Visible to the public Biblio

Filters: Author is Yu, Yang  [Clear All Filters]
2020-01-21
Yu, Yang, Hou, Jing, Li, Huan.  2019.  Study on Continuous Internal Audit System Modeling and Application. Proceedings of the 2019 International Conference on Artificial Intelligence and Advanced Manufacturing. :1–6.
Under the information environment the development of Continuous internal audit business model is inevitable and it will generally become the mainstream model. Based on the understanding of internal audit development in enterprises, it's found that most of current internal audit systems stay at post audit as an auxiliary tool of internal auditors in the auditing process, which hastens the application of continuous internal audit. Emerging computer technology is combined in this paper to build an universal continuous internal audit model, which is divided into four phases, based on internal audit system. Finally, based on the tracking error of index fund, this paper makes an applied research on the framework of the established continuous internal audit system.
2019-05-08
Zhang, Dongxue, Zheng, Yang, Wen, Yu, Xu, Yujue, Wang, Jingchuo, Yu, Yang, Meng, Dan.  2018.  Role-based Log Analysis Applying Deep Learning for Insider Threat Detection. Proceedings of the 1st Workshop on Security-Oriented Designs of Computer Architectures and Processors. :18–20.
Insider threats have shown their great destructive power in information security and financial stability and have received widespread attention from governments and organizations. Traditional intrusion detection systems fail to be effective in insider attacks due to the lack of extensive knowledge for insider behavior patterns. Instead, a more sophisticated method is required to have a deeper understanding for activities that insiders communicate with the information system. In this paper, we design a classifier, a neural network model utilizing Long Short Term Memory (LSTM) to model user log as a natural language sequence and achieve role-based classification. LSTM Model can learn behavior patterns of different users by automatically extracting feature and detect anomalies when log patterns deviate from the trained model. To illustrate the effective of classification model, we design two experiments based on cmu dataset. Experimental evaluations have shown that our model can successfully distinguish different behavior pattern and detect malicious behavior.