ID 原文 译文
56668 且MHC-NSA的时间复杂度由传统NSA算法的指数阶降低到多项式阶.在UCI数据集Balance Scale上的实验表明, MHC-NSA的检测器训练时间较典型NSA算法代表RNSA, V-Detector和BIORV-NSA在分别降低53.73%, 96. 43%和92. 66%的同时,检测率分别提升69.57%, 44%和17. 54%. Moreover, relative totraditional NSAs, the time complexity of MHC-NSA decreases from the exponential level to the polynomial level. Experiments on the UCI dataset show that compared with RNSA, V-Detector, and BIORV-NSA, the trainingtime of MHC-NSA is reduced by 53. 73%, 96. 43%, and 92. 66% and the detection rate of MHC-NSA increases by69. 57%, 44%, and 17. 54%, respectively.
56669 作为推荐系统的重要组成部分,协同过滤已成为了当今主流的推荐方法之一. Collaborative filtering, an important part of the recommendation system, has become a mainstreamrecommendation method.
56670 其中基于潜在因子的协同过滤常采用SVD推荐模型分析用户喜好. In collaborative filtering methods based on potential factors, SVD recommendationmodels are often used to analyze user preferences.
56671 近年来,随着SVD推荐模型研究的深入, SVD++,TrustSVD等一类带有隐式反馈的SVD推荐模型被相继提出. 此类模型能更有效地从有限的数据源中挖掘有用信息并取得了较好的效果,因此受到了人们广泛关注. With the recent research of SVD recommendation models,some SVD recommendation models with implicit feedback, such as SVD++ and TrustSVD, have been successivelyproposed. These types of models can more effectively mine useful information from limited data sources andachieve better results than traditional SVD recommendation model, thereby garnering widespread attention.
56672 然而,现有文献大多关注于模型设计,缺乏专门针对带有隐式反馈的SVD推荐模型的高效求解算法. 为此,本文首先研究了一般性的SVD推荐模型梯度求解框架,然后以SVD++推荐模型为突破口,基于块梯度下降方法设计了高效求解算法BCDSVD++并解决了容量矩阵求逆、稀疏数据优化处理等两个关键问题. 实验表明,本文所设计的BCDSVD++算法具有比传统的并行梯度下降法更高效的求解效率及收敛能力. However, most existing papers focus on model design and the lack of efficient algorithms for SVD recommendationmodels with implicit feedback. Therefore, this paper first studies the general gradient solution framework of theSVD recommendation model. Then, it considers the SVD++ recommendation model as a breakthrough anddesigns an efficient solution algorithm, namely, BCDSVD++, based on the block coordinate descent method. Furthermore, we solve the two key problems of capacity matrix inversion and sparse data optimization processing. Experiments show that the proposed BCDSVD++ algorithm yields better solution efficiency and convergenceability than the traditional parallel gradient descent method.
56673 液体运载火箭作为卫星运载的重要平台,对其结构安全性进行准确的评估极其重要. Liquid launch rockets are used as one of the most important platforms for carrying satellites. Thus,its structural safety must be accurately assessed.
56674 本文针对液体运载火箭实际工作过程中监测数据易受到噪声干扰的问题,基于置信规则库(belief rule base,BRB)提出了一种考虑环境干扰的液体运载火箭结构安全性评估方法. To address the environmental disturbance for observation datain engineering practice, a new safety assessment method is proposed in this paper with belief rule base (BRB) andenvironmental disturbance consideration.
56675 首先,通过分析火箭监测数据的特征,提出了一种基于距离的监测指标不确定度求取方法,通过指标不确定度来反映噪声对于监测数据的影响程度; First, the characteristic analysis is conducted for the rocket, and anuncertainty calculation method based on distance for observation index is provided. The uncertainty index is usedto present the disturbance degree of observation data from environmental noise.
56676 然后,针对目前置信规则库模型中仅考虑了定性知识的不确定,无法处理监测数据不确定的问题,提出了一种考虑指标不确定度的输入数据匹配度转换方法,监测数据中包含不确定性信息的大小通过剩余匹配度表示; Second, only the uncertaintyof qualitative knowledge is considered in current studies of the BRB model, and the uncertainty in quantitativedata cannot be addressed. Thus, a new calculation method for matching degrees of input data is developed,where the uncertainty of the observation data can be represented using the residual matching degree.
56677 最后,基于BRB构建了液体运载火箭安全性评估模型,对存在不确定性的监测数据进行融合,得到液体运载火箭的安全性状态. Finally,to integrate the uncertainty of the observation data, a safety assessment model for the liquid launch rocket isdeveloped based on the BRB model.