Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2020-09-28 Cooperative journals: 《计算机应用研究》
Abstract： Image classification techniques based on deep neural networks have achieved great success in recent years. However, recent studies have shown that deep neural network are vulnerable to the attack of adversarial examples. To solve this problem, some works train networks by adding Gaussian noise to the image. Thereby improving the ability of the network to defend adversarial examples, but the method does not consider the sensitivity of the network to different areas in the image when adding noise. To solve this problem, this paper proposed an adversarial training algorithm based on gradient guidance noise addition. When training the network, adding adaptive noise to different areas based on the sensitivity, adding large noise to the more sensitive areas, suppressing the sensitivity of the network to image changes, adding less noise to the less sensitive areas and improves the network classification accuracy. Compared with the existing algorithms on the cifar-10 dataset, the experimental results show that the proposed method effectively improved the accuracy of neural networks when classifying adversarial examples.
Subjects: Library Science，Information Science >> Information Science submitted time 2017-10-11 Cooperative journals: 《数据分析与知识发现》
Abstract： [Objective] This paper aims to identify sentiment propensity accurately with the help of a new method based on dependency parsing. [Methods] First, we extracted the sentiment stems of the sentences. Second, we defined sentiment-computing rules. Finally, we calculated sentiment propensity of each sentence. [Results] The proposed method achieved an overall accuracy of 84.46%. The average precision rate and recall rate for bullish class were 82.84% and 87.14% respectively, with an F-measure of 84.94%. In the mean time, bearish class got a precision rate of 86.28%, a recall rate of 81.74% and an F-measure of 83.95%. [Limitations] The proposed method did not consider the relevance among clauses. [Conclusions] The dependency parsing can effectively improve the accuracy of sentiment analysis of textual message from financial forum.
Hits
Hits
Downloads
Comment