ID 原文 译文
18365 第二，如何在抑制噪声样本影响力的同时，保留正常样本的作用力。 The second is how to retain the effect of normal points while suppressing the impact of noise.
18366 针对这两问题，该文提出了基于自适应松弛的鲁棒模糊C均值聚类算法(AR-RFCM)。 To solve these two problems,Robust Fuzzy C-Means based on Adaptive Relaxation (AR-RFCM) is proposed.
18367 新模型基于K最近邻的方式(KNN)来估计样本的可靠性，自适应地调整松弛参数，从而实现在降低噪声样本影响力的同时，保留可靠样本的作用力。 The new model estimates thereliability of sample points by the method of the K-Nearest Neighbor (KNN). It adjusts adaptively therelaxation parameters to reduce the impact of noise, and keeps the effect of reliable sample points at the sametime.
18368 此外，AR-RFCM利用了C均值聚类模型中隶属度的稀疏性来提高可靠样本的作用力，从而提高数据簇的内聚程度，进而降低噪声样本的影响。 In addition, AR-RFCM utilizes the sparsity of membership in K-means to improve the effect of reliable sample points. Therefore, the compactness of clusters is improved and the impact of noise is suppressed.
18369 实验表明，AR-RFCM不仅在处理噪声样本时具有良好的鲁棒性，同时在25个UCI 数据集实验中，分类正确率(兰德指数)平均高于FCM算法7.7864%。 Experiments demonstrate that AR-RFCM has a good robustness for noise, and also achieves higher rand index in all 25 UCI data sets, even averagely higher than FCM 7.7864%.
18370 在加密邮件系统中，公钥可搜索加密技术可以有效地解决在不解密的情况下搜索加密邮件的问题。 In encrypted email system, the public key searchable encryption technology can effectively solve theproblem of searching for encrypted emails without decryption.
18371 针对公钥可搜索加密复杂的密钥管理问题，该文在加密邮件系统中引入了基于身份的密码体制。 In view of the complex key management problemof public key searchable encryption, an identity-based cryptosystem is introduced in the encrypted mail system.
18372 针对可搜索加密的离线关键字猜测攻击问题，该文采用了在加密关键字和生成陷门的同时进行认证，并且指定服务器去搜索加密电子邮件的方法。 For the offline keyword guessing attack problem of searchable encryption, the method of encrypting keywordsand generating trapdoors are adopted at the same time, and the server is designated to search for encryptedemails.
18373 同时，在随机预言机模型下，基于判定性双线性Diffie-Hellman假设，证明方案满足陷门和密文不可区分性安全。 At the same time, under the random oracle model, based on the decisional bilinear Diffie-Hellman assumption, the scheme is proved to satisfy the trapdoor and ciphertext indistinguishable security.
18374 数值实验结果表明，在陷门生成和关键字密文检测阶段，该方案与现有方案相比在计算效率上较高。 The numerical experiments show that the scheme has higher computational efficiency than the existing schemes in the keyword trapdoor generation and keyword ciphertext test phase.