Complex networks exist in a wide range of real world systems, such as social networks, technological networks, and biological networks. During the last decades, many researchers have concentrated on exploring some common things contained in those large networks include the small-world property, power-law degree distributions, and network connectivity. In this paper, we will investigate another important issue, community discovery, in network analysis. We choose Nonnegative Matrix Factorization (NMF) as our tool to find the communities because of its powerful interpretability and close relationship between clustering methods. Targeting different types of networks (undirected, directed and compound), we propose three NMF techniques (Symmetric NMF, Asymmetric NMF and Joint NMF). The Responsible editor: Eamonn Keogh. 123 494 F. Wang et al.correctness and convergence properties of those algorithms are also studied. Finally the experiments on real world networks are presented to show the effectiveness of the proposed methods.

Sexual health education is an essential part of quality-oriented education for college students. It aims to help these students to acquire knowledge of sexual physiology, sexual psychology, and sexual social norms that is consistent with the maturity of the students. Along with college students’attitudes toward sex, their perceptions regarding sexual behavior have also undergone profound changes. The importance of safe sexual behavior, sexual taboos, and sexual autonomy are gaining increasing attention as Chinese society is becoming more open. For college students who have just reached adulthood and have full autonomy of themselves, however, are they really going to have sexual behavior without careful consideration? Or is it something they have planned to do in the first place? To answer the above questions, this study was conducted to understand the relationship between college students’ attitudes toward sex, subjective norms, and behavioral control of their sexual behavior intentions by applying the Theory of Planned Behavior. In this study, 460 valid questionnaires were collected from Chinese college students and analyzed with partial least squares structural equation modeling (PLS-SEM). This study analyzes the relationship of multiple factors, including those influencing college students’ sexual behavior intentions. Meanwhile, it also compares the differences in factors affecting sexual behavior intentions between college students with or without sexual experience and those of different genders. Based on the results of the study, it was found that, first, subjective norms and perceived behavioral control of college students had a significant effect on safe sexual behavior intentions, while attitudes did not have a significant effect on safe sexual behavior intentions. Second, the gender and sexual experience of college students had a significant effect on safe sexual behavior intentions. Third, non-sexually experienced college students were more likely to be influenced by external factors. Relevant future research suggestions will be proposed based on the results of this study. Finally, this study helps to provide substantive suggestions for enhancing safe sexual behavior among college students in the context of universal higher education, as well as strengthening the self-protection of college students and providing practical advice for the development of sex education in China.

Efficiency is crucial to the online recommender systems, especially for the ones which needs to deal with tens of millions of users and items. Because representing users and items as binary vectors for Collaborative Filtering (CF) can achieve fast user-item affinity computation in the Hamming space, in recent years, we have witnessed an emerging research effort in exploiting binary hashing techniques for CF methods. However, CF with binary codes naturally suffers from low accuracy due to limited representation capability in each bit, which impedes it from modeling complex structure of the data.In this work, we attempt to improve the efficiency without hurting the model performance by utilizing both the accuracy of real-valued vectors and the efficiency of binary codes to represent users/items. In particular, we propose the Compositional Coding for Collaborative Filtering (CCCF) framework, which not only gains better recommendation efficiency than the state-of-theart binarized CF approaches but also achieves even higher accuracy than the real-valued CF method. Specifically, CCCF innovatively represents each user/item with a set of binary vectors, which are associated with a sparse real-value weight vector. Each value of the weight vector encodes the importance of the corresponding binary vector to the user/item. The continuous weight vectors greatly enhances the representation capability of binary codes, and its sparsity guarantees the processing speed. Furthermore, an integer weight approximation scheme is proposed to further accelerate the speed. Based on the CCCF framework, we design an efficient discrete optimization algorithm to learn its parameters. Extensive experiments on three real-world datasets show that our method outperforms the state-of-the-art binarized CF methods (even achieves better performance than the real-valued CF method) by a large margin in terms of both recommendation accuracy and efficiency. We publish our project at https://github.com/3140102441/CCCF.

The current study investigated what influences college students’ behavioral intention and behavior towards sports gambling using the theory of planned behavior (TPB) as a theoretical framework. The study also explored the moderation effect of problem gambling severity in the relationships between TPB determinants, behavioral intention, and sports gambling behavior. Data were collected from 334 college students from four different universities in the U.S. and analyzed through partial least squares structural equation modeling (PLS-SEM) and multi-group analysis. The results indicated that attitude was the most critical determinant of college students’ sports gambling intentions, followed by the subjective norms, while both behavioral intention and perceived behavioral control were significant predictors of sports gambling behavior. The study also found some meaningful moderation effects of problem gambling severity. Subjective norms were influential on college students with greater problem gambling severity, while attitude was the strongest predictor of recreational sports gamblers. Suggestions on prevention and treatment programs regarding sports gambling and problem gambling are discussed.

Anomalous urban mobility pattern refers to abnormal human mobility flow in a city. Anomalous urban mobility pattern detection is important in the study of urban mobility. In this paper, a framework is proposed to identify anomalous urban mobility patterns based on taxi GPS trajectories and Point of Interest (POI) data. In the framework, functional regions are first generated based on the distribution of POIs by the DBSCAN clustering algorithm. A Weighted Term Frequency-Inverse Document Frequency (WTF-IDF) method is proposed to identify function values in each region. Then, the Origin-Destination (OD) of trips between functional regions is extracted from GPS trajectories to detect anomalous urban mobility patterns. Mobility vectors are established for each time interval based on the OD of trips and are classified into clusters by the mean shift algorithm. Abnormal urban mobility patterns are identified by processing the mobility vectors. A case study in the city of Wuhan, China, is conducted; the experimental results show that the proposed method can effectively identify daily and hourly anomalous urban mobility patterns.