更强，最终增强消费者的信任行为。
{"title":"The Effects of Consumer Perceived Different Service of Trusted Third Party on Trust Intention: An Empirical Study in Australia","authors":"Cong Cao, Jun Yan, Mengxiang Li","doi":"10.1109/ICEBE.2017.19","DOIUrl":"https://doi.org/10.1109/ICEBE.2017.19","url":null,"abstract":"With the advantages including the high interaction, convenience, transparency and individualization, the Internet makes the online shopping become increasingly popular among businesses and consumers. The online transaction environment is characterized by virtuality and anonymity, so the trust plays an indispensable role. Currently the trust issue has become one of the main reasons that troubles and hinders the online shopping behavior of consumers. In this paper, the trusted third party (TTP) in Australia's business is studied and the factors influencing consumers' trust behavior are examined from the perspective of consumers online trust of online shopping. Based on the literature review and combined with the development status and background of Australia's e-commerce, underpinned by the Theory of Planned Behavior, the Institutional Theory and Signaling Theory, an integrative service framework of TTP is proposed and empirically tested in this paper. Partial Least Squares Structural Equation Modelling was conducted to assess the hypotheses. This study proposes that an integrative service framework of TTP should include the following five aspects: payment service, logistics & distribution service, guarantee & insurance service, certification service, and rating & recommendation service. The empirical results, obtained in a sample of 282 online shopping users, indicate that the above-mentioned five different types of TTPs, which have combined effect on the trust motivation and behavior of consumers. Through building an integrative service framework of TTP, the level of consumer trust can be more fully strengthened. When consumers think that more services from TTP are owned by an enterprise, they have smaller barrier in their expected future trust and stronger control of their own trust behavior, and finally enhance the consumers' trust behavior.","PeriodicalId":347774,"journal":{"name":"2017 IEEE 14th International Conference on e-Business Engineering (ICEBE)","volume":"177 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134292654","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 wide adoption of social networks, people are accustomed to post their ideas and thinking via these platforms. Tweets or comments online usually come with individual sentiment, which are time consuming to be analyzed by human labor. This study encapsulates a prototype Chinese sentiment mining system and takes a global hotel reviewing website TripAdvisor as the evaluation sample. The proposed sentiment mining model is compared with logistic regression and support vector machine models based on their performances. This proposed model outperforms LR and SVM in all datasets in terms of classification accuracy and F-measure. An additional module embedded in proposed system enables expansion of novel or undefined terms to the dictionary referred (NTUSD). With this Word2Vec-based module, the system further improves accuracy while reduces both type I and type II error for at least 5%.
{"title":"A Rule-Based Chinese Sentiment Mining System with Self-Expanding Dictionary - Taking TripAdvisor as an Example","authors":"Jung-Bin Li, Li-Bing Yang","doi":"10.1109/ICEBE.2017.45","DOIUrl":"https://doi.org/10.1109/ICEBE.2017.45","url":null,"abstract":"With the wide adoption of social networks, people are accustomed to post their ideas and thinking via these platforms. Tweets or comments online usually come with individual sentiment, which are time consuming to be analyzed by human labor. This study encapsulates a prototype Chinese sentiment mining system and takes a global hotel reviewing website TripAdvisor as the evaluation sample. The proposed sentiment mining model is compared with logistic regression and support vector machine models based on their performances. This proposed model outperforms LR and SVM in all datasets in terms of classification accuracy and F-measure. An additional module embedded in proposed system enables expansion of novel or undefined terms to the dictionary referred (NTUSD). With this Word2Vec-based module, the system further improves accuracy while reduces both type I and type II error for at least 5%.","PeriodicalId":347774,"journal":{"name":"2017 IEEE 14th International Conference on e-Business Engineering (ICEBE)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129074322","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：