Abstract Bag-of-features (BoF) is one of the most well-known methods used to represent digital image features because of its simplicity and efficiency. A variety of improved algorithms have been employed to enhance the performance of BoF in characterization. However, challenges in the application of BoF in the field still exist. This study focused on BoF by decomposing local features and presented... Bag-of-features (BoF) is one of the most well-known methods used to represent digital image features because of its simplicity and efficiency. A variety of improved algorithms have been employed to enhance the performance of BoF in characterization. However, challenges in the application of BoF in the field still exist. This study focused on BoF by decomposing local features and presented a novel framework for BoF on the basis of low-rank and sparse matrix decomposition to obtain a more robust and discriminative digital image classification. First, the local feature matrix of a digital image is decomposed into a low-rank matrix and a sparse matrix. Then, the BoF model was constructed in each part. Finally, the multiple kernel learning method was applied to combine the two models and the digital images were classified by using the support vector machine. Compared with existing methods in five public data sets, results show that the method proposed in this study is superior to the baseline algorithm and other coding algorithms by improving local features, with an improved classification performance of 17.68% in maximum and 0.01% in minimum. Compared with similar methods (such as leveraging the low-rank and sparse matrix decomposition and group sparse coding for image classification), this method is superior, with an improved classification performance of 2.76% in maximum and 0.08% in minimum, and obtains the highest average correct rate of classification. Therefore, the proposed method in this study is effective in improving the BoF in the feature extraction stage and has a better image classification performance. Read More Read Less

Abstract The involvement of a huge volume of man-made models and frames in 3D digital animations results in high coding costs using high efficiency video coding (HEVC). Studies identified the unique features of textures in 3D models that aids in predicting the depth of coding units (CU). The fast algorithm for CU partition, which was based on the texture features of... The involvement of a huge volume of man-made models and frames in 3D digital animations results in high coding costs using high efficiency video coding (HEVC). Studies identified the unique features of textures in 3D models that aids in predicting the depth of coding units (CU). The fast algorithm for CU partition, which was based on the texture features of the 3D models, was proposed to decrease coding complexity and shorten coding time. Canny operator was introduced to facilitate the segmentation of key frames in 3D animation. The optimal depth of the current CU was predicted based on its location and that of adjacent CUs, as well as the CU located at the current position of the reference frame. To reduce prediction errors in complex areas, rate-distortion optimization method was used to conduct precise partition. Compared with the performance of the original high efficiency video coding algorithm, the proposed algorithm shortened encoding time by 39.20% when the peak-signal-to-noise ratio of brightness was decreased by 0.07 dB and the encoding bit rate was increased by 0.84%. These results indicate the practical utility of the proposed algorithm and its excellent performance in improving coding time. Read More Read Less

Abstract As important channels of information transmission and knowledge flow, collaboration network and knowledge network can aid network actors in information update and knowledge creation. However, only a few comparative analyses have been conducted on knowledge flow in these two networks. To identify the knowledge flow in collaboration network and knowledge network, this study deployed a database of nanometer photoelectron... As important channels of information transmission and knowledge flow, collaboration network and knowledge network can aid network actors in information update and knowledge creation. However, only a few comparative analyses have been conducted on knowledge flow in these two networks. To identify the knowledge flow in collaboration network and knowledge network, this study deployed a database of nanometer photoelectron patents taken out by China at the State Intellectual Property Office over the period 2000 to 2015, and analyzed the evolution of these two networks in terms of network typology, node centrality and core-periphery structure. Results indicate that collaboration network and knowledge network are decoupling and developing in negative directions. Specifically, the collaboration networks of nanometer photoelectron patents decrease in scale, which hinders information transmission, whereas the knowledge networks increase in scale, which is beneficial to knowledge flow. In addition, the degree centrality and betweenness centrality of most nodes in collaboration network and knowledge network are positively related. Universities and ultra-micro techniques maintain core positions in the evolution of collaboration network and knowledge network and exert significant effects on knowledge flow. This study offers a viable method of knowledge flow identification based on the joint investigation of knowledge network and collaboration network and sheds light on efficiency enhancement of knowledge management within these two networks. Read More Read Less

Abstract Calculating similarity for multidimensional data is one of the key problems that must be addressed in order to promote the development of data clustering algorithms. In this study, we developed and tested a new similarity calculation index to improve the accuracy of multidimensional data clustering. First, the information divergence (ID) and generalized gradient angle (GGZ) were explored in detail.... Calculating similarity for multidimensional data is one of the key problems that must be addressed in order to promote the development of data clustering algorithms. In this study, we developed and tested a new similarity calculation index to improve the accuracy of multidimensional data clustering. First, the information divergence (ID) and generalized gradient angle (GGZ) were explored in detail. Second, the ID and GGZ were combined to calculate the similarity of multidimensional data, thus enabling a new algorithm for data clustering. Finally, two experiments were conducted to evaluate the performance of the proposed algorithm. The results of the experiments demonstrate that our proposed similarity calculation index for multidimensional data is both accurate and effective, providing better performance as measured by the metrics of accuracy (ACC), normalized mutual information (NMI), and purity (PUR). Based on this research, we conclude that the application of the proposed similarity calculation index is conducive to the improvement of data clustering for multidimensional data. Read More Read Less

Abstract In the article, possible applications of the transmission technology in the context of work of the contact center operator are reviewed. As a result, it becomes clear that the use of the streaming data transmission technology in the direct communication has long been present on the market in a variety of commercial products. In the article, a new scenario is... In the article, possible applications of the transmission technology in the context of work of the contact center operator are reviewed. As a result, it becomes clear that the use of the streaming data transmission technology in the direct communication has long been present on the market in a variety of commercial products. In the article, a new scenario is proposed for the use of streaming technology to provide context about the user. Further, on the basis of the analysis of typical calls of customers to the contact center, a conclusion is made that the use of the streaming data transmission technology allows improving the efficiency from 67% to 93% of all calls to the contact center, depending on the subject area of the contact center's work. Read More Read Less