TXTang Xiaosheng

Tang Xiaosheng

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Tang Xiaosheng has a total of 12 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Student Academic Performance Prediction Using Deep Multi-source Behavior Sequential Network

    by Xiang Li, Xinning Zhu, Xiaoying Zhu, Yang Ji, Xiaosheng Tang

    Online education is becoming increasingly popular and often combined with traditional place-based study to improve learning efficiency for university students. Since students have left a large amount of online learning data, it provides an effective way to predict students’ academic performance and enable pre-intervention for at-risk students. Current data sources used to predict students’ performance are limited to data just from the corresponding learning platform, from which only learning behaviors on that course can be observed. However, students’ academic performance will be related to other behavioral factors, especially the patterns of using Internet. In this paper, we utilize two types of datasets from 505 university students, i.e., online learning records for a project-based course, and network logs of university campus network. A deep learning framework: Sequential Prediction based on Deep Network (SPDN) is proposed to predict students’ performance in the course. SPDN models students’ online behavioral sequences by utilizing multi-source fusion CNN technique, and incorporates static information based on bidirectional LSTM. Experiments demonstrate that the proposed SPDN model outperforms the baselines and has a significant improvement on early-warning. Furthermore, it can be learned that Internet access patterns even have a greater impact on students’ academic performance than online learning activities.


  • Frequent Subgraph Mining Based Collaboration Pattern Analysis for Wikipedia

    by Zhonghu Zuo, Chunhong Zhang, Xiaosheng Tang, Zheng Hu, Yuqian Tang

    Online knowledge collaborations, where distributed members without hierarchies self-organize themselvesto create valuable contents, are prevalent in many open production systems such as Wikipedia, GitHub andsocial networks. While many existing studies from network science have been brought to analyse the general interactivebehavioural patterns embedded in these systems, how the collaborations influence the achievement outcomes hasnot been thoroughly investigated. In this paper, we mine the collaboration patterns from a micro perspective to deeplyunderstand the relationships between the collaboration among participants and the qualities of theWikipedia articles.In particular, the subgraphs contained in the collaboration networks derived from theWikipedia revision histories aretaken as the fundamental units to analyse the collaboration diversities from the subgraph properties such as size andtopology. In contrast to the predefined static motifs adopted by the previous works, the collaboration subgraphs aredirectly found from Wikipedia dataset by a frequent subgraph mining algorithm GRAMI, which is able to capturethe real dynamic collaboration patterns. Moreover, the relationships between the co-authors in the subgraphs are alsodiscriminated to further explore the collaboration patterns. The experiments exhibit the statistical properties of thecollaboration subgraphs and the efficiency of them as the metrics for the article quality assessments. We concludethat a small group of editors with relative frequent fixed collaboration patterns contribute more to the excellent articlequality than the professional extents of arbitrary individuals in the collaboration group. This discovery confirms thecommonly insight about collaboration that many heads are always better than one and concretely suggests a potentialexplanation for the increasing prevalence and success of the online knowledge collaborations


  • A General Architecture in Support of Personalized, Interactive Multimedia Services in the Mobile Broadcast Convergent Environment

    by Wang Hui, Song Yali, Tang Xiaosheng, Zhang Ping

    The widespread use of 2G cellular systems and the availability of 3G cellular systems and mobile digital broadcast systems are facilitating the development of new personalized, interactive applications on mobile devices with limitations on computational resources and battery. However, as of today, most of these applications are unidirectional or target vehicle device as the terminal. This article describes a modular and general architecture that supports the development of interactive, multimedia, personalized applications, providing an extra level of service to the users. The performance evaluation of the architecture demonstrates that our prototype is friendly and easy to use, and saves power consumption.