Friend Recommendation Research Articles

From Consumer's Attitude towards Online Purchase Decision: Influence of Friends and Reference Groups

The rapid advancement of e-commerce has piqued researchers' interest in identifying the determining factors of online purchase decisions. They have found abundant information about the influence of consumer attitudes but still have divergent results, indicating a gap in studies with mediating variables. The study aims to examine the mediating effect of the opinions of friends and reference groups on the relationships between consumer attitudes and online purchase decisions. The study surveyed 654 respondents and analyzed the data using partial least squares structural equations (PLS-SEM). The results show that consumers' attitudes significantly positively influence their online purchase decisions, which are mediated by the opinions of friends and influence groups. The opinions or recommendations of friends or other reference groups with similar purchasing experiences influence consumers. The research establishes a connection between constructs that influence online purchase decisions, with the main theoretical contribution being the analysis of the effect of friends' opinions and reference groups on generating a better consumer predisposition to make decisions in the virtual environment. The empirical contribution offers information that marketing managers should consider based on the results achieved, proposing strategies focused on reference groups that facilitate consumer decision-making.

Neighbor-Enhanced Representation Learning for Link Prediction in Dynamic Heterogeneous Attributed Networks

Dynamic link prediction aims to predict future connections among unconnected nodes in a network. It can be applied for friend recommendations, link completion, and other tasks. Network representation learning algorithms have demonstrated considerable effectiveness in various prediction tasks. However, most network representation learning algorithms are based on homogeneous networks and static networks for link prediction that do not consider rich semantic and dynamic information. Additionally, existing dynamic network representation learning methods neglect the neighborhood interaction structure of the node. In this work, we design a neighbor-enhanced dynamic heterogeneous attributed network embedding method (NeiDyHNE) for link prediction. In light of the impressive achievements of the heuristic methods, we learn the information of common neighbors and neighbors’ interaction in heterogeneous networks to preserve the neighbors proximity and common neighbors proximity. NeiDyHNE encodes the attributes and neighborhood structure of nodes as well as the evolutionary features of the dynamic network. More specifically, NeiDyHNE consists of the hierarchical structure attention module and the convolutional temporal attention module. The hierarchical structure attention module captures the rich features and semantic structure of nodes. The convolutional temporal attention module captures the evolutionary features of the network over time in dynamic heterogeneous networks. We evaluate our method and various baseline methods on the dynamic link prediction task. Experimental results demonstrate that our method is superior to baseline methods in terms of accuracy.

Cooperation and social organization depend on weighing private and public reputations

To avoid exploitation by defectors, people can use past experiences with others when deciding to cooperate or not (‘private information’). Alternatively, people can derive others’ reputation from ‘public’ information provided by individuals within the social network. However, public information may be aligned or misaligned with one’s own private experiences and different individuals, such as ‘friends’ and ‘enemies’, may have different opinions about the reputation of others. Using evolutionary agent-based simulations, we examine how cooperation and social organization is shaped when agents (1) prioritize private or public information about others’ reputation, and (2) integrate others’ opinions using a friend-focused or a friend-and-enemy focused heuristic (relying on reputation information from only friends or also enemies, respectively). When agents prioritize public information and rely on friend-and-enemy heuristics, we observe polarization cycles marked by high cooperation, invasion by defectors, and subsequent population fragmentation. Prioritizing private information diminishes polarization and defector invasions, but also results in limited cooperation. Only when using friend-focused heuristics and following past experiences or the recommendation of friends create prosperous and stable populations based on cooperation. These results show how combining one’s own experiences and the opinions of friends can lead to stable and large-scale cooperation and highlight the important role of following the advice of friends in the evolution of group cooperation.

BERT4FCA: A method for bipartite link prediction using formal concept analysis and BERT.

Link prediction in bipartite networks finds practical applications in various domains, including friend recommendation in social networks and chemical reaction prediction in metabolic networks. Recent studies have highlighted the potential for link prediction by maximal bi-cliques, which is a structural feature within bipartite networks that can be extracted using formal concept analysis (FCA). Although previous FCA-based methods for bipartite link prediction have achieved good performance, they still have the problem that they cannot fully capture the information of maximal bi-cliques. To solve this problem, we propose a novel method for link prediction in bipartite networks, utilizing a BERT-like transformer encoder network to enhance the contribution of FCA to link prediction. Our method facilitates bipartite link prediction by learning more information from the maximal bi-cliques and their order relations extracted by FCA. Experimental results on five real-world bipartite networks demonstrate that our method outperforms previous FCA-based methods, a state-of-the-art Graph Neural Network(GNN)-based method, and classic methods such as matrix-factorization and node2vec.

Factors of pre-war educational migration: an investigation of polish medical and dental students in Ukraine

BackgroundLow admission rates at Polish medical universities led many students to study abroad. Ukraine provided an appealing option for years. The purpose of the study is to investigate the most important factors that motived Polish students enrolled at I. Horbachevsky Ternopil National Medical University in Ukraine to pursue medical and dentistry studies in lower middle-income countries, just before the outbreak of the war. It has become incredibly important to determine crucial economic and non-economic factors.MethodsThe paper draws on a quantitative survey (N = 94) conducted among medical and dental students from Poland who had studied at I. Horbachevsky Ternopil National Medical University in Ukraine using a semi-structured questionnaire. To test whether there was a relationship between the importance of the motives, Pearson’s chi-square independence test was employed.ResultsThe study found the key economic and non-economic factors for pursuing Ukrainian studies were rejection from Polish tuition-free programs, lower Ukrainian tuition and living costs compared to Poland and to other European Union countries. Determining reasons for choosing Ternopil University were recruiter and friend recommendations and its prestige among Ukrainian medical schools. Gender influenced the ranking of motives.ConclusionThe article examines the unique pre-war educational migration from Poland to Ukraine – occurring counter to typical flows from lower to higher a lower income country. The study showed that universities should strengthen recruiting efforts and highlight competitive tuition and living costs to attract international students, especially from relatively more developed nations.

Reusable Menstrual Management Products: Perceptions and Use Among Hispanic Women in Puerto Rico [ID 2683545

INTRODUCTION: Reusable menstrual products offer much-needed eco-friendly and cost-effective alternatives for managing menstruation, yet disposables continue to dominate the market and public awareness. This study aims to explore women’s perceptions and knowledge of menstrual care options, revealing barriers to reusable products and the need for enhanced education. METHODS: A cross-sectional study was done using Survey Monkey. A 53-item questionnaire was administered covering sociodemographics, practices, and general knowledge. Recruitment primarily targeted university-aged women (21+) living in Puerto Rico via social media flyers. We employed descriptive statistics for analysis. The research was IRB-approved. RESULTS: Among 208 participants (85% completion rate), 20% combined reusable and disposable products. Motivations included waste reduction (31%), cost savings (26%), and friend recommendations (15%). Intriguingly, a substantial segment (30–44%) of nonusers (66%) expressed a willingness to switch, citing environmental concerns (42%) and desiring a progressive choice (18%). Hindrances to adoption encompassed perceived unhygienic aspects (29%) and lack of awareness (25%). Participants advocated for increased education, with only 12% having received information from their primary care physicians, 50% feeling their physicians lacked knowledge about available products, 81% expressing their doctors should initiate these conversations, and 87% believing menstrual management should be included in school curricula. CONCLUSION: Although the majority of respondents use disposable products, a noteworthy proportion is exploring reusable options or willing to switch to eco-friendly alternatives for waste reduction. Participants highlighted the need for primary care physicians and educators to incorporate menstrual management alternatives in their discussions, which could potentially result in large-scale cost savings and environmental benefits.

Tail-STEAK: Improve Friend Recommendation for Tail Users via Self-Training Enhanced Knowledge Distillation

Graph neural networks (GNNs) are commonly employed in collaborative friend recommendation systems. Nevertheless, recent studies reveal a notable performance gap, particularly for users with limited connections, commonly known as tail users, in contrast to their counterparts with abundant connections (head users). Uniformly treating head and tail users poses two challenges for tail user preference learning: (C1) Label Sparsity, as tail users typically possess limited labels; and (C2) Neighborhood Sparsity, where tail users exhibit sparse observable friendships, leading to distinct preference distributions and performance degradation compared to head users. In response to these challenges, we introduce Tail-STEAK, an innovative framework that combines self-training with enhanced knowledge distillation for tail user representation learning. To address(C1), we present Tail-STEAK-base, a two-stage self-training framework. In the first stage, only head users and their accurate connections are utilized for training, while pseudo links are generated for tail users in the second stage. To tackle (C2), we propose two data augmentation-based self-knowledge distillation pretext tasks. These tasks are seamlessly integrated into different stages of Tail-STEAK-base, culminating in the comprehensive Tail-STEAK framework. Extensive experiments, conducted on state-of-the-art GNN-based friend recommendation models, substantiate the efficacy of Tail-STEAK in significantly improving tail user performance. Our code and data are publicly available at https://github.com/antman9914/Tail-STEAK.

Instagram Influencers and Consumer Behaviour: A Study of Youth in Delhi

The traditional buying and selling process has undergone a radical change. Instagram has gained popularity and established itself as a hub for stylish women who only use it to read fashion blogs. Fashion blogs have become one of the most effective platforms for influencers to interact with their audience by giving their insights and advice. The study, titled "Instagram influencers' & consumer behavior: A study of youth in Delhi," aimed to find out how Instagram influencers influence young adults. An attempt was made to comprehend the factors that influence young adults' purchasing behavior. The study contributes to the literature by seeking responses from young adults. The major findings of the study indicate that male respondents spent most time on Instagram—five hours or more. The majority of respondents, both males and females, follow Instagram influencers, with 36% following Instagram influencers to get inspiration, and respondents stated that they are interested in trends, fashion knowledge, information, and entertainment, in addition to the options provided. Clothes were the most commonly purchased lifestyle product by respondents, as they constitute an important human need, followed by accessories, skincare, and cosmetics. The recommendation or suggestion of friends was considered the most essential aspect of the purchasing process.

Meta-path aware dynamic graph learning for friend recommendation with user mobility

Recently, friend recommendation has gained widespread popularity in location-based social networks (LBSNs), which provides more opportunities for users to forge new friendships. Most existing studies exploit user trajectories or check-ins of Point-Of-Interests (POIs) to predict friendships based on geographic homophily. However, the dynamics of social relationships are left insufficiently considered in modeling user preferences. In this paper, we explore how geographical and social preferences influence each other in a dynamic manner. Specifically, we propose a Meta-path aware Dynamic Graph with Subgraph Inference, named MDyGSI, which models the evolution of user preferences with time-phased sequences of POIs and social relationships for friend recommendation in LBSNs. In each time step of the evolution, geographical and social preferences are modeled through behavior-specific meta-paths in a dynamic heterogeneous graph. The formations of different meta-paths are facilitate by each other to explore mutual influences of dual preferences. To keep the dynamics of social relationships aligned with check-in history, reliable User-User connections are sampled from social graphs based on geographical collaborative filtering in each step, which also avoids noisy social interactions. Furthermore, the dual preferences are concatenated with evolutionary weights measured by a Gated Recurrent Unit for final recommendation. Experimental results on two real-world datasets show significant improvements in MDyGSI over the state-of-the-art methods.

Friends Recommendation on Social Networks using the Bayesian Personalized Ranking-Matrix Factorization

In the digital landscape of social networking, the challenge of improving friend recommendation systems is pivotal for enhancing user interaction and fostering social connections. Addressing this challenge, the current study innovates by fusing Bayesian Personalized Ranking (BPR) with Matrix Factorization (MF), culminating in a novel BPR-MF model designed for the intricacies of social network relationships. The study harnesses a rich dataset from LastFM, comprising 27,806 interactions among 7,624 users, to analyze mutual follower patterns and augment the precision of friend recommendations. Through rigorous preprocessing and systematic evaluation of the BPR-MF model against different numbers of latent factors, the research uncovers that a configuration of 20 latent factors is most effective, achieving an RMSE of 0.156 and an AUC ROC of 0.800. This discovery addresses the critical problem of balancing computational complexity with prediction accuracy in recommendation models. It also demonstrates the necessity for a nuanced, data-driven approach to generate relevant social connections. The research sets a new direction for future studies aiming to capitalize on user interaction data to offer precise friend suggestions, all while upholding user privacy and avoiding reliance on personal data.

Innovation of Mental Health Education Path in Colleges and Universities under the Perspective of “Internet +”

Abstract The traditional mental health education model has made it difficult to cope with the increasingly serious psychological problems of college students. In this paper, under the perspective of “Internet+”, based on the social network of college students, we constructed a propagation dynamics model of depression in colleges and universities and utilized the heterogeneous mean field to detect the propagation status of students’ depression. After calculating the comprehensive centrality degree of students with depression, the community mining of peer psychological adjustment is completed. The C-P similarity of students under the mental health perspective was defined by the ordering of the comprehensive centrality degree and the friend recommendation results were obtained by the distance measure of similarity. In the network of students without depressed mood, after a period of oscillation initially, the end state can be basically reached within 1000 steps, and the infected nodes reach 1000. When the depressed mood factor is small, such as S=0.02 and S=0.04, it has been able to prevent the transformation of depressed mood to the state of all 0, and the proportion occupied by positive and negative moods in the network is basically stable. Internet technology effectively improves the problem of identifying students with psychological abnormalities and implementing targeted mental health education.

Social Biases and Equity Investment Decisions of Individual Investors: Behavior Finance Perspective

Individual investors may not always pose a rational decision on stock market and heavily guided by their behavioral and social biases. This study is delineated to examine the role of social factors in the decision-making process of individual investors. This study is based on quantitative research approach and utilized a 5-points Likert type closed-end structured questionnaire survey to collect the data from 269 individuals, invested in the firms listed at Nepal Stock Exchange (NEPSE). This study utilized reliability analysis, descriptive statistics, Pearson correlation coefficient and ordinary least square (OLS) method. Evidence indicates that herding, social interaction and media effect shows the positive influences on equity investment decision process of the investors. Finally, this study justified that social biases play the significant role for equity investment decisions of the individual investors through excessive follow the humor of the market place, over or under reaction of the information about securities, follow the suggestions and recommendations of peers, friends, family members and neighbors along with highly depend upon news, social medias and information of so-called experts of capital market that leads to investment mistakes. Hence, investors should strongly follow the proper information about stock and market, utilize their own knowledge about information processing and analysis, and try to reduce the biases while making decisions through information analysis and processing that may downsize the irrational decisions.

Social Network Recommendations for Friends with Neo4j Graph Database

In recent years, along with the development of the internet, the number of social network users is increasing day by day. Through social networks, users can share, exchange information or make friends with each other. However, with new relationships, users often have a need to assess the credibility of a new friend before making friends on social networks. This paper proposes a friend recommendation system on social networks based on how to calculate the reliability between users. The recommendation system is implemented using the Neo4j graph database. The results of the truth algorithm proposed in this paper are higher than other similar algorithms.

Social Network Recommendations for Friends with Neo4j Graph Database

In recent years, along with the development of the internet, the number of social network users is increasing day by day. Through social networks, users can share, exchange information or make friends with each other. However, with new relationships, users often have a need to assess the credibility of a new friend before making friends on social networks. This paper proposes a friend recommendation system on social networks based on how to calculate the reliability between users. The recommendation system is implemented using the Neo4j graph database. The results of the truth algorithm proposed in this paper are higher than other similar algorithms.

The application of friend recommendation algorithm in the design of traditional handicraft art communication platform

Abstract Cultural awareness is a context in which the integration and development of traditional arts and crafts can be promoted by establishing an exchange platform. In this paper, the exchange platform is designed through the exploration of traditional handicraft art resources and the study of different classifications. The main points of the platform design and key operational processes are explored based on the analysis of the users’ needs for the traditional handicraft art platform. In response to the lack of accuracy of the traditional buddy recommendation algorithm, a buddy recommendation algorithm incorporating user interests was proposed based on big data to carve out different user preferences for traditional arts and crafts. The user interaction response rate of the platform was good, with TPS of 98.7, 137.4 and 75.6 on the sending side, service side and receiving side, respectively, in stable operation. The friend recommendation algorithm incorporating user interests can effectively dig out user groups with similar interests, which is optimized by 11.83% compared to the traditional friend recommendation algorithm. The exchange platform supported by big data, based on friend recommendations to strengthen user interaction, can effectively promote the exchange and integration of traditional handicraft arts.

Find Another me Across the World - Large-Scale Semantic Trajectory Analysis Using Spark

In today's society, location-based services are widely used which collect a huge amount of human trajectories. Analyzing semantic meanings of these trajectories can benefit numerous real-world applications, such as product advertisement, friend recommendation, and social behavior analysis. However, existing works on semantic trajectories are mostly centralized approaches that are not able to keep up with the rapidly growing trajectory collections. In this paper, we propose a novel large-scale semantic trajectory analysis algorithm in Apache Spark. We design a new hash function along with efficient distributed algorithms that can quickly compute semantic trajectory similarities and identify communities of people with similar behavior across the world. The experimental results show that our approach is more than 30 times faster than centralized approaches without sacrificing any accuracy like other parallel approaches.

RETRACTED ARTICLE: E-commerce User Recommendation Algorithm Based on Social Relationship Characteristics and Improved K-Means Algorithm

In the era of the Internet, information data continue to accumulate, and the explosive growth of network information explosion leads to the reduction of the accuracy of users’ access to information. To enhance the user experience and purchasing desire of e-commerce users, a e-commerce user recommendation algorithm based on social relationship characteristics and improved K-means algorithm is proposed. It combines the Automatic Time Division Dynamic Topic Model based on adaptive time slice division for building a strength calculation model in view of the characteristics of social relations. Then, it proposes an e-commerce user recommendation algorithm in view of the improved K-means algorithm to improve the accuracy of topic feature extraction and user recommendation. The experiment illustrates that there is no fluctuation in the clustering function of the improved K-means algorithm, and the highest, lowest, and average accuracy remain consistent under the three datasets, with average accuracy of 78.9%, 84.5%, and 95.9%, respectively. The community discovery-based friend recommendation algorithm presented in the study has the highest accuracy, illustrating that improving the K-means algorithm can further improve recommendation accuracy. The accuracy of the feature extraction method in view of alternative cost is 0.63, which improves the accuracy by about 9%. The results indicate that this study can provide technical support for user recommendations on e-commerce platforms.

Exploring the Relationship between the Factors Affecting Consumer Behavior concerning Mobile Phones in Nepal

This study examined the relationship between publicity, company website, advertisement, brand ambassador, friends' recommendations, and social media while making buying decisions by consumers concerning mobile phone purchases. The population of the study comprised university students of Nepal. A structured questionnaire was sent to respondents. Out of the returned responses, 165 responses were selected for this study. A correlation test was employed to examine the relationship between the stated variables of the study. The finding demonstrated positive correlations between the variables of the test. All the variables had a statistically positive correlation with each other at <.001, except (friends' recommendation and company website; and friend recommendation and brand ambassador).

Novel Friend Recommendations Based on User-Generated Contents in Online Social Media Sites

Novel Friend Recommendations Based on User-Generated Contents in Online Social Media Sites

Graph Structure Learning on User Mobility Data for Social Relationship Inference

With the prevalence of smart mobile devices and location-based services, uncovering social relationships from human mobility data is of great value in real-world spatio-temporal applications ranging from friend recommendation, advertisement targeting to transportation scheduling. While a handful of sophisticated graph embedding techniques are developed for social relationship inference, they are significantly limited to the sparse and noisy nature of user mobility data, as they all ignore the essential problem of the existence of a large amount of noisy data unrelated to social activities in such mobility data. In this work, we present Social Relationship Inference Network (SRINet), a novel Graph Neural Network (GNN) framework, to improve inference performance by learning to remove noisy data. Specifically, we first construct a multiplex user meeting graph to model the spatial-temporal interactions among users in different semantic contexts. Our proposed SRINet tactfully combines the representation learning ability of Graph Convolutional Networks (GCNs) with the power of removing noisy edges of graph structure learning, which can learn effective user embeddings on the multiplex user meeting graph in a semi-supervised manner. Extensive experiments on three real-world datasets demonstrate the superiority of SRINet against state-of-the-art techniques in inferring social relationships from user mobility data. The source code of our method is available at https://github.com/qinguangming1999/SRINet.