Virtual friend recommendations in virtual worlds

A survey of graph neural network based recommendation in social networks

Friend recommendation is a fundamental service for both social networks and practical applications. The majority of existing friend-recommendation methods utilize user profiles, social relationships, or static post content data, but rarely consider the semantic intentions and dynamic behaviors of users. In this paper, we propose FRec++, a friend recommendation method based on semantic and dynamic information. In FRec++, the first plus stands for new friend recommendation using a semantic model based on LDA while the next plus stands for the use of dynamic user attributes (e.g., behaviors or positions). More specifically, we first use the LDA method to generate semantic topics for user interests, and then compute the topic similarities between the target user and candidate friends. Next, we calculate the similarities of dynamic behaviors (i.e., forwarding, making comments, liking, and replying) and investigate the static attributes of users to measure the relevance of preferences. Finally, the weighted average method is used to integrate the above factors. We conducted experiments on the Weibo dataset and the results show that FRec++ outperforms several existing methods.

Friend recommendation is an important real-world application in Location-based Social Networks (LBSN), helping users discover potential friends and enhance their overall happiness. LBSN mainly comprises two distinct data structures: spatio-temporal data for human mobility and graph data for social networks. These two data structures make it challenging to model the complex relationships between them, which are essential for comprehensively understanding users’ lives. Previous studies have either modeled user trajectories and social networks separately or used classical simple graph-based methods, where a simple edge links only two nodes, failing to capture the multiple relationships inherent in LBSN. Furthermore, most studies have relied on Euclidean space to train their graph models, which could result in significant distortion because of tree-like social network data structure. To address these limitations, we propose a novel heterogeneous LBSN hypergraph that represents user check-in records and continuous trajectories—comprising multiple Points of Interest (POI)—as hyperedges, enabling the representation of complex spatio-temporal relationships. This approach enables us to link multiple nodes of different types by hyperedges and use hyperbolic spaces to create more efficient graph representations. Additionally, we devise a new type-specific attention mechanism for our Heterogeneous Hyperbolic Hypergraph Neural Network (H 3 GNN), which is end-to-end trainable and employs supervised contrastive learning to learn hypergraph node embeddings for the subsequent friend recommendation task with the help of hyperbolic space. Finally, our model H 3 GNN achieves better results than existing methods on six real-world city datasets, and our ablation studies demonstrate the effectiveness of each component. Additionally, our experiments indicate that H 3 GNN requires less data storage and training time compared to previous methods.

How to recommend friends in social network has attracted many research efforts. Most current friend recommendation methods are just based on the assumption that people will become friends if they have common interests which are usually estimated with the contents of their published posts and following relationships. However, friends recommended by these methods are only suitable for virtual social space instead of the real world. In this paper, we propose a new method to recommend friends in social network from the perspective of not just common interests, but also real-life needs. That is, we focus on finding friends that they can communicate with each other by social network and participate in some real-life activities face to face. The central idea of our approach is that we suppose people are more likely to be friends if their lives share more location overlaps besides the common interests. Currently, most people publish posts containing their real-time location information at any time, which makes it possible to detect and use the location information to recommend friends. Thus, our method combines users’ published posts, their location sequences detected from the posts and how active they are in Sina Weibo to estimate whether they can become friends in not only social network but also the real world. Experiments on Sina Weibo dataset demonstrate that our method can significantly outperform the traditional friend recommendation methods in terms of Precision, Recall and F1 measures.

Friend recommendation with content spread enhancement in social networks

Recently, communication and sharing the information through online social network (OSN) grows very fast in day to day life. As user share important data through online social network, therefore providing security and privacy to the online social network becomes the challenging task for the researchers. Moreover, for increasing the social connections and get the information from the particular group of people user needs to search the new friends and add them in the friend list. Now a days there isa friend recommendation method which was used by many online social networks, this method recommended the friend list to the user. As the privacy of these friend recommendation methods is the most significance task for the researchers and for the online social networks. Basically an online social network focused on the millions people's profile to spend time on an online social network, also implementing an OSN clients social circles, by using recommended suggestions. In this system, the main focus is on privacy and security of the online social network. This system basically focuses to support the users of online social network by secure trust creation with a stranger. This is achieved with multi-hop recommendation process. In the proposed method, the system provides privacy to the online social network from which system is allowing users to enhance their social networks. Existing system used secure KNN algorithm for securing social network. By using KNN algorithm, distance based learning is unclear. It is sort of distance and their component to use for the best results and calculation is very expensive. To overcome this limitation and to improve the precision, the proposed frame-work makes use of secure social coordinate matching with a Naive Bayes classifier. Also the privcy of social coordinates is maintained with Pailier encryption algorithm. Experimental results demonstrate that the proposed system is secure and time efficient than the existing one. Memory consumption is also improved with the proposed system.

There is widespread interest in the development and use of decentralized systems and virtual world environments as possible new places for engaging in collaborative work activities. Similarly, there is widespread interest in stimulating new technological innovations that enable people to come together through social networking, file/media sharing, and networked multi-player computer game play. A decentralized virtual activity system (DVAS) is a networked computer supported work/play system whose elements and social activities can be both virtual and decentralized (Scacchi et al. 2008b). Massively multi-player online games (MMOGs) such as World of Warcraft and online virtual worlds such as Second Life are each popular examples of a DVAS. Furthermore, these systems are beginning to be used for research, deve-lopment, and education activities in different science, technology, and engineering domains (Bainbridge 2007, Bohannon et al. 2009; Rieber 2005; Scacchi and Adams 2007; Shaffer 2006), which are also of interest here. This chapter explores two case studies of DVASs developed at the University of California at Irvine that employ game-based virtual worlds to support collaborative work/play activities in different settings. The settings include those that model and simulate practical or imaginative physical worlds in different domains of science, technology, or engineering through alternative virtual worlds where players/workers engage in different kinds of quests or quest-like workflows (Jakobsson 2006).

The tremendous growth of social media has been accompanied by highly advanced online social network (OSN) technologies. Such advanced technologies have been heavily utilized by perpetrators as convenient tools for deceiving people in online worlds. Social capital has been discussed as a powerful mechanism to leverage interpersonal relationships in social networks in order for an individual to achieve his/her goal. The beauty of social capital is the ability to materialize non-monetary, less costly, and non-economic resources into tools to solve social problems. In this paper, we aim to leverage social capital (SC) to minimize online users' vulnerabilities to online deception. In particular, we propose a Social cApital-based FriEnd Recommendation scheme, called SAFER, that can protect OSN users from phishing attacks. We quantify three dimensions of social capital, namely, structural, cognitive, and relational, based on user features obtained from real datasets and model a user's friending behavior based on their social capital. In addition, to model a user's behavior upon being attacked by phishing attacker, we developed the so-called SER-SEIR (Susceptible, Exposed, Recovered-Susceptible, Exposed, Infected, and Recovered) model as a variant of the SEIR model. Via extensive simulation experiments based on two real datasets considering bot-based and human-based attackers performing phishing attacks, we demonstrate the performance of four SC-based friend recommendation schemes with three non-SC-based comparable counterparts in terms of the ratio of detecting attackers and the fraction of users in the states of S, E, I, and R. Based on the performance comparison, we analyze the overall trends of their performance in terms of the extent of resistance against phishing attacks by bot or human attackers.

Virtual worlds for children are becoming increasingly popular, and yet there are few accounts of children’s use of these worlds. Young children are spending increasing amounts of time online as technology continues to create significant changes in social and cultural practices in the 21st century. Some of children's online interactions can be categorized as playful in nature; however, play and technology are frequently positioned as oppositional. In this article, I explore the tensions surrounding the relationship between play and technology and relate it to similar discourses concerning the concepts of ‘real’ and ‘virtual’. I then move on to consider the growing popularity of virtual worlds with young children and examine the way in which the worlds have been marketed to children and parents/carers on the basis of their propensity to offer online play in a safe environment. The article provides an overview of two virtual worlds currently targeted at young children and draws on a survey of primary children’s use of virtual worlds in order to identify the nature of play in these environments. One hundred and seventy-five children aged 5—11 completed an online survey and 15 took part in group interviews in which their use of virtual worlds was explored. This article focuses on the data relating to 17 children aged from five to seven years who used virtual worlds. Findings indicate that virtual worlds offered these young children a wide range of opportunities for play and that the types of play in which they engaged relate closely to ‘offline’ play. The implications for early years educators are considered.

The paper outlines a new paradigm and its underlying rationales for implementing networked learning environments that is emerging from new technologies such as multi-user platform, virtual worlds, virtual learning community, and intelligent agents. The proposed paradigm of the networked learning environments is described as inhabited virtual learning worlds (IVLW), which is a shared learning space in 3-D format and populated with avatars that are the representations of learners who are geographically dispersed around the world. The virtual learning worlds are also composed of objects such as intelligent agents and learning materials. A pilot system is created based on the discussed rationales of inhabited virtual learning worlds. A preliminary empirical study focusing on the selected learning behaviors in young learners also has been conducted with the pilot system. The results of the empirical study and suggestions for enhancing the pilot system are discussed in the closing section of the article.

Social networks have become very popular in today’s age. As a result, Friend Recommendation Systems (FRSs) have gained immense importance. Teenagers today are often depressed and feel lonely. Social Networks, if used wisely can help a user connect to the right set of friends which can help to solve the above-mentioned challenges. Most popular social media platforms rely on homophily-based FRSs for their friend recommendations. That is, they use the user’s social graph to match him/her with other users. Users now are concerned about how their personal data is being used and its privacy. Therefore, most users do not reveal their personal information on social networks. Also, there is no evidence to the accuracy of user information. Therefore, the existing recommendation systems in social media prove to be inaccurate in a lot of cases. Furthermore, the main purpose of these traditional recommendation systems is to connect pre-existing friends online rather than suggesting users that could be potential friends. To connect users to a new set of people, there is a need to design a friend recommendation system that matches users based on something other than existing connections. In this paper, we propose a system which uses personality analysis along with hybrid filtering to make personalized friend recommendations along with other filtering options for age range, location, gender, etc. This system would solve the cold start problem which occurs because in the present scenario people don’t like spending time to specify their preferences for friend recommendations. The filtering options are of further interest to the user as there is a possibility of connecting to a new bunch of friends in person too.

Social Desirability in Virtual Communities

There are benefits for social and behavioral researchers to conduct studies in online virtual worlds. However, typically learning scripting takes additional time or money to hire a consultant. We propose a prototype Virtual World Toolkit for to help researchers design, set up, and run experiments in Virtual Worlds, with little coding or scripting experience needed. We explored three types of prototype designs, focused on a traditional interface with pilot results. We also present results of initial expert user study of our toolkit to determine the learnability, usability, and feasibility of our toolkit to conduct experiments. Results suggest that our toolkit requires little training and sufficient capabilities for a basic experiment. The toolkit received a great feedback from a number of expert users who thought that it is a promising first version that lays the foundation to more future improvements. This toolkit prototype contributes to enhancing researchers’ capabilities in conducting social/behavioral studies in virtual worlds and hopefully will empower social and behavioral researchers by proving a toolkit prototype that requires less time, efforts and costs to setup stimulus responses types of human subject studies in virtual worlds.

With hundreds of millions of users already today,virtual worlds will become an important factor in tomorrow's media landscape. In a virtual world, users are represented by so-called avatars. These avatars move around the virtual world, communicate with each other,and interact with the virtual world. The movements of these avatars can be tracked precisely, and useful information can be inferred from analyzing these movements. In this paper, we analyze a large data set (≫200 million records) of position data describing the movements of avatars in the virtual world Second Life.The dataset was derived from in-world sensors that had been deployed beforehand, but also so-called bots can be used to gather such information. From this data, we can track usage patterns of avatars (and therefore users) overtime. We can also identify regions of high interest where a large number of users gather frequently (which would be important for planning advertising in the virtual world), and visualize this statistical analysis using heat maps. By combining the position data with information about the language spoken by the avatars, we can label these regions according to the language predominantly spoken there. Analyzing incidents of co-location of avatars over a period of time, we can automatically infer friends, and eventually social networks. Using additional metadata such as language we can label clusters in this automatically generated social network.

In recent years, virtual communities have become the topic of countless books, journal articles and television shows, but what are they, and where did they come from? According to Preece, Maloney-Krichmar, and Abras (2003), the roots of virtual communities date back to as early as 1971 when e-mail first made its appearance on the Advanced Research Projects Agency Network (ARPANET), which was created by the United State’s Department of Defense. This network would lead to the development of dial-up bulletin board systems (BBSs) which would allow people to use their modems to connect to remote computers and participate in the exchange of e-mail and the first discussion boards. From these beginnings a host of multi user domains (MUDs) and multi-user object oriented domains (MOOs) would spring up all over the wired world. These multi-user environments would allow people to explore an imaginary space and would allow them to interact both with the electronic environment and other users. Additionally, listservs (or mailing lists) sprang up in 1986, and now, almost two decades later, they are still in use as the major method of communication among groups of people sharing common personal or professional interests (L-Soft, 2003). Since then the Internet has exploded due to the development of Web browsers as well as the development of communications technologies such as broadband, digital subscriber line (DSL), and satellite communications. Groups of people from as few as two and reaching to many thousands now communicate via email, chat, and online communities such as the Whole Earth ‘Lectronic Link (WELL) and such services as MSN, Friendster, America Online (AoL), Geocities, and Yahoo! Groups. Other examples of online communities are collaborative encyclopedias like Wikipedia. Web logs (Blogs) like Slashdot.com and LiveJournal allow users to create their own content and also to comment on the content of others. They also allow the users to create identities and to make virtual “friends” with other users. The definition of virtual community itself becomes as convoluted as the multitude of technologies that drives it. Are e-mail lists, message boards, and chat rooms online communities or are they virtual communities? Virtual communities might be persistent worlds as those found in popular online games (Everquest, 2004, Ultima Online, 2004) or virtual worlds (such as MUDs and MOOs) where the user is able to explore a simulated world or to take on a digital “physicality” in the form of an avatar. It becomes clear from the literature that the terms are still used interchangeably.