کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
4946848 | 1439557 | 2017 | 11 صفحه PDF | دانلود رایگان |
With the development of active distribution networks, data transmission is facing a severe security challenge. Secure data transmission is crucial for the real-time and exact control of active distribution networks. However, traditional data encryption methods have difficulty with the real-time control and mass data transmission of the active distribution networks. Additionally, content filtering based on text classification has a strong dependence on the size and type of data. To solve these problems, this paper proposes a novel distributed content filtering algorithm based on data labeling and policy expression (DCF-DLPE). In DCF-DLPE, we design a secure private protocol with data labeling and build a policy rule expression. Four representative datasets are used to evaluate the performance of the proposed algorithm. The comparative results show that for the larger dataset, DCF-DLPE outperforms the DES, AES (256-bit) and Blowfish encryption methods in the average time-consumption. Experimental results also show that compared with text classification algorithms, DCF-DLPE has a clear advantage in terms of filtering accuracy, sensitivity and precision. It is more important that, compared with text classification algorithms, performance of the DCF-DLPE algorithm is independent of the size and type of the dataset.
Journal: Neurocomputing - Volume 270, 27 December 2017, Pages 159-169