{"title":"Dynamic Behavior Prediction via Heterogeneity and Community Feature Preserving Network Embedding","authors":"Shan Lu, Sisi Chen, Dehong Qiu","doi":"10.1109/ICEBE52470.2021.00012","DOIUrl":"https://doi.org/10.1109/ICEBE52470.2021.00012","url":null,"abstract":"Network embedding is a promising technique and has achieved impressive performance in predicting customer behavior in electronic commerce networks. Most current studies focus on the behavior prediction in homogeneous static networks and only use the feature of low order proximity within neighbors. However, electronic commerce networks are heterogeneous and change over time. The information of community and heterogeneity is also crucial for customer behavior prediction in dynamic networks. In this paper, we propose a novel framework termed as Dynamic Community aware Heterogeneous Network Embedding (DyCHNE), which combines the structural feature of community and heterogeneity instead of only the proximity feature of nodes. Specifically, the framework is consisted by two levels. The first-level enables each node to learn the dynamic feature of the heterogeneity from its neighbors in a continuous snapshot of the network; the second-level makes each node to learn the high order structural information of its community. We test our method on a real-world dataset which contains the customer-commodity relations in continuous network snapshots. Experimental result shows that our method outperforms the state-of-art methods.","PeriodicalId":136390,"journal":{"name":"2021 IEEE International Conference on e-Business Engineering (ICEBE)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122542219","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}
Hanchao Li, Xiang Fei, Ming Yang, K. Chao, Chaobo He
{"title":"From Music Information Retrieval to Stock Market Analysis: Theoretical Discussion on Feature Extraction Transfer","authors":"Hanchao Li, Xiang Fei, Ming Yang, K. Chao, Chaobo He","doi":"10.1109/ICEBE52470.2021.00027","DOIUrl":"https://doi.org/10.1109/ICEBE52470.2021.00027","url":null,"abstract":"Finding similar objects and patterns based on the similarity score is one of the fundamental and useful tasks in Data Mining. Different applications may introduce different features that need to be extract and analyzed. However, if two applications share some similar core concepts, it is possible to transfer some features that will be beneficial to transfer learning. Thus, this paper uses some features from Music Information Retrieval and Stock Market Analysis to theoretically illustrate the possibility of Feature Extraction Transfer. We use a 3-tuple or 6-tuple vector to record the music fundamental melody whereas a 5-tuple vector to record the daily behavior of the stock market from the candlestick chart. Hence, the flow of one music melody and the flow of one stock market can be treat as a time series vector sequence. Using this linkage, we have computed some feature exaction from Music Information Retrieval onto Stock Market Analysis and obtained some positive results. For example, the similarity between Activision Blizzard Inc and Zynga Inc have achieved a similarity score of 0.6250. Moreover, these positive results gave some ideas on implementing a self-supervised learning based system to manage your stock market and the potential of implementing a transfer learning between these two applications.","PeriodicalId":136390,"journal":{"name":"2021 IEEE International Conference on e-Business Engineering (ICEBE)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133062598","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}
{"title":"Improved Game Units Balancing In Game Design Through Combinatorial Optimization","authors":"Wei Wang, Ran Zhang","doi":"10.1109/ICEBE52470.2021.00022","DOIUrl":"https://doi.org/10.1109/ICEBE52470.2021.00022","url":null,"abstract":"Game balancing is an essential part of game design, and it plays a vital role as the balancing results directly affect the players' experiences. At present, there is no well-admitted definition of game balance. Generally, the game designers hold their own envisions of “balancing” in specific contextualized environments. Accordingly, it is a great challenge for game designers/developers to understand and establish. Since game balance is a multifaceted concept, there is no prescriptive guideline either. In this work, numerical balancing is selected to investigate the possibility of applying engineering optimization techniques to the game unit balancing in a quantitative manner. Currently, although the definite rules of game mechanics could be expressed mathematically, the game developers tune different attributes manually to evaluate the changes in an iterative way, which is tedious and time-consuming. Moreover, the balancing quality and efficiency strongly rely on the game developers' personal skills. To address this issue, we modeled the balancing process as a combinatorial optimization problem, in which the assessment metrics of game units are optimization objectives, and the attributes of game units are variables. The “balancing” of game units is interpreted as minimizing the standard deviations of the assessment metrics, and the evolutionary algorithm such as NSGA-II is used to find the optimization solutions. A typical case study is employed to demonstrate the proposed idea, and the optimization results show that the proposed method can simultaneously improve the numerical balancing in terms of quality and efficiency. At last, we summarize the contributions of the work and discuss how the proposed method can be further improved for game balancing in future work.","PeriodicalId":136390,"journal":{"name":"2021 IEEE International Conference on e-Business Engineering (ICEBE)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122340242","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}
{"title":"Detecting Fake Reviews via Verbal and Nonverbal Features","authors":"Guangxu Wang, Jinfeng Zhang, Xiong Zhang","doi":"10.1109/ICEBE52470.2021.00013","DOIUrl":"https://doi.org/10.1109/ICEBE52470.2021.00013","url":null,"abstract":"Online reviews are important to help customers to make purchasing decisions. However, opinion spamming exists in online review hosting websites, which has biased consumers' understanding of targeted products and service. Artificial Intelligence algorithms have been designed and adopted to automatically detect fake reviews. In this study, we evaluate the detection performance of various classic algorithms in a dataset crawled from one of largest online review hosting websites. We generate both verbal features from review text and nonverbal features from review related attributes. Experiment results show nonverbal features have higher impacts on fake review detection. Besides, GBDT could achieve stable detection performance in terms of four common metrics but using fewer features.","PeriodicalId":136390,"journal":{"name":"2021 IEEE International Conference on e-Business Engineering (ICEBE)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128975401","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}
Yanhui Su, P. Backlund, H. Engström, Mattias Strand
{"title":"Data-driven Method for Mobile Game Publishing Marketing Promotion","authors":"Yanhui Su, P. Backlund, H. Engström, Mattias Strand","doi":"10.1109/ICEBE52470.2021.00016","DOIUrl":"https://doi.org/10.1109/ICEBE52470.2021.00016","url":null,"abstract":"Marketing is vital for game businesses. However, it is hard to reach the target audiences and choose the right marketing strategies for different user acquisition channels in mobile game publishing. This paper shows how business intelligence can guide independent (indie) mobile game developers to do marketing promotion based on a data-driven method. We apply and extend the ARM funnel model from social game publishing to mobile game publishing and provide a data-driven method to evaluate the marketing promotion performance. First, in order to address indie game marketing challenges, we propose a data-driven method by defining key metrics and applying two different analysis methods for assessing mobile game marketing promotion. Second, we apply our data-driven method to an indie mobile game project and guide the indie game studio to determine the potential marketing promotion issues. Third, we also provide suggestions on improving marketing performance based on analysis results, especially by differentiating between channels. Finally, we summarize how our method can be generalized and used to guide indie mobile game marketing promotion, find high-quality channels for user acquisition, improve the marketing promotion effect, and discuss future research work.","PeriodicalId":136390,"journal":{"name":"2021 IEEE International Conference on e-Business Engineering (ICEBE)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133940392","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}
{"title":"Exploring A CAM - Based Approach for Weakly Supervised Fire Detection Task","authors":"Lvlong Lai, Jian Chen, Hui-juan Huang, Qingyao Wu","doi":"10.1109/ICEBE52470.2021.00035","DOIUrl":"https://doi.org/10.1109/ICEBE52470.2021.00035","url":null,"abstract":"Most existing works in fire detection literature use available detectors like Faster RCNN, SSD, YOLO, etc. to localize the fire in images. These approaches work well but require object-level annotation for training, which is created manually and is very expensive. In this paper, we explore the weakly supervised fire detection task (WSFD) in which only the image-level annotation is given. We propose an approach based on class activation map (CAM). The CAM-based approach firstly trains a deep neural network as the classifier for identifying fire and non-fire images. For a fire image in the inference stage, it uses the classifier to create a CAM and then further generates the bounding boxes according to the CAM. To evaluate the effectiveness of our approach, we collect and construct a benchmark dataset named WS-FireNet and conduct comprehensive experiments on it. The experiment results show that in a way the performance of our approach is satisfactory.","PeriodicalId":136390,"journal":{"name":"2021 IEEE International Conference on e-Business Engineering (ICEBE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130897561","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}
{"title":"A Vision-based Lane Departure Warning Framework","authors":"Jiaju Wu, Pengshuai Yin, Xin Shu, Huichou Huang, Fei Liu, Qingyao Wu","doi":"10.1109/ICEBE52470.2021.00029","DOIUrl":"https://doi.org/10.1109/ICEBE52470.2021.00029","url":null,"abstract":"Lane departure warning is an essential function of driver assistance systems. In this paper, we propose a vision-based lane departure warning framework, which can determine whether the vehicle deviates from the lane only with images as input. Our framework combines the deep learning method and traditional line detection algorithm. Specifically, the framework consists of a Lane Boundaries Localization module, a Lane Boundaries Generator module, and a Departure Warning module. The Lane Boundaries Localization module is a deep neural network for finding the area containing lane Boundaries. Then the Lane Boundaries Generator module performs backbone detection and line detection to generate lane boundaries on the located area. After identifying the lane boundaries, the Departure Warning module can quickly determine whether the vehicle deviates from the lane. Additionally, we implemented and deployed our framework to an Android platform and tested its running speed.","PeriodicalId":136390,"journal":{"name":"2021 IEEE International Conference on e-Business Engineering (ICEBE)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115902014","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}
{"title":"The Upward Development Strategy of Fruit Agricultural Products under the Background of E-commerce","authors":"Xinyu Zhang, Zhuo Dai","doi":"10.1109/ICEBE52470.2021.00024","DOIUrl":"https://doi.org/10.1109/ICEBE52470.2021.00024","url":null,"abstract":"This article discusses the development background and development status of fruit agricultural products under the background of e-commerce. The overall trend is stable and improving. The development dilemma of fruit agricultural products under the background of e-commerce includes the following problems. Firstly, the market supervision system of fruit agricultural products is not perfect. Second, the construction of storage, fresh-keeping, cold chain, and logistics facilities is not sound. Third, rural e-commerce professionals are still lacking. Development countermeasures include cultivating the standardization, differentiation, branding of fruit agricultural products, standardizing market management of agricultural products; improving rural logistics and other infrastructure; and actively training and introducing rural e-commerce talents. Finally, this paper explains the current diversified sales channels of fruit agricultural products on online platforms and novel publicity methods. Among them, “self-media bringing goods” has become a common trend in online agricultural product sales. In response to the national poverty alleviation policies, many e-commerce companies have assisted the sales of fruit agricultural products in various ways. Based on the current development status of fruit agricultural products e-commerce, this paper analyzes the main development dilemmas of the e-commerce of fruit agricultural products, and proposes development strategies for reference, in order to promote the more prosperous development of fruit agricultural products under the background of the e-commerce.","PeriodicalId":136390,"journal":{"name":"2021 IEEE International Conference on e-Business Engineering (ICEBE)","volume":"119 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116379658","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}
Yihong Jin, Guanshujie Fu, Li Qian, Hanwen Liu, Hongwei Wang
{"title":"Representation and Extraction of Diesel Engine Maintenance Knowledge Graph with Bidirectional Relations Based on BERT and the Bi-LSTM-CRF Model","authors":"Yihong Jin, Guanshujie Fu, Li Qian, Hanwen Liu, Hongwei Wang","doi":"10.1109/ICEBE52470.2021.00025","DOIUrl":"https://doi.org/10.1109/ICEBE52470.2021.00025","url":null,"abstract":"The effective and efficient maintenance of diesel engines in a power plant highly relies upon the knowledge accumulated in previous maintenance and overhaul practices. A large part of this knowledge exists in maintenance reports in the form of engineering know-what, know-how and know-why with complex elements and relationships. As such, the need of supporting automatic extractions of such knowledge using state-of-the-art machine learning and NLP techniques is raised. This paper proposes a framework for the presentation and extraction of the knowledge graph from unstructured text in maintenance reports. Different from previous work, this framework supports extractions of bidirectional relations through a novel combination of reports prepossessing, the BERT model and the Bi-LSTM-CRF model. Specifically, the BERT model is applied to get character vectors and the Bi-LSTM-CRF model is applied to realize automatic extractions of entities and relations. We use an application named Neo4j to construct and storage knowledge graphs, and an application named Protégé to construct bidirectional relations. All the results from users' search is ranked by a scoring formula developed by Yahya et al. Conducted experiments over both public and private data set show that the proposed framework achieves good performance and hence provide a promising solution.","PeriodicalId":136390,"journal":{"name":"2021 IEEE International Conference on e-Business Engineering (ICEBE)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133324159","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}
{"title":"Current and Future Direction of Social Media Marketing in SMEs: A Systematic Literature Review","authors":"Yuehan Chen, Atichart Harnchamchai, Teeraporn Saeheaw","doi":"10.1109/ICEBE52470.2021.00017","DOIUrl":"https://doi.org/10.1109/ICEBE52470.2021.00017","url":null,"abstract":"In the digital transformation, SMEs have gradually realized the benefits of social media as a marketing tool to enhance their competitiveness. Studies about SMEs' social media marketing are still in the initial stage and fragmented in their focus. This study conducts a systematic literature review for the social media marketing of SMEs. A total of sixteen articles were included after the selection with eligibility criteria from 179 literature published between 2016 and 2020 and synthesized to identify the current research about the social media marketing of SMEs, including the methods, theories, platforms of current research, and the motivations that drove the SMEs to adopt social media marketing. It was also found that the social media marketing with customer knowledge management and TikTok marketing of SMEs may be of potential for future research and practice.","PeriodicalId":136390,"journal":{"name":"2021 IEEE International Conference on e-Business Engineering (ICEBE)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123928997","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}