Abstract
Online social networks (OSN) are well-known platforms for exchanging various information. However, one of the existing OSN challenges is the issue of fake accounts. The attacker harnesses malicious accounts in the infected system to spread false information, such as malware, viruses, and harmful URLs. Based on the vast triumphs of deep learning in several fields, mainly automated representation, we propose RunFake, a convolutional neural network (CNN) to handle malicious account classification. We build a dynamic CNN to train a classification model instead of using regular machine learning. In particular, we create a general activation function called RunMax as a new element of the neural network's final layer. We improve accuracy in the training and testing procedure by utilizing the proposed activation layer instead of the traditional function. Based on the experimental result, our method can yield Precision = 94.00, Recall = 93.21, and F1-Score = 93.42 with a better area under curve (AUC) score = 0.9547 using user profile data as features. We harvest a promising outcome with greater accuracy with tiny loss than common learning architecture in a fake account classification problem.
Similar content being viewed by others
References
Abeer A-M, Maha H, Nada A-S, Hemalatha M (2016) Security issues in social networking sites. Int J Appl Eng Res 11(12):7672–7675
Al-Qurishi M, Al-Rakhami M, Alamri A, Alrubaian M, Rahman SMM, Hossain MS (2017) Sybil defense techniques in online social networks: a survey. IEEE Access 5:1200–1219
Al-Zoubi AM, Alqatawna J, Faris H, Hassonah MA (2021) Spam profiles detection on social networks using computational intelligence methods: the effect of the lingual context. J Inf Sci 47(1):58–81
Anglano C, Canonico M, Guazzone M (2017) Analysis of telegram messenger on android smartphones. Digital Investig 23:31–49
Baingana B, Giannakis GB (2016) Joint community and anomaly tracking in dynamic networks. IEEE Trans Signal Process 64(8):2013–2025
Bindu P, Thilagam PS (2016) Mining social networks for anomalies: Methods and challenges. J Netw Comput Appl 68:213–229
Castillo C, Mendoza M, Poblete B (2011) Information credibility on twitter. WWW Conference
Chiluka N, Andrade N, Pouwelse J, Sips H (2015) Social networks meet distributed systems: towards a robust sybil defense under churn. In: Proceedings of the 10th ACM symposium on information, computer and communications security, ASIA CCS '15, ACM: 505–518
Durst S, Zhu L (2016) The darpa twitter bot challenge
Egele M, Stringhini G, Kruegel C, Vigna G (2017) Towards detecting compromised accounts on social networks. IEEE Trans Dependable Secure Comput 14(4):447–460
He C et al (2020) CIFEF: combining implicit and explicit features for friendship inference in location-based social networks. KSEM, Texas
Jie HJ, Wanda P (2020) RunPool: a dynamic pooling layer for convolution neural network. Int J Comput Intell Syst 13(1):66–76
Kevin K, Alexander D, Matthias S (2021) Does my social media burn?—identify features for the early detection of company-related online firestorms on twitter. Online Soc Netw Media 25:100151
Kökciyan N, Yolum P (2016) ProGuard: a semantic approach to detect privacy violations in online social networks. IEEE Trans Knowl Data Eng 28(10):2724–2737
Liu B-H, Hsu Y-P, Ke W-C (2014) Virus infection control in online social networks based on probabilistic communities. Int J Commun Syst 27:4481–4491
Meier A, Johnson BK (2022) Social comparison and envy on social media: a critical review. Curr Opin Psychol 45:101302. https://doi.org/10.1016/j.copsyc.2022.101302
Mohaisen A, Hollenbeck S 2013 Improving social network-based sybil defenses by rewiring and augmenting social graphs. In: Revised selected papers of the 14th international workshop on information security applications, WISA: 8267
Muftic S, Abdullah N, Kounelis I (2016) Business information exchange system with security, privacy, and anonymity. J Electr Comput Eng 251:7093642
Nadav V, Gal-Oz N, Ehud G (2021) A trust based privacy providing model for online social networks. Online Soc Netw Media 24:100138
Ni X, Luo J, Zhang B, Teng J, Bai X, Liu Bo, Xuan D (2016) A mobile phone-based physical-social location proof system for mobile social network service. Sec Commun Netw 9:1890–1904
Prabhu Kavin B, Sagar Karki S, Hemalatha DS, Vijayalakshmi R, Thangamani M, Haleem SLA, Jose D, Tirth V, Kshirsagar PR, Adigo AG (2022) Machine learning-based secure data acquisition for fake accounts detection in future mobile communication networks. Wireless Commun Mobile Comput 2022:1
Ruan X, Wu Z, Wang H, Jajodia S (2016) Profiling online social behaviors for compromised account detection. IEEE Trans Inf Forensics Secur 11(1):176–187
Satish Kumar A, Revathy S (2022) A hybrid soft computing with big data analytics based protection and recovery strategy for security enhancement in large scale real world online social networks. Theor Comput Sci 927:15–30. https://doi.org/10.1016/j.tcs.2022.05.018
Sharma V, You I, Kumar R (2017) ISMA: intelligent sensing model for anomalies detection in cross platform OSNs with a case study on IoT. IEEE Access 5:3284–3301
Shu K, Wang S, Tang J, Zafarani R, Liu H (2017) User identity linkage across online social networks: a review. ACM SIGKDD Explor Newsl 18(2):5–17
Simon WJ, Krupnik TJ, Aguilar-Gallegos N, Halbherr L, Groot JCJ (2021) Putting social networks to practical use: improving last-mile dissemination systems for climate and market information services in developing countries. Climate Serv 23:215
Takahashi T, Panta B, Kadobayashi Y, Nakao K (2018) Web of cybersecurity Linking, locating, and discovering structured cybersecurity information. Int J Commun Syst 31:3470
Tan Z, Ning J, Liu Y, Wang X, Yang G, Yang W (2016) ECRModel: An elastic collision-based rumor-propagation model in online social networks. IEEE Access 4:6105–6120
Uppada SK, Manasa K, Vidhathri B et al (2022) Novel approaches to fake news and fake account detection in OSNs: user social engagement and visual content centric model. Soc Netw Anal Min 12:52
Vigliotti MG, Hankin C (2015) Discovery of anomalous behaviour in temporal networks. Soc Netw 41:18–25
Wanda P, Jie HJ (2020) DeepProfile: finding fake profile in online social network using dynamic CNN. J Inf Secur Appl 52:102465
Wanda P, Jie HJ (2021) DeepFriend: finding abnormal nodes in online social networks using dynamic deep learning. Soc Netw Anal Min 11:34
Wanda P, Endah MH, Jie HJ (2020) DeepOSN: Bringing deep learning as malicious detection scheme in online social network. IAES Int J Artif Intell (IJ-AI) 9(1):146
Yang Z, Dai Z, Salakhutdinov R, Cohen WW (2018) Breaking the softmax bottleneck: a high-rank RNN language model. In: International conference on learning representations. https://arxiv.org/abs/1711.03953.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix
Appendix
1.1 RunMax: complement material
To deal with the SoftMax bottleneck problem, we propose RunMax given as follows:
Definition 1
RunMax is defined as
where \({\mathcal{G}}\left( \cdot \right)\) represents a Gaussian function \(f\left( z \right) = \exp \left( { - z^{2} } \right)\) with derivative \(f^{\prime}\left( z \right) = - 2z \left( {\exp \left( { - z^{2} } \right)} \right)\)
Theorem 1
Let \(z \in S\) as the input of RunMax \(f\left( z \right)\) and SoftMax \(f_{s} \left( z \right)\). Let S as a d-dimensional vector space \(1 \in S\), thus the range of Softmax is a subset of RunMax.
Proof
If we have \(1 \in S\), it can be described as \(S = \left\{ {\mathop \sum \limits_{l = 1}^{d - 1} k^{^{\prime}\left( l \right)} u^{^{\prime}\left( l \right)} + k^{^{\prime}\left( d \right)} 1 | k^{^{\prime}\left( l \right)} \in R} \right\}\) where \(u^{^{\prime}\left( l \right)}\) is \(\left( {l = 1, \ldots ,d - 1} \right)\) and 1 are linearly independent vectors. The arbitrary part of S can be represented as \(\mathop \sum \limits_{l = 1}^{d - 1} k^{^{\prime}\left( l \right)} u^{^{\prime}\left( l \right)} + k^{^{\prime}\left( d \right)} 1\), and thus, we can write \(z = \mathop \sum \limits_{l = 1}^{d - 1} k^{^{\prime}\left( l \right)} u^{^{\prime}\left( l \right)} + k^{^{\prime}\left( d \right)} 1\). For the output of softmax,
By substituting \(z = \mathop \sum \limits_{l = 1}^{d - 1} k^{^{\prime}\left( l \right)} u^{^{\prime}\left( l \right)} + k^{^{\prime}\left( d \right)} 1\) to the \(\left[ {f_{s} \left( z \right)_{i} } \right]\) function, we have:
Thus, the range of SoftMax become as follows:
Besides, by replacing \(z = \mathop \sum \limits_{l = 1}^{d - 1} k^{^{\prime}\left( l \right)} u^{^{\prime}\left( l \right)} + k^{^{\prime}\left( d \right)} 1\) to the \(\left[ {f\left( z \right)_{i} } \right]\) RunMax function, output of RunMax becomes as follows:
When \(k^{^{\prime}\left( l \right)}\) are fixed for \(\left( {l = 1, \ldots ,d - 1} \right)\) and \(k^{^{\prime}\left( d \right)} \to + \infty\), we get the following equality:
hence \({\text{lim}}_{{{\text{k}} \to + \infty }} {\mathcal{G} }\left( {{\text{v}},{\text{k}}} \right) = 1\) when \(v\) is fixed. Considering Eq. 17, RunMax has following relation:
We can calculate \(S^{\prime} = \left\{ {\mathop \sum \limits_{l = 1}^{d - 1} k^{^{\prime}\left( l \right)} u^{^{\prime}\left( l \right)} + k^{^{\prime}\left( d \right)} 1 | k^{^{\prime}\left( l \right)} \in R for \left( {l = 1, \ldots ,d - 1, k^{^{\prime}\left( d \right)} \to + \infty } \right) \subset S} \right\}\). Based on Eq. (16), we can look that the range of RunMax contains the range of SoftMax. Therefore, we get \(\left\{ {f_{s} \left( z \right)\left| { z \in S \subseteq f\left( z \right)} \right| z \in S} \right\}\). Theorem 1 describes that if \(1 \in S\), so the range of RunMax is able to be larger than that of SoftMax. The assumption \(1 \in S\) means that there exist inputs of which outputs are the equal probabilities for all labels as \(p_{\theta } \left( {y_{i} |x} \right) = \frac{1}{M}\) for all \(i\).
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Wanda, P. RunMax: fake profile classification using novel nonlinear activation in CNN. Soc. Netw. Anal. Min. 12, 158 (2022). https://doi.org/10.1007/s13278-022-00983-9
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s13278-022-00983-9