Collaborative Filtering Recommendation Algorithm Based on Multi-Relationship Social Network

Recommendation system is one of the most common applications in the field of big data. The traditional collaborative filtering recommendation algorithm is directly based on user-item rating matrix. However, when there are huge amounts of user and commodities data, the efficiency of the algorithm will be significantly reduced. Aiming at the problem, a collaborative filtering recommendation algorithm based on multi-relational social networks is proposed. The algorithm divides the multi-relational social networks based on the multi-subnet complex network model into communities by using information dissemination method, which divides the users with similar degree into a community. Then the user-item rating matrix is constructed by choosing the k-nearest neighbor set of users within the community, in this case, the collaborative filtering algorithm is used for recommendation. Thus, the execution efficiency of the algorithm is improved without reducing the accuracy of recommendation.     

设计师与用户对产品意象的基因网络对比

目的研究设计师与用户对产品意象认知方面的差异性,挖掘用户的隐性需求。方法分别以设计竞赛作品与市场畅销产品作为主体对产品感知意象的映射载体,提出了基于同构与异构两种编码原则的产品基因网络模型。通过绘制同构基因网络图谱与异构基因网络频谱,分别对比分析两类基因节点的重要性与关联性。以保温杯为实例对模型的可行性进行验证。结论通过对同类产品在竞赛与市场环境下表现出的不同信息所构成的基因网络进行对比研究,可以降低设计师与用户对产品意象的认知差异,辅助设计师从宏观角度用联系的观点把握用户需求,从而提高设计方案的市场接受度。     

C-to-C电子商务的四流合一管理模式探讨

通过对200个C-to-C交易的使用者的调查,对现行C-to-C电子商务中信息流、物流和资金流这三个基本流运行中存在问题的分析,提出了如何运用网络平台更有效地整合三者的设想,并提出为使C-to-C电子商务健康发展,政府的管理流也应是C-to-C电子商务运行中不可缺少的组成成分,提出了用市场模式管理买卖双方的设想和四流合一的C-to-C电子商务管理模式.     

Semi-GSGCN: Social Robot Detection Research with Graph Neural Network

Malicious social robots are the disseminators of malicious information on social networks, which seriously affect information security and network environments. Efficient and reliable classification of social robots is crucial for detecting information manipulation in social networks. Supervised classification based on manual feature extraction has been widely used in social robot detection. However, these methods not only involve the privacy of users but also ignore hidden feature information, especially the graph feature, and the label utilization rate of semi-supervised algorithms is low. Aiming at the problems of shallow feature extraction and low label utilization rate in existing social network robot detection methods, in this paper a robot detection scheme based on weighted network topology is proposed, which introduces an improved network representation learning algorithm to extract the local structure features of the network, and combined with the graph convolution network (GCN) algorithm based on the graph filter, to obtain the global structure features of the network. An end-to-end semi-supervised combination model (Semi-GSGCN) is established to detect malicious social robots. Experiments on a social network dataset (cresci-rtbust-2019) show that the proposed method has high versatility and effectiveness in detecting social robots. In addition, this method has a stronger insight into robots in social networks than other methods.     

Computing the User Experience via Big Data Analysis: A Case of Uber Services

As of 2020, the issue of user satisfaction has generated a significant amount of interest. Therefore, we employ a big data approach for exploring user satisfaction among Uber users. We develop a research model of user satisfaction by expanding the list of user experience (UX) elements (i.e., pragmatic, expectation confirmation, hedonic, and burden) by including more elements, namely: risk, cost, promotion, anxiety, sadness, and anger. Subsequently, we collect 125,768 comments from online reviews of Uber services and perform a sentiment analysis to extract the UX elements. The results of a regression analysis reveal the following: hedonic, promotion, and pragmatic significantly and positively affect user satisfaction, while burden, cost, and risk have a substantial negative influence. However, the influence of expectation confirmation on user satisfaction is not supported. Moreover, sadness, anxiety, and anger are positively related to the perceived risk of users. Compared with sadness and anxiety, anger has a more important role in increasing the perceived burden of users. Based on these findings, we also provide some theoretical implications for future UX literature and some core suggestions related to establishing strategies for Uber and similar services. The proposed big data approach may be utilized in other UX studies in the future.     

An Attention-Based Friend Recommendation Model in Social Network

In social networks, user attention affects the user’s decision-making, resulting in a performance alteration of the recommendation systems. Existing systems make recommendations mainly according to users’ preferences with a particular focus on items. However, the significance of users’ attention and the difference in the influence of different users and items are often ignored. Thus, this paper proposes an attention-based multi-layer friend recommendation model to mitigate information overload in social networks. We first constructed the basic user and item matrix via convolutional neural networks (CNN). Then, we obtained user preferences by using the relationships between users and items, which were later inputted into our model to learn the preferences between friends. The error performance of the proposed method was compared with the traditional solutions based on collaborative filtering. A comprehensive performance evaluation was also conducted using large-scale real-world datasets collected from three popular location-based social networks. The experimental results revealed that our proposal outperforms the traditional methods in terms of recommendation performance.     

Network model of knowledge diffusion

This paper introduces a diffusion network model: an individual-citation-based directed network model with a time dimension, as a potentially useful approach to capture the diffusion of research topics. The approach combines social network analysis, network visualization and citation analysis to discuss some of the issues concerning the spread of scientific ideas. The process of knowledge diffusion is traced from a network point of view. Using research on the h-index as a case study, we built detailed networks of individual publications and demonstrated the feasibility of applying the diffusion network model to the spread of a research. The model shows the specific paths and associations of individual papers, and potentially complementing issues raised by epidemic models, which primarily deal with average properties of entire scientific communities. Also, based on the citation-based network, the technique of main path analysis identified the articles that influenced the research for some time and linked them into a research tradition that is the backbone of the h-index field.     

基于用户兴趣本体的数字图书馆个性化检索研究

为满足数字图书馆信息检索中用户个性化需求,本文描述了一个基于用户兴趣本体的个性化检索模型．该模型将用户兴趣本体应用于图书馆检索,包括用户兴趣模型的构建、用户兴趣查询扩展以及个性化检索结果排序等。在用户访问数字图书馆过程中,利用用户兴趣本体来匹配和扩展用户关键词,实现语义化的检索,满足用户个性化需求。     

Adaptive bandwidth sharing mechanism for quality of service administration in infrastructure wireless networks

In infrastructure wireless networks, the wireless hop can be considered as another hop of the transmission path. With the rapid growth of wireless traffics, the future wireless network is expected to provide services for heterogeneous data traffics with different quality of service (QoS) requirements. Most proposed schemes do not have adaptive mechanisms to deal with the environment changes. In real situation, bandwidths, error rates and loss rates of wireless links vary frequently. We will base on the differentiated service model and propose a wireless differentiation (WD) scheme for user datagram protocol (UDP) flows and a wireless differentiation with prioritised ACK scheme for connections with transmission control protocol (TCP) flows. Both schemes provide QoS support for IEEE 802.11b and do not change the basic access mechanism of IEEE 802.11b.     

An analytical solution based on mobility and multicriteria optimization for access selection in heterogeneous environment

Providing global connectivity with high speed and guaranteed quality at any place and any time is now becoming a reality due to the integration and co-ordination of different radio access technologies. The internetworking of existing networks with diverse characteristics has been considered attractive to meet the incredible development of interactive multimedia services and ever-growing demands of mobile users. Due to the diverse characteristics of heterogeneous networks, several challenges have to be addressed in terms of quality of service (QoS), mobility management and user preferences. To achieve this goal, an optimal network selection algorithm is needed to select the target network for maximizing the end user satisfaction. The existing works do not consider the integration of utility function with mobile terminal mobility characteristics to minimize ping-pong effects in the integrated networks. An integrated multicriteria network selection algorithm based on multiplicative utility function and residual residence time (RRT) estimation is proposed to keep the mobile users always best connected. Multiplicative weighted utility function considers network conditions, application QoS and user preferences to evaluate the available networks. In this paper, the proposed scheme is implemented with two mainstreams (pedestrian users and high-velocity users). For high-velocity users, RRT and adaptive residence time threshold are also considered to keep the probability of handover failures and unnecessary handovers within the limits. Monte-Carlo simulation results demonstrate that the proposed scheme outperforms against existing approaches.     

Structure and pattern of social tags for keyword selection behaviors

This article identifies patterns and structures in the social tagging of scholarly articles in CiteULike. Using a dataset of 4,215 tags attributed to 1,600 scholarly articles from 15 library and information science journals, a network was built to understand users?? information organization behavior. Social network analysis and the frequent-pattern tree method were used to discover the implicit patterns and structures embedded in social tags as well as in their use, based on 26 proposed tag categories. The pattern and structure of this network of social tags is characterized by power-law distribution, centrality, co-used tag categories, role sharing among tag categories, and similar roles of tag categories in associating distinct tag categories. Furthermore, researchers generated 21 path-based decision-making sub-trees providing valuable insights into user tagging behavior for information organization professionals. The limitations of this study and future research directions are discussed.     

Identifying genuine protein–protein interactions within communities of gene co‐expression networks using a deconvolution method

Direct relationships between biological molecules connected in a gene co‐expression network tend to reflect real biological activities such as gene regulation, protein–protein interactions (PPIs), and metabolisation. As correlation‐based networks contain numerous indirect connections, those direct relationships are always ‘hidden’ in them. Compared with the global network, network communities imply more biological significance on predicting protein function, detecting protein complexes and studying network evolution. Therefore, identifying direct relationships in communities is a pervasive and important topic in the biological sciences. Unfortunately, this field has not been well studied. A major thrust of this study is to apply a deconvolution algorithm on communities stemming from different gene co‐expression networks, which are constructed by fixing different thresholds for robustness analysis. Using the fifth Dialogue on Reverse Engineering Assessment and Methods challenge (DREAM5) framework, the authors demonstrate that nearly all new communities extracted from a ‘deconvolution filter’ contain more genuine PPIs than before deconvolution.Inspec keywords: proteins, deconvolution, genetics, bioinformatics, biology computing, molecular biophysicsOther keywords: identifying genuine protein–protein interactions, gene co‐expression network, deconvolution method, direct relationships, biological molecules, biological activities, gene regulation, correlation‐based networks, numerous indirect connections, global network, network communities, biological significance, protein function, protein complexes, studying network evolution, biological sciences, different gene co‐expression networks     

Why do people switch mobile payment service platforms? An empirical study in Taiwan

Although an increasing number of studies have recently investigated mobile payment service (MPS) platform user behaviours, the majority of these studies have focused on user adoption. However, user switching behaviour has been paid limited attention as yet. The current study investigates critical antecedents of user switching intention of MPS platform drawing on the perspective of the push-pull-mooring framework. The proposed model was empirically tested with 612 valid responses collected using on-line questionnaire survey in Taiwan. The results of this study revealed that user regret, alternative attractiveness, perceived complementarity of the alternative, and perceived network size of the alternative positively relate to user intention to switch. The most critical push effect driving MPS platform users to switch is user regret caused by dissatisfaction with system quality and information quality, especially in terms of system stability and visual attractiveness. In terms of pull effects, the improvement of alternative attractiveness and perceived network size of the alternative should be paid more attention, so thereby expanding their provision of complementary resources can be considered as an effective tactic. Furthermore, inertia is also negatively associated with user switching intention. Of all of the sources of inertia, uncertainty costs are a major crucial mooring effect. The performance of sunk costs still leaves much to be increased because of the insufficient user use and investment in the incumbent MPS platform. The findings will be helpful for MPS platform practitioners when fully comprehending users’ switching behaviour and formulating appropriate strategies to retain existing users while still attracting potential new users.     

On spatially explicit models of cholera epidemics

We generalize a recently proposed model for cholera epidemics that accounts for local communities of susceptibles and infectives in a spatially explicit arrangement of nodes linked by networks having different topologies. The vehicle of infection (Vibrio cholerae) is transported through the network links that are thought of as hydrological connections among susceptible communities. The mathematical tools used are borrowed from general schemes of reactive transport on river networks acting as the environmental matrix for the circulation and mixing of waterborne pathogens. Using the diffusion approximation, we analytically derive the speed of propagation for travelling fronts of epidemics on regular lattices (either one-dimensional or two-dimensional) endowed with uniform population density. Power laws are found that relate the propagation speed to the diffusion coefficient and the basic reproduction number. We numerically obtain the related, slower speed of epidemic spreading for more complex, yet realistic river structures such as Peano networks and optimal channel networks. The analysis of the limit case of uniformly distributed population sizes proves instrumental in establishing the overall conditions for the relevance of spatially explicit models. To that extent, the ratio between spreading and disease outbreak time scales proves the crucial parameter. The relevance of our results lies in the major differences potentially arising between the predictions of spatially explicit models and traditional compartmental models of the susceptible–infected–recovered (SIR)-like type. Our results suggest that in many cases of real-life epidemiological interest, time scales of disease dynamics may trigger outbreaks that significantly depart from the predictions of compartmental models.     

Identification of technological knowledge intermediaries

Intermediaries in a technological knowledge network have recently been highlighted as crucial innovation drivers that accelerate technological knowledge flows. Although the patent network analysis has been frequently used to monitor technological knowledge structures, it has examined only sources or recipients of the technological knowledge by mainly estimating technological knowledge inflows or outflows of a network node. This study, therefore, aims to identify technological knowledge intermediaries when a technology-level knowledge network is composed of several industries. First, types of technological knowledge flows are deductively classified into four types by highlighting industry affiliations of source technologies and recipient technologies. Second, a directed technological knowledge network is generated at the technology class level, using patent co-classification analysis. Third, for each class, mediating scores are measured according to the four types. The empirical analysis illustrates the Korea’s technological knowledge network between 2000 and 2008. As a result, the four types of mediating scores are compared between industries, and industry-wise technological knowledge intermediaries are identified. The proposed approach is practical to explore converging processes in technology development where technology classes act as technological knowledge intermediaries among diverse industries.     

Using eigenvectors of perturbed and collapsed adjacency matrices to explore bowtie structures in directed networks

The bowtie structure can illustrate not only the accessibility of the World Wide Web, but also the reachability of other directed networks. In this paper, we use the principal eigenvectors of the adjacency matrix with the unique largest eigenvalue to identify the strongly connected component of a directed network and fit the network into the bowtie structure. To ensure that the largest eigenvalue is unique, we add a little perturbation to the matrix before the eigen analysis. After the revelation of the bowtie structure centered on the strongly connected component with the largest unique eigenvalue, a directed network may have other bowtie structures centered on strongly connected components with smaller eigenvalues. To reveal other bowtie structures, we collapse the perturbed matrix by aggregating nodes of the strongly connected component with the largest eigenvalue into a supernode. Hence, the principal eigenvectors of the perturbed and collapsed matrix can be used to reveal the bowtie structure centered on the strongly connected component with the second largest eigenvalue. Furthermore, repeating the process of collapsing a strongly connected component and finding principal eigenvectors of the perturbed and collapsed matrix, we can reveal all the bowtie structures of a directed network.     

Animal transportation networks

Many group-living animals construct transportation networks of trails, galleries and burrows by modifying the environment to facilitate faster, safer or more efficient movement. Animal transportation networks can have direct influences on the fitness of individuals, whereas the shape and structure of transportation networks can influence community dynamics by facilitating contacts between different individuals and species. In this review, we discuss three key areas in the study of animal transportation networks: the topological properties of networks, network morphogenesis and growth, and the behaviour of network users. We present a brief primer on elements of network theory, and then discuss the different ways in which animal groups deal with the fundamental trade-off between the competing network properties of travel efficiency, robustness and infrastructure cost. We consider how the behaviour of network users can impact network efficiency, and call for studies that integrate both network topology and user behaviour. We finish with a prospectus for future research.     

Advanced Community Identification Model for Social Networks

Community detection in social networks is a hard problem because of the size, and the need of a deep understanding of network structure and functions. While several methods with significant effort in this direction have been devised, an outstanding open problem is the unknown number of communities, it is generally believed that the role of influential nodes that are surrounded by neighbors is very important. In addition, the similarity among nodes inside the same cluster is greater than among nodes from other clusters. Lately, the global and local methods of community detection have been getting more attention. Therefore, in this study, we propose an advanced community-detection model for social networks in order to identify network communities based on global and local information. Our proposed model initially detects the most influential nodes by using an Eigen score then performs local expansion powered by label propagation. This process is conducted with the same color till nodes reach maximum similarity. Finally, the communities are formed, and a clear community graph is displayed to the user. Our proposed model is completely parameter-free, and therefore, no prior information is required, such as the number of communities, etc. We perform simulations and experiments using well-known synthetic and real network benchmarks, and compare them with well-known state-of-the-art models. The results prove that our model is efficient in all aspects, because it quickly identifies communities in the network. Moreover, it can easily be used for friendship recommendations or in business recommendation systems.     

Dynamics of deceptive interactions in social networks

In this paper, we examine the role of lies in human social relations by implementing some salient characteristics of deceptive interactions into an opinion formation model, so as to describe the dynamical behaviour of a social network more realistically. In this model, we take into account such basic properties of social networks as the dynamics of the intensity of interactions, the influence of public opinion and the fact that in every human interaction it might be convenient to deceive or withhold information depending on the instantaneous situation of each individual in the network. We find that lies shape the topology of social networks, especially the formation of tightly linked, small communities with loose connections between them. We also find that agents with a larger proportion of deceptive interactions are the ones that connect communities of different opinion, and, in this sense, they have substantial centrality in the network. We then discuss the consequences of these results for the social behaviour of humans and predict the changes that could arise due to a varying tolerance for lies in society.     

Investigating online social media users’ behaviors for social commerce recommendations

Online social media create virtual communities and network platforms that people use to create, share, and exchange opinions, views and experiences. With social networks, social commerce not only relies on commerce, but online social media can also promote the sale of goods or services online. Many online operators have begun to use recommendation systems to analyze customer purchase history and identify individual products that customers may purchase. This enables the company to send product information to consumers to attract their attention. In addition, consumers have a higher purchase rate for recommended products based on consumer data. Based on a survey in Taiwan society, this study uses the questionnaire survey method to collect data on a relational database. This study investigates Taiwan online social media users’ behaviors using data mining methods, including clustering analysis and association rules. Clustering analysis is to investigate possible profiles of users and association rules are to find knowledge patterns and rules of user profiles, online social media usage motivation/preferences and social commerce behavior in order to generate social commerce recommendations in terms of social technology development in the modern society.