ID 原文 译文
58238 工厂环境的电磁噪声会对以低功耗无线传感设备为核心的工业物联网技术产生巨大的影响. Factory-level electromagnetic noise has a huge impact on industrial Internet of things whichmainly uses low-power wireless sensor devices.
58239 为此，应用对数周期天线与频谱仪，从时域和频域 2 个角度在某汽车厂的焊接车间和办公区测量电磁噪声. For this reason，a log periodic antenna and a spectrumanalyzer are used to measure the electromagnetic noise in the welding workshop and the office area of acar factory.
58240 频域测量得到了0.3 ～ 3 GHz 频段工厂中电磁噪声的频点、功率等信息，噪声频谱呈现尖峰形状，可运用截断拉普拉斯分布进行建模分析. We obtain the frequency，power and other information of electromagnetic noise in 0. 3 ～3 GHz bands from the frequency domain measurement. The noise spectrum shows a spike shape，and thetruncated Laplace distribution can be used for modeling.
58241 时域测量采集了 315、779、916 MHz 几个频点的噪声数据，提取了噪声的幅度概率分布、脉冲持续时间分布、脉冲间隔时间分布 3 个参数. In time domain measurement，the noise data of315 MHz，779 MHz and 916 MHz are collected，and we also get the three parameters of amplitude probability distribution，pulse duration distribution and pulse separation distribution.
58242 测量结果表明，天线水平、垂直极化方式测得噪声信号相近; The measurement resultsindicate that the results of horizontal and vertical polarization are similar.
58243 办公区受噪声影响比焊接车间内部小. The influence of noise in theoffice area is less than that in the welding workshop.
58244 针对海量新闻数据给用户带来的困扰,为提升用户阅读新闻的个性化体验,提出了融合向量空间模型和 Bi鄄 secting K鄄means 聚类的新闻推荐方法. Personalized recommendation technology is a good solution to the problem of information overload. In order to improve the users personalized experience of reading news, a news recommendation method based on the vector space model and Bisecting Kmeans clustering is proposed.
58245 首先进行新闻文本向量化,使用向量空间模型和 TF鄄IDF 算法构建出新闻特 征向量; 采用 Bisecting K鄄means 聚类算法对新闻特征向量集进行聚类; Firstly, the news text vectorization is carried out: using the vector space model and TFIDF algorithm to construct news fea ture vectors; then Bisecting Kmeans clustering algorithm is utilized to cluster the news feature vector set;
58246 然后将已聚类的新闻集分为训练集和测试 集,根据训练集构建“用户—新闻类别—新闻冶三层层次结构的用户兴趣模型;最后采用余弦相似度方法得出新闻 推荐结果,并与测试集进行对比分析. after that, the clustered news set is divided into training set and test set, according to the training set, a “user news category news three level structure of the user interest model is built; finally, the cosine similarity method is used to calculate news recommendation results.
58247 实验以基于用户的协同过滤算法、基于物品的协同过滤算法、结合向量空间 模型和 K鄄means 聚类的推荐方法为基准,实验结果表明,该方法具有可行性,在准确率、召回率和 F 值上都有所提高. The experiments are based on user based collaborative filtering algorithm, item based collaborative filtering algorithm, combined vector space model and means clustering recommendation method, and the results show that the proposed method is feasible, and the accuracy rate, recall rate and F value all have been improved.