Behavior In Online Social Networks Research Articles

Beliefs trump facts: Effect of Frequent Corrections on Misinformation Beliefs during Extended Extreme Events

Providing corrections to people who have engaged with false claims in Online Social Networks (OSN) is a form of cognitive intervention employed to address the spread of misinformation. Although there is a large body of work that has studied the effectiveness of corrections for promoting accurate beliefs, there is still much uncertainty around the precise effects of corrections on individuals’ behaviors in OSNs. Notably, the effect of offering frequent corrections on discerning information and identifying misinformation remains uncertain. We conducted two laboratory experiments to test whether experiencing frequent corrections to misinformation improved peoples’ ability to discriminate between true and false news claims during extended extreme events like the COVID-19 pandemic. All participants recruited for the experiments were from USA. They received corrections at varying frequencies, depending on their assigned experimental condition. Results from both experiments suggest that increasing frequency of corrections may not affect people’s ability to correctly assess information (or misinformation). Participant’s beliefs (vaccine hesitancy, belief in mask effectiveness, and trust in fact-checking organization) were found to be the most significant contributing factors to the ability to learn from corrections. We discuss the implications of the findings from these experiments.

Interaction behavior enhanced community detection in online social networks

Community detection in online social networks (OSNs) has attracted a great deal of attention in recent years due to its potential prospects in applications such as recommendation systems, spam detection, targeted advertising, etc. However, community detection methods that have been proposed so far rely mostly on static information, such as the topology of the network, the attributes of the users, etc. In this paper, we propose a framework that enhances community detection with user interaction behaviors in OSNs since such behaviors should be important factors in characterizing the communities. In the framework, the various features of user interaction behaviors are expressed in the form of interaction indices through utilizing the cumulative distribution function and interaction divergence. The indices are then transformed into edge weights through a logistic regression model and combined with the static information to construct a weighted graph for the OSN. Finally, community detection is realized by applying a modularity maximization algorithm. A comprehensive evaluation using diverse real-world datasets shows that our proposed framework has superior performance over comparable ones in deriving high-quality communities. Critical metrics, such as modularity, average clustering coefficient, and average conductance, are evaluated to demonstrate the effectiveness of our approach. Furthermore, the synergy between network topology and user interaction behavior in community detection is also investigated.

Deep convolutional neural network based real-time abnormal behavior detection in social networks

In this study, we introduce a novel real-time approach for detecting abnormal behaviors in online social networks, leveraging both dynamic user activities and their associated profiles through Convolutional Neural Networks (CNN). Social networks, which serve as hubs for communication, discussion, and multimedia sharing, amass a plethora of user data. This vast reservoir of information, while invaluable for genuine interactions, is susceptible to exploitation by malicious entities. Despite numerous previously proposed detection methods, the challenge of ensuring security persists. Our distinctive framework, rigorously tested using an extensive dataset sourced from the web and executed via the CNN toolbox in MATLAB, consistently demonstrates an impressive 99.9% classification accuracy. This achievement not only underscores the method's effectiveness but also marks a significant stride towards proactively identifying and neutralizing potential threats in digital social platforms.

Understanding Collective Human Behavior in Social Media Networks Via the Dynamical Hypothesis: Applications to Radicalization and Conspiratorial Beliefs.

The dynamical hypothesis has served to explore the ways in which cognitive agents can be understood dynamically and considered dynamical systems. Originally used to explain simple physical systems as a metaphor for cognition (i.e., the Watt governor) and eventually more complex animal systems (e.g., bird flocks), we argue that the dynamical hypothesis is among the most viable approaches to understanding pressing modern-day issues that arise from collective human behavior in online social networks. First, we discuss how the dynamical hypothesis is positioned to describe, predict, and explain the time-evolving nature of complex systems. Next, we adopt an interdisciplinary perspective to describe how online social networks are appropriately understood as dynamical systems. We introduce a dynamical modeling approach to reveal information about emergent properties in social media, where radicalized conspiratorial beliefs arise via coordination between user-level and community-level comments. Lastly, we contrast how the dynamical hypothesis differs from alternatives in explaining collective human behavior in social networks.

Characterizing community behavior in OSNs: Modeling and forecasting activity on Facebook using the SARIMA model

Online social networks (OSNs) occupy an important part in users’ daily life as they maintain the flow of interaction and information exchange on all local, national, and global scales.This work develops a time series model of interactions on Facebook using the SARIMA (seasonal autoregressive integrated moving average) time series modeling technique based on empiricism with the theoretical model of regular user behavior. A case study of the Moroccan community, which has a high rate of interactions, is carried out to test the conformity of the model with the measurements. The results show that the SARIMA model is better suited to modeling the flow of interactions. The application of the SNR method on the signal energies has allowed to measure the usage damping in the users. The multitude of applied approaches have allowed to extract some main characteristics of this large and complex network.

Days with and without self-injurious thoughts and behaviors: Impact of childhood maltreatment on adolescent online social networking.

Among adolescents, an increasing rate of interpersonal interactions occurs online. Previous research has shown that interpersonal context exerts a robust impact on suicidal thoughts or behaviors, yet little attention has focused on examining the content of online interactions surrounding self-injurious thoughts or behaviors. As such, the present study sought to compare online social networking behaviors among adolescents on days with and without experiencing self-injurious thoughts or behaviors, as influenced by childhood maltreatment history. Adolescents aged 13-18 hospitalized for self-injurious thoughts or behaviors were recruited as part of an ongoing longitudinal study. A subsample (N = 22) of adolescents provided data from their online social networking platforms (i.e., text messages, Facebook, Instagram, and Twitter). Using a mixed-methods approach, online social networking data on days of experiencing self-injurious thoughts or behaviors and days of not experiencing self-injurious thoughts or behaviors were compared. Results indicate the frequency and content of online social networking messaging do not change by day of self-injurious thoughts or behaviors or history of childhood maltreatment. However, childhood maltreatment predicts received conflictual messages as well as sent symptomatic messages on days of experiencing self-injurious thoughts or behaviors. Childhood maltreatment may play a role in the content of adolescent online behaviors, particularly on days when they experience self-injurious thoughts or behaviors. Implications for intervention are discussed.

Feature extractions and selection of bot detection on Twitter A systematic literature review

Abstract Automated or semiautomated computer programs that imitate humans and/or human behavior in online social networks are known as social bots. Users can be attacked by social bots to achieve several hidden aims, such as spreading information or influencing targets. While researchers develop a variety of methods to detect social media bot accounts, attackers adapt their bots to avoid detection. This field necessitates ongoing growth, particularly in the areas of feature selection and extraction. The study's purpose is to provide an overview of bot attacks on Twitter, shedding light on issues in feature extraction and selection that have a significant impact on the accuracy of bot detection algorithms, and highlighting the weaknesses in training time and dimensionality reduction. To the best of our knowledge, this study is the first systematic literature review based on a preset search-strategy that encompasses literature published between 2018 and 2021 which are concerned with Twitter features (attributes). The key findings of this research are threefold. First, the paper provides an improved taxonomy of feature extraction and selection approaches. Second, it includes a comprehensive overview of approaches for detecting bots in the Twitter platform, particularly machine learning techniques. The percentage was calculated using the proposed taxonomy, with metadata, tweet text, and merging (meta and tweet text) accounting for 37%, 31%, and 32%, respectively. Third, some gaps are also highlighted for further research. The first is that public datasets are not precise or suitable in size. Second, the use of integrated systems and real-time detection is uncommon. Third, detecting each bots category identified separately is needed, rather than detecting all categories of bots using one generic model and the same features' values. Finally, extracting influential features that assist machine learning algorithms in detecting Twitter bots with high accuracy is critical, especially if the type of bot is pre-determined.

Identification of Bots and Cyborgs in the #FeesMustFall Campaign

Bots (social robots) are computer programs that replicate human behavior in online social networks. They are either fully automated or semi-automated, and their use makes online activism vulnerable to manipulation. This study examines the existence of social robots in the #FeesMustFall movement by conducting a scientific investigation into whether social bots were present in the form of Twitter bots and cyborgs. A total of 576,823 tweets posted between 15 October 2015 and 10 April 2017 were cleaned, with 490,449 tweets analyzed for 90,783 unique persons. Three separate approaches were used to screen out suspicious bot and cyborg activity, supplemented by the DeBot team’s methodology. User 1 and User 2, two of the 90,783 individuals, were recognized as bots or cyborgs in the study and contributed 22,413 (4.57 percent) of the 490,449 tweets. This confirms the existence of bots throughout the campaign, which aided in the #FeesMustFall’s amplification on Twitter, complicating sentiment analysis and invariably making it the most popular and lengthiest hashtag campaign in Africa, particularly at the time of data collection.

Characterizing user behavior in online social networks: Analysis of the regular use of Facebook

The analysis of user behaviour in online social networks (OSNs) is one of the important research interests related to human-computer interactions. OSNs gives a large space to share news with no limits around the world and allows user to benefit from properties of this interactive and dynamic system. The study of user behaviour on a social and popular platform characterized by the use of new technologies requires to understand and the analysis of collective behaviour on Facebook. This paper aims to analyse the usage patterns in OSNs using the visible interactions of Facebook, by studying the time of activity and the evolution of human behaviour through a process of detection of visible and non-volatile interactions. In the first step, we perform a data collection process based on breadth first search algorithm (BFS) and semi-supervised crawler agent. In the second step, we build an interaction quantification process to measure users’ activities and analysis related time series. The study of the frequency of periodic use has shown that the communities monitored follow a weekly rhythm that decreases over time to reach a frequency of daily use, which reflects a stability of activities and a case of dependency of use.

Predicting Mental Health From Followed Accounts on Twitter

The past decade has seen rapid growth in research linking stable psychological characteristics (i.e., traits) to digital records of online behavior in Online Social Networks (OSNs) like Facebook and Twitter, which has implications for basic and applied behavioral sciences. Findings indicate that a broad range of psychological characteristics can be predicted from various behavioral residue online, including language used in posts on Facebook (Park et al., 2015) and Twitter (Reece et al., 2017), and which pages a person ‘likes’ on Facebook (e.g., Kosinski, Stillwell, & Graepel, 2013). The present study examined the extent to which the accounts a user follows on Twitter can be used to predict individual differences in self-reported anxiety, depression, post-traumatic stress, and anger. Followed accounts on Twitter offer distinct theoretical and practical advantages for researchers; they are potentially less subject to overt impression management and may better capture passive users. Using an approach designed to minimize overfitting and provide unbiased estimates of predictive accuracy, our results indicate that each of the four constructs can be predicted with modest accuracy (out-of-sample R’s of approximately .2). Exploratory analyses revealed that anger, but not the other constructs, was distinctly reflected in followed accounts, and there was some indication of bias in predictions for women (vs. men) but not for racial/ethnic minorities (vs. majorities). We discuss our results in light of theories linking psychological traits to behavior online, applications seeking to infer psychological characteristics from records of online behavior, and ethical issues such as algorithmic bias and users’ privacy.

Fusing Behavioral Projection Models for Identity Theft Detection in Online Social Networks

We aim at exploiting users’ coarse behavioral records for identity theft detection in online services. We concentrate on this issue in online social networks (OSNs) that users’ behavioral records usually consist of multiple dimensional behavior data. The behavioral records in each dimension are possibly coarse and insufficient for effectively modeling users’ behavioral patterns. In this paper, we investigate whether there is a complementary effect among different dimensions of records for modeling users’ behavioral patterns. We focus on three typical dimensions of behaviors in OSNs, i.e., offline check-ins, online tip-postings, and social contacts. We devise the dedicated behavior models based on each dimension of data, i.e., users’ behavioral projection models. Then, by examining all feasible logical combinations of them, we find the optimal ones for two real-world data sets: Foursquare and Yelp. Notably, we analyze the potential correlation between customized demand and optimal logical fusion scheme. As an insightful result, we find that the correlation is independent of the specific data. This study would give the cybersecurity community new insights into the possibility and methodology to achieve a customized identity theft detection in OSNs by integrating multiple behavioral projection models.

The reflection of offline activities on users’ online social behavior: An observational study

The ever increasing presence of online social networks in users’ daily lives has led to the interplay between users’ online and offline activities. There have already been several works that have studied the impact of users’ online activities on their offline behavior, e.g., the impact of interaction with friends on an exercise social network on the number of daily steps. In this paper, we consider the inverse to what has already been studied and report on our extensive study that explores the potential causal effects of users’ offline activities on their online social behavior. The objective of our work is to understand whether the activities that users are involved with in their real daily life, which place them within or away from social situations, have any direct causal impact on their behavior in online social networks. Our work is motivated by the theory of normative social influence, which argues that individuals may show behaviors or express opinions that conform to those of the community for the sake of being accepted or from fear of rejection or isolation. We have collected data from two online social networks, namely Twitter and Foursquare, and systematically aligned user content on both social networks. On this basis, we have performed a natural experiment that took the form of an interrupted time series with a comparison group design to study whether users’ socially situated offline activities exhibited through their Foursquare check-ins impact their online behavior captured through the content they share on Twitter. Our main findings can be summarised as follows (1) a change in users’ offline behavior that affects the level of users’ exposure to social situations, e.g., starting to go to the gym or discontinuing frequenting bars, can have a causal impact on users’ online topical interests and sentiment; and (2) the causal relations between users’ socially situated offline activities and their online social behavior can be used to build effective predictive models of users’ online topical interests and sentiments.

A new user behavior evaluation method in online social network

Due to the limitations of existing user evaluation in online social network (OSN), a new user behavior evaluation model is proposed. Considering the high degree of uncertainty, complexity, dynamics of user behavior in OSN, cloud model theory is innovatively introduced to evaluate the user behavior, on this basis, a credible evaluation scheme concerning user behavior combined with entropy weighting is proposed in this paper. The user's original behavior is classified as the evaluation factors according to the attribute, then the behavior attribute cloud is established via Reverse Cloud Generator algorithm; At the same time, the user is divided into different trust levels according to the actual situation, and hierarchical cloud is generated through Positive Cloud Generator algorithm. The membership matrix of the behavior attribute cloud to the hierarchical cloud is calculated by employing a mathematical formula, and the weight of each attribute is adaptively calculated by the entropy method, overcoming the limitations of subjective weight assignation. This model provides a new thinking of user behavior evaluation in OSN, meanwhile, also expands the application of cloud model theory and fuzzy comprehensive evaluation method. Finally, as an example of embodiment, a case study is presented for user behavior evaluation in Facebook, both the feasibility and effectiveness are verified.

Implicit User Trust Modeling Based on User Attributes and Behavior in Online Social Networks

In this paper, we present a new user trustworthiness estimation model for social networks (SN), whereas most of existing researches have been focused on the user-user/item relationship trustworthiness estimation. Users share information of their interest on various social media without their trustworthiness verification. Therefore, SN are susceptible to malicious users for misinformation spreading. In SN, the original information source is generally unknown and the user who is sharing the contents is the only known information about the source. Therefore, the user's trustworthiness is an effective criterion for SN content's trustworthiness estimation. However, the users are unable to identify trustworthy/untrustworthy users, and the existing user-user/item relationship models do not provide user trustworthiness information. Our proposed model provides a systematic way to assess the user trustworthiness based on user attributes and interaction behavior. The proposed model is helpful to avoid the trust sparsity (implicit trust model), trust subjectivity (user's objective/collective trustworthiness estimation model) and cold-start user's trustworthiness (user's attributes-based trust modeling) problems. We employ friends-recommendation (FR) as an exemplary application to evaluate the performance of our proposed model in trust-aware recommendations. Simulation results illustrate that our trust-aware FR model outperformed the existing trust and FR models.

A Class of Rumor Spreading Models with Population Dynamics**Supported by National Natural Science Foundation of China under Grant Nos. 11275017 and 11173028

Based on the characteristics of rumor spreading in online social networks, this paper proposes a new rumor spreading model. This is an improved SIS rumor spreading model in online social networks that combines the transmission dynamics and population dynamics with consideration of the impact of both of the changing number of online social network users and different levels of user activity. We numerically simulate the rumor spreading process. The results of numerical simulation show that the improved SIS model can successfully characterize the rumor spreading behavior in online social networks. We also give the effective strategies of curbing the rumor spreading in online social networks.

User modeling and usage profiling based on temporal posting behavior in OSNs

In this paper, we study the posting behavior of online social network (OSN) users, in particular the posting frequency and temporal patterns, and consider possible interpretations of how users use the platform. At the aggregate (macro) level, we find two distinct peaks of traffic, one during morning working hours, and one in the evening. The morning peak is more pronounced for frequent posters, while the evening peak is pronounced for the remaining users. We postulate that this difference results from different usage purposes of the OSN platform (e.g. for work, with customers, etc.) than purely social interactions (e.g., friends, family, etc.). We also study user posting behavior at an individual (micro) level and model the user posting sequences as generated by a Hidden Markov Model. We compare the results of using a simple zeroth order model (which is equivalent to a topic model such as LDA), and a first-order model, in terms of their effectiveness in clustering and predicting user types, and show the advantage gained by the first-order HMM. Overall, our study provides new insights into user activity in today’s OSNs, and suggests a framework for profiling users based on their posting activities. We believe our approach will complement other methods of user profiling based on static demographic information and friendship network information.

Fine-grained Emotion Role Detection Based on Retweet Information

User behaviors in online social networks convey not only literal information but also one’s emotional attitudes towards the information. To compute this attitude, we define the concept of emotion role as the concentrated reflection of a user’s online emotional characteristics. Emotion role detection aims to better understand the structure and sentiments of online social networks and support further analysis, e.g., revealing public opinions, providing personalized recommendations, and detecting influential users. In this article, we first introduce the definition of a fine-grained emotion role, which consists of two dimensions: emotion orientation (i.e., positive, negative, and neutral) and emotion influence (i.e., leader and follower). We then propose a Multi-dimensional Emotion Role Mining model (MERM) to determine a user’s emotion role in online social networks. Specifically, we tend to identify emotion roles by combining a set of features that reflect a user’s online emotional status, including degree of emotional characteristics, accumulated emotion preference, structural factor, temporal factor, and emotion change factor. Experiment results on a real-life micro-blog reposting dataset show that the classification accuracy of the proposed model can achieve up to 90.1%.

A physical pathway to understand individual's labeling behavior in signed social networks

In this paper we propose a reshuffling approach to empirical analyze individual's labeling behavior in signed social networks. In our approach, each individual is assumed to have the ability to re-label his/her neighbors randomly with the parameters ps and p+. Many reshuffled networks, which have the same topological structure and different signs' configuration, are built through applying our approach to the given three signed social networks. The entropy Sout and the giant component ρG for each reshuffled networks are calculated and analyzed. We find that there exist two kinds of individual's labeling behavior according to the suppressed effect of Sout and the exponent α in the relationship of ρG and q+. Additionally, the suppressed effect of Sout shows the non-randomness factor in individual's labeling behavior. These results offer new insights to understand human's behavior in online social networks.

Learning to lurker rank: an evaluation of learning-to-rank methods for lurking behavior analysis

While being long researched in social science and computer–human interaction, lurking behaviors in online social networks (OSNs) have been computationally studied only in recent years. Remarkably, determining lurking behaviors has been modeled as an unsupervised, eigenvector-centrality-based ranking problem, and it has been shown that lurkers can effectively be ranked according to the link structure of an OSN graph. Although this approach has enabled researchers to overcome the lack of ground-truth data at a large scale, the complexity of the problem hints at the opportunity of learning from past lurking experiences as well as of using a variety of behavioral features, including any available, possibly platform-specific information on the activity and interaction of lurkers in an OSN. In this paper, we leverage this opportunity in a principled way, by proposing a machine-learning framework which, once trained on lurking/non-lurking examples from multiple OSNs, allows us to predict the ranking of unseen lurking behaviors, ultimately enabling the prioritization of user engagement tasks. Results obtained on 23 network datasets by state-of-the-art learning-to-rank methods, using different optimization and evaluation criteria, show the significance of the proposed approach.

Privacy-preserved community discovery in online social networks

Community detection is a pivotal task for understanding user behaviors in online social networks, in which a third-party server can partition the users with close social relationships and similar behaviors into a same group. The existing approaches for community detection usually request full access to detailed social connections among users, which are usually sensitive. How to derive a meaningful community structure while not disclosing sensitive information remains unsettled. In this work, a novel framework is proposed to discover community structure in online social networks while preserving sensitive link information. The framework takes both social connections and users’ published contents into consideration. It also provides the flexibility in which a third-party server can adaptively select the concerned subgraph. The experiment results towards a real world dataset show that the proposed framework outperforms the baseline algorithm and can achieve a high accuracy on the discovered community structure.