Spammer Detection and Fake User Identification on Social Networks

Zhou

et al. 2021

Scientific Programming

Self Cite

Social Internet of Things (SIoT) is a variation of social networks that adopt the property of peer-to-peer networks, in which connections between the things and social actors are automatically established. SIoT is a part of various organizations that inherit the social interaction, and these organizations include industries, institutions, and other establishments. Triadic closure and homophily are the most commonly used measures to investigate social networks’ formation and nature, where both measures are used exclusively or with statistical models. The triadic closure patterns are mapped for actors’ communication behavior over a location-based social network, affecting the homophily. In this study, we investigate triads emergence in homophilic social networks. This evaluation is based on the empirical review of triads within social networks (SNs) formed on Big Data. We utilized a large location-based dataset for an in-depth analysis, the Chinese telecommunication-based anonymized call detail records (CDRs). Two other openly available datasets, Brightkite and Gowalla, were also studied. We identified and proposed three social triad classes in a homophilic network to feature the correlation between social triads and homophily. The study opened a promising research direction that relates the variation of homophily based on closure triads nature. The homophilic triads are further categorized into transitive and intransitive groups. As our concluding research objective, we examined the relative triadic throughput within a location-based social network for the given datasets. The research study attains significant results highlighting the positive connection between homophily and a specific social triad class.

Section: Resultssupporting

confidence: 58%

Section: Introductionmentioning

confidence: 99%

Correlation between Triadic Closure and Homophily Formed over Location-Based Social Networks

Zhou

et al. 2021

Scientific Programming

Self Cite

“…Experimental results on multiple benchmark datasets of deceptive reviews performed well. F et al [28] used dynamic knowledge graphs to detect fake reviews F. Masood [29] , A. Rastogi [30] , Rodrigo Barbado [31] and others identified fake reviews by combining multiple features such as text features and user behavior information.…”

Section: B Identify Fake Reviews From the Reviewer's Behavior Characmentioning

confidence: 99%

Fake Review Detection Based on Multiple Feature Fusion and Rolling Collaborative Training

et al. 2020

Fake reviews may mislead consumers. A large number of fake reviews will even cause huge property losses and public opinion crises. Therefore, it is necessary to detect and filter fake reviews. However, most existing methods have lower accuracy in detecting fake reviews due to they just use single features and lack of labeled experimental data. To solve this problem, we propose a novelty method to detect fake reviews based on multiple feature fusion and rolling collaborative training. First, the method requires an initial index system with multiple features such as text features, sentiment features of reviews and behavior features of reviewers. Second, the method needs an initial training sample set. Thus, we designed related algorithms to extract all the features of a review. Then the classification of the review is labeled manually. Finally, the method uses the initial sample set to train 7 classifiers, and the most accurate classifier will be selected to classify new reviews. The novelty of the method lies in that the features and the classification labels of the new reviews will be added into the initial sample set as new samples. So the size of the sample set will increase automatically. The experimental results in the reviews of yelp shopping website show that the accuracy of the proposed method for detecting fake reviews is 84.45%, which is 3.5% higher than the baseline methods. And compared with the latest deep learning model, its baseline precision has increased by 5.3%. According to the Friedman test, the support vector machine (SVM) classifier and random forest (RF) classifier has been proven to be the best one by statistical means. It means our method which uses multiple features has higher accuracy than the baseline models. Meanwhile, it also resolves the problem of lacking labeled training samples in fake reviews detection. INDEX TERMS Fake review detection, machine learning, multiple feature fusion, feature extraction, rolling collaborative training

“…Indeed, as documented in the scientific literature, misinformation originates often from users with questionable credibility, unknown identities and a limited number of followers (Gupta et al, 2013;Vosoughi et al, 2018), and false news become viral quickly only once they reach more popular users. In some cases, these source users may be likened or may coincide with automated spamming accounts ("bots"), which have similar features (Eshraqi et al, 2015;Masood et al, 2019;Wang, 2010).…”

Section: Experiments With Different Sources Selectionmentioning

confidence: 99%

Contrasting the Spread of Misinformation in Online Social Networks

Amoruso¹,

Anello²,

Auletta³

et al. 2020

jair

Online social networks are nowadays one of the most effective and widespread tools used to share information. In addition to being employed by individuals for communicating with friends and acquaintances, and by brands for marketing and customer service purposes, they constitute a primary source of daily news for a significant number of users. Unfortunately, besides legit news, social networks also allow to effectively spread inaccurate or even entirely fabricated ones. Also due to sensationalist claims, misinformation can spread from the original sources to a large number of users in a very short time, with negative consequences that, in extreme cases, can even put at risk public safety or health. In this work we discuss and propose methods to limit the spread of misinformation over online social networks. The issue is split in two separate sub-problems. We first aim to identify the most probable sources of the misinformation among the subset of users that have been reached by it. In the second step, assuming to know the misinformation sources, we want to locate a minimum number of monitors (that is, entities able to identify and block false information) in the network in order to prevent that the misinformation campaign reaches some “critical” nodes while maintaining low the number of nodes exposed to the infection. For each of the two issues, we provide both heuristics and mixed integer programming formulations. To verify the quality and efficiency of our suggested solutions, we conduct experiments on several real-world networks. The results of this extensive experimental phase validate our heuristics as effective tools to contrast the spread of misinformation in online social networks. Regarding the source identification step, our approach showed success rates above 80% in most of the considered settings, and above 60% in almost all of them. With respect to the second issue, our heuristic proved to be able to obtain solutions that exceeded (in terms of number of required monitors) the ones obtained through our MILP-based approach of more than 20% in only few test scenarios. Our heuristics for both problems also proved to outperform significantly some previously proposed algorithms.