协同过滤算法中新项目推荐方法的研究

为了有效地解决协同过滤算法中新项目难以推荐的问题,文中提出了一种对项目矩阵进行划分的方法。其基本思想是,首先利用分类树算法划分项目矩阵并计算项目间的相似度,在此基础上缩小近邻搜索的范围和需要预测的资源数目。通过用户对已有项目的评分排列顺序和项目间相似性预测用户对新项目的评分。实验结果表明:基于项目矩阵划分的协同过滤算法有效地解决新项目推荐困难的问题,显示出了比传统推荐算法更好的推荐质量和扩展性。     

用磁声法MAE检测钢结构构件应力的机理和应用

分析了磁声发射的产生与接收、磁声发射与磁致伸缩、磁声发射的应力检测机理，应力对磁声发射的影响，磁声发射的检测系统等几个方面．从这些方面说明了用磁声发射来实施应力检测是可行的．证明磁声发射的强度既受外磁场变化影响，又受应力变化影响，并且磁声发射的变化幅度与所受应力大小有一定的关系，据此可为磁声发射检测铁磁材料的应力状态提供依据．但由于磁声发射的强度和施加于构件上应力状态的关系比较复杂，所以结合磁声发射和巴克素森BN两种方法各自优点的应力检测更比看好．     

Spectrum cartography techniques,challenges, opportunities,and applications: A survey

The spectrum cartography finds applications in several areas such as cognitive radios, spectrum aware communications, machine-type communications, Internet of Things, connected vehicles, wireless sensor networks, and radio frequency management systems, etc. This paper presents a survey on state-of-the-art of spectrum cartography techniques for the construction of various radio environment maps (REMs). Following a brief overview on spectrum cartography, various techniques considered to construct the REMs such as channel gain map, power spectral density map, power map, spectrum map, power propagation map, radio frequency map, and interference map are reviewed. In this paper, we compare the performance of the different spectrum cartography methods in terms of mean absolute error, mean square error, normalized mean square error, and root mean square error. The information presented in this paper aims to serve as a practical reference guide for various spectrum cartography methods for constructing different REMs. Finally, some of the open issues and challenges for future research and development are discussed.     

基于项目评分预测的协同过滤推荐算法

推荐系统是电子商务系统中最重要的技术之一.随着电子商务系统用户数目和商品数目的日益增加,在整个商品空间上用户评分数据极端稀疏,传统的相似性度量方法均存在各自的弊端,导致推荐系统的推荐质量急剧下降.针对用户评分数据极端稀疏情况下传统相似性度量方法的不足,提出了一种基于项目评分预测的协同过滤推荐算法,根据项目之间的相似性初步预测用户对未评分项目的评分,在此基础上,采用一种新颖的相似性度量方法计算目标用户的最近邻居.实验结果表明,该算法可以有效地解决用户评分数据极端稀疏情况下传统相似性度量方法存在的问题,显著地提高推荐系统的推荐质量.     

Extraction of polycyclic aromatic hydrocarbons from cookies: A comparative study of ultrasound and microwave-assisted procedures

The chromatographic determination of 15 polycyclic aromatic hydrocarbons (PAHs) in cookies has been improved in order to obtain a fast method with a low limit of detection through the combination of microwave-assisted extraction (MAE), oil saponification and solid-phase extraction clean-up before the injection of purified extracts in a C18 201TP52 (5 µm, 250  × 2.1 mm) column. Using acetonitrile-water as mobile phase, with a 50% to 95% w/w acetonitrile gradient for a fixed flow of 0.250 ml min^-1, 15 PAHs were separated in 45 min. The column temperature was maintained at 15°C; and fluorimetric detection was made at a fixed excitation wavelength of 264 nm and emission measurements at the best wavelength for each analyte, from 352 nm for 11H-benzo[b]fluorene to 500 nm for indeno[1,2,3-c,d]pyrene. Recoveries for all 15 PAHs varied between 96  ± 4 and 105%  ± 4%; and the limits of detection ranged from 0.015 ng g^-1 for chrysene to 0.7 ng g^-1 for phenantrene. Results were compared with those obtained by conventional Soxhlet extraction during 8-h refluxing with toluene, demonstrating that the methodology proposed is appropriate to quantify PAHs in cookies. Furthermore, the microwave-assisted method was faster and used less solvent than the conventional and ultrasound-assisted methods. The extraction time was reduced to 9 min compared with the 8 h required for Soxhlet extraction and 60 min required for ultrasound-assisted treatment, and the solvent consumption has been reduced to 25 ml compared with the 155 and 90 ml required using Soxhlet and ultrasound, respectively.     

小麦麸皮中总黄酮微波辅助提取技术研究

应用微波辅助提取麦麸中的黄酮类化合物.通过正交实验的方法得到从小麦麸皮中提取总黄酮的最佳工艺为:微波功率560 W,微波辐射时间30 s,乙醇体积分数为80%,料液比为1:20.在该工艺条件下总黄酮提取率为0.351 2%.     

基于加权相似度的用户协同过滤方法

协同过滤算法中用户相似性度量的准确性对推荐质量有显著影响。为了提高用户协同过滤算法中近邻选择的准确率,提出一种加权的皮尔逊相关系数（PCC）,可根据用户-项目的评分数,直接计算出PCC加权因子。将改进的皮尔逊相似度机制用于MovieLens,Douban和Epinions数据集进行实证分析。结果表明,提出的算法可以有效提高协同过滤推荐的平均绝对误差（MAE）和准确度。     

基于中值和多尺度的组合优化滤波器

考虑到一般机器人视觉导航滤波算法的通用性和有效性比较差的缺点,结合中值滤波器和多尺度自适应融合滤波器这两种算法的优点,提出一种基于中值和多尺度的组合优化滤波器。算法首先应用中值滤波,然后进行多尺度滤波处理,最后根据最小平均绝对误差准则,将滤波后结果进行融合处理。实验结果表明：该算法能很好地滤除机器人道路图像各种常见的噪声,从而提高算法的有效性和通用性。     

Comparative studies with Kollicoat MAE 30 D and Kollicoat MAE 30 DP in aqueous spray dispersions and enteric coatings on highly swellable caffeine cores

A film formulation containing Kollicoat MAE 30 D, Kollidon 30, Sicovit Rot 30, titanium dioxide, talc, and a plasticizer for the aqueous manufacture of enteric coatings was studied for the coagulations occurring with certain plasticizers and for differences in resistance on highly swellable caffeine cores. Also included in these investigations were the latices Kollicoat MAE 30 DP and Eudragit L 30 D-55. The coagulations occurring with all three film latices can be attributed to the presence of Kollidon 30 together with certain excipients. Preparations with Kollidon 30, but without color pigments, showed no tendency to coagulate. The advantage of propylene glycol (PG) compared to other plasticizers such as triethyl citrate (TEC) is that no coagulations occurred, even in the presence of Kollidon 30 and color pigments. Among the Kollidon 30-free film formulations examined, a plasticizer content of 10-15% PG or TEC gave the best results. Optimal pigment distribution in the coat originally produced by Kollidon 30 can optionally be achieved by prolonged stirring of the pigment suspension. The resistance can be further improved by inclusion of a subcoating with Kollidon VA 64. Kollicoat MAE 30 D and MAE 30 DP and Eudragit L 30 D-55 showed identical behavior in this study.     

基于时间加权的协同过滤算法研究

协同过滤算法是目前个性化推荐系统中应用最成功的推荐算法之一,但传统的算法没有考虑用户兴趣漂移的问题,导致推荐系统的推荐质量下降.针对这个问题,提出了基于时间加权的协同过滤算法.实验表明,改进的算法提高了推荐系统的推荐质量.