Home Tsinghua Science and Technology Article
PDF (1.1 MB)
Cite
EndNote(RIS) BibTeX
Collect
Submit Manuscript
Show Outline
Figures (4)

Tables (6)
Table 1
Table 2
Table 3
Table 4
Table 5
Show 1 more tables Hide 1 tables
Open Access

Role Identification Based Method for Cyberbullying Analysis in Social Edge Computing

School of Computer Science, Fudan University, Shanghai 200433, China
Show Author Information

Abstract

Over the past few years, many efforts have been dedicated to studying cyberbullying in social edge computing devices, and most of them focus on three roles: victims, perpetrators, and bystanders. If we want to obtain a deep insight into the formation, evolution, and intervention of cyberbullying in devices at the edge of the Internet, it is necessary to explore more fine-grained roles. This paper presents a multi-level method for role feature modeling and proposes a differential evolution-assisted K-means (DEK) method to identify diverse roles. Our work aims to provide a role identification scheme for cyberbullying scenarios for social edge computing environments to alleviate the general safety issues that cyberbullying brings. The experiments on ten real-world datasets obtained from Weibo and five public datasets show that the proposed DEK outperforms the existing approaches on the method level. After clustering, we obtain nine roles and analyze the characteristics of each role and their evolution trends under different cyberbullying scenarios. Our work in this paper can be placed in devices at the edge of the Internet, leading to better real-time identification performance and adapting to the broad geographic location and high mobility of mobile devices.

Keywords

role identificationcyberbullyingsocial edge computingonline community

References

[1]

M. G. S. Murshed, C. Murphy, D. Hou, N. Khan, G. Ananthanarayanan and F. Hussain, Machine learning at the network edge: A survey, ACM Comput. Surv., vol. 54, no. 8, pp. 1–37, 2021.

Crossref Google Scholar
[2]
Mulder-Hardenberg, Edge computing in industrial automation, Mh-monitoringcontrol.com, 2022.
[3]
D. Wang, Social Edge Computing : Empowering Human-Centric Edge Computing, Learning and Intelligence, Springer Nature, 2023.
Crossref
[4]

S. Rathore, P. K. Sharma, V. Loia, Y. S. Jeong and J. H. Park, Social network security: Issues, challenges, threats, and solutions, Inf. Sci., vol. 421, pp. 43–69, 2017.

Crossref Google Scholar
[5]

H. Saini, H. Mehra, R. Rani, G. Jaiswal, A. Sharma and A. Dev, Enhancing cyberbullying detection: A comparative study of ensemble CNN-SVM and BERT models, Soc. Netw. Anal. Min., vol. 14, no. 1, p. 1, 2023.

Crossref Google Scholar
[6]

F. Shi, H. Ning, L. Chen and S. Dhelim, Cyber-syndrome: Concept, theoretical characterization, and control mechanism, Tsinghua Science and Technology, vol. 29, no. 3, pp. 721–735, 2024.

Crossref Google Scholar
[7]
D. Yang, R. E. Kraut, T. Smith, E. Mayfield and D. Jurafsky, Seekers, providers, welcomers, and storytellers: Modeling social roles in online health communities, in Proc. CHI Conference on Human Factors in Computing Systems, 2019, pp. 1−14.
Crossref
[8]
L. Sun, R. E. Kraut and D. Yang, Multi-level modeling of social roles in online micro-lending platforms, in Proc. ACM on Human-Computer Interaction, vol. 3, no. CSCW, pp. 1−25, 2019.
Crossref
[9]

T. K. Chan, C. M. Cheung and Z. W. Lee, Cyberbullying on social networking sites: A literature review and future research directions, Inf. Manag., vol. 58, no. 2, p. 103411s, 2021.

Crossref Google Scholar
[10]
R. Zhu, Research on detection methods for cyberbullying based on deep neural network, (in Chinese), master’s dissertation, Nanjing Normal University, Nanjing, China, 2021.
[11]

M. A. Al-Garadi, M. R. Hussain, N. Khan, G. Murtaza, H. F. Nweke, I. Ali, G. Mujtaba, H. Chiroma, H. A. Khattak and A. Gani, Predicting cyberbullying on social media in the Big Data Era using machine learning algorithms: Review of literature and open challenges, IEEE Access, vol. 7, pp. 70701–70718, 2019.

Crossref Google Scholar
[12]
Y. Chen, Y. Zhou, S. Zhu and H. Xu, Detecting offensive language in social media to protect adolescent online safety, in Proc. of International Conference on Privacy, Security, Risk and Trust and International Conference on Social Computing, 2012, pp. 71−80.
Crossref
[13]

M. A. Al-Garadi, K. D. Varathan and S. D. Ravana, Cybercrime detection in online communications: The experimental case of cyberbullying detection in the Twitter network, Comput. Hum. Behav., vol. 63, pp. 433–443, 2016.

Crossref Google Scholar
[14]
M. Dadvar and F. D. Jong, Cyberbullying detection: A step toward a safer internet yard, in Proc. of the 21st International Conference on World Wide Web, 2012, pp. 121−126.
Crossref
[15]

A. Perera and P. Fernando, Accurate cyberbullying detection and prevention on social media, Procedia Comput. Sci., vol. 181, pp. 605–611, 2021.

Crossref Google Scholar
[16]
K. Dinakar, R. Reichart and H. Lieberman, Modeling the detection of textual cyberbullying, in Proc. of the International AAAI Conference on Web and Social Media, 2021, pp. 11−17.
Crossref
[17]
D. Mouheb, R. Albarghash, M. F. Mowakeh, Z. Al Aghbari and I. Kamel, Detection of arabic cyberbullying on social networks using machine learning, in Proc. of IEEE/ACS International Conference on Computer Systems and Applications, 2019, pp. 1–5.
Crossref
[18]

N. Novalita, A. Herdiani, I. Lukmana and D. Puspandari, Cyberbullying identification on twitter using random forest classifier, J. Phys. Conf. Ser., vol. 1192, no. 1, p. 012029, 2019.

Crossref Google Scholar
[19]
L. Cheng, A. Mosallanezhad, Y. N. Silva, D. L. Hall and H. Liu, Mitigating bias in session-based cyberbullying detection: A non-compromising approach, in Proc. 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, 2021, pp. 2158−2168.
Crossref
[20]
T. Davidson, D. Warmsley, M. Macy and I. Weber, Automated hate speech detection and the problem of offensive language, in Proc. International AAAI Conference on Web and Social Media, 2017, pp. 512−515.
Crossref
[21]
Y. Liu, P. Zavarsky and Y. Malik, Non-linguistic features for cyberbullying detection on a social media platform using machine learning, in Proc. of Cyberspace Safety and Security : 11th International Symposium, 2019, pp. 391−406.
Crossref
[22]

V. Balakrishnan, S. Khan and H. R. Arabnia, Improving cyberbullying detection using Twitter users’ psychological features and machine learning, Comput. Secur., vol. 90, p. 101710, 2020.

Crossref Google Scholar
[23]

N. Lu, G. Wu, Z. Zhang, Y. Zheng, Y. Ren and K. K. R. Choo, Cyberbullying detection in social media text based on character-level convolutional neural network with shortcuts, Concurr. Comp. Pract. E., vol. 32, no. 23, p. e5627, 2020.

Crossref Google Scholar
[24]

D. Sultan, M. Mendes, A. Kassenkhan and O. Akylbekov, Hybrid CNN-LSTM network for cyberbullying detection on social networks using textual contents, Int. J. Adv. Comput. Sci. Appl., vol. 14, no. 9, pp. 748–756, 2023.

Crossref Google Scholar
[25]

S. Chen, J. Wang and K. He, Chinese cyberbullying detection using XLNet and deep Bi-LSTM hybrid model, Information, vol. 15, no. 2, p. 93, 2024.

Crossref Google Scholar
[26]

A. Kumar and N. Sachdeva, A Bi-GRU with attention and CapsNet hybrid model for cyberbullying detection on social media, World Wide Web, vol. 25, no. 4, pp. 1537–1550, 2022.

Crossref Google Scholar
[27]
L. Cheng, K. Shu, S. Wu, Y. N. Silva, D. L. Hall and H. Liu, Unsupervised cyberbullying detection via time-informed Gaussian mixture model, in Proc. 29th ACM International Conference on Information & Knowledge Management, 2020, pp. 185–194.
Crossref
[28]
P. Yi and A. Zubiaga, Cyberbullying detection across social media platforms via platform-aware adversarial encoding, in Proc. 16th International AAAI Conference on Web and Social Media, 2022, pp. 1430−1434.
Crossref
[29]

K. Maity, T. Sen, S. Saha and P. Bhattacharyya, MTBullyGNN: A graph neural network-based multitask framework for cyberbullying detection, IEEE Trans. Comput. Soc. Syst., vol. 11, no. 1, pp. 849–858, 2022.

Crossref Google Scholar
[30]

T. Li, Z. Zeng, Q. Li and S. Sun, Integrating GIN-based multimodal feature transformation and multi-feature combination voting for irony-aware cyberbullying detection, Inf. Process. Manag., vol. 61, no. 3, p. 103651, 2024.

Crossref Google Scholar
[31]

I. Jahnke, Dynamics of social roles in a knowledge management community, Comput. Human Behav., vol. 26, no. 4, pp. 533–546, 2010.

Crossref Google Scholar
[32]
E. Gleave, H. T. Welser, T. M. Lento and M. A. Smith, A conceptual and operational definition of ‘social role’ in online community, in Proc. 42nd Hawaii International Conference on System Sciences, 2009, pp. 1−11.
Crossref
[33]
P. Wijenayake, D. D. Silva, D. Alahakoon and S. Kirigeeganage, Automated detection of social roles in online communities using deep learning, in Proc. 3rd International Conference on Software Engineering and Information Management, 2020, pp. 63−68.
Crossref
[34]
S. H. Mukta, M. E. Ali and J. Mahmud, Deriving interpersonal role identities from social network interactions, in Proc. 6th International Conference on Networking, Systems and Security, 2019, pp. 39−47.
Crossref
[35]
J. Sun, T. Wu, Y. Jiang, R. Awalegaonkar, X. V. Lin and D. Yang, Pretty princess vs. successful leader: Gender roles in greeting card messages, in Proc. CHI Conference on Human Factors in Computing Systems, 2022, pp. 1−15.
Crossref
[36]

N. Gondal, Multiplexity as a lens to investigate the cultural meanings of interpersonal ties, Soc. Netw., vol. 68, pp. 209–217, 2022.

Crossref Google Scholar
[37]

A. Erfani, P. J. Hickey and Q. Cui, Likeability versus competence dilemma: Text mining approach using LinkedIn data, J. Manag. Eng., vol. 39, no. 3, p. 04023013, 2023.

Crossref Google Scholar
[38]
D. Yang, M. Wen and C. P. Rosé, Weakly supervised role identification in teamwork interactions, in Proc. Annual Meeting of the Association for Computational Linguistics, 2015, pp. 1671−1680.
Crossref
[39]

V. Balakrishnan, S. Khan, T. Fernandez and H. R. Arabnia, Cyberbullying detection on twitter using Big Five and Dark Triad features, Pers. Individ. Differ., vol. 141, pp. 252–257, 2019.

Crossref Google Scholar
[40]

G. Jacobs, C. V. Hee and V. Hoste, Automatic classification of participant roles in cyberbullying: Can we detect victims, bullies, and bystanders in social media text?, Nat. Lang. Eng., vol. 28, no. 2, pp. 141–166, 2020.

Crossref Google Scholar
[41]

M. Mohd, F. Fauzi and W. N. H. W. Ali, Identification of profane words in cyberbullying incidents within social networks, J. Inf. Sci. Theory Pract., vol. 9, no. 1, pp. 24–34, 2021.

Google Scholar
[42]
M. D. Capua, E. D. Nardo and A. Petrosino, Unsupervised cyber bullying detection in social networks, in Proc. 23rd International Conference on Pattern Recognition, 2016, pp. 432−437.
Crossref
[43]

A. Bozyiit, S. Utku and E. Nasibov, Cyberbullying detection: Utilizing social media features, Expert Syst. Appl., vol. 179, p. 115001, 2021.

Crossref Google Scholar
[44]
V. K. Singh, S. Ghosh and C. Jose, Toward multimodal cyberbullying detection, in Proc. CHI Conference Extended Abstracts on Human Factors in Computing Systems, 2017, pp. 2090−2099.
Crossref
[45]
R. Zhao, A. Zhou and K. Mao, Automatic detection of cyberbullying on social networks based on bullying features, in Proc. 17th International Conference on Distributed Computing and Networking, 2016, pp. 1−6.
Crossref
[46]
AllSlang, List of swear words, bad words, & curse words, https://www.noswearing.com/dictionary, 2023.
[47]
S. Sood, J. Antin and E. Churchill, Profanity use in online communities, in Proc. of the SIGCHI Conference on Human Factors in Computing Systems, 2012, pp. 1481−1490.
Crossref
[48]
Tencent AI Lab, Tencent AI Lab embedding corpora for Chinese and English words and phrases, (in Chinese), https://ai.tencent.com/ailab/nlp/en/embedding.html, 2022.
[49]
M. Dadvar, D. Trieschnigg, R. Ordelman and F. De Jong, Improving cyberbullying detection with user context, in Proc. Advances in Information Retrieval : 35th European Conference on IR Research, 2013, pp. 693−696.
Crossref
[50]
J. O. Atoum, Detecting Cyberbullying from Tweets through machine learning techniques with sentiment analysis, in Proc. Future of Information and Communication Conference, 2023, pp. 25−38.
Crossref
[51]
V. Nahar, S. Unankard, X. Li and C. Pang, Sentiment analysis for effective detection of cyber bullying, in Proc. 14th Asia-Pacific Web Conference, 2012, pp. 767−774.
Crossref
[52]
Baidu, Inc., BAIDU AI CLOUD, (in Chinese), https://intl.cloud.baidu.com/, 2022.
[53]
Y. J. Foong and M. Oussalah, Cyberbullying system detection and analysis, in Proc. European Intelligence and Security Informatics Conference, 2017, pp. 40−46.
Crossref
[54]
DUTIR, ZaneMuir/DLUT-Emotionontology, (in Chinese), https://github.com/zanemuir/dlut-emotionontology, 2017.
[55]
R. P. Jacob, K. Manoj, D. Mohan, S. Issac and D. Sudarsan, Cyberbullying detection and prevention using artificial intelligence, in Proc. of the International Conference on Soft Computing for Security Applications, 2022, pp. 905−914.
Crossref
[56]
P. J. Lee, Y. H. Hu, K. Chen, J. M. Tarn and L. E. Cheng, Cyberbullying detection on social network services, in Proc. 22nd Pacific Asia Conference on Information Systems, 2018, vol. 61, p. 2018.
[57]
Z. Wahid and A. Al Imran, Multi-feature transformer for multiclass cyberbullying detection in Bangla, in Proc. IFIP International Conference on Artificial Intelligence Applications and Innovations, 2023, pp. 439−451.
Crossref
[58]
M. Raphel, P. Parvathi, R. Y. Hashim, R. J. Thevara and P. D. Varma, Analysing gender and age aspects of cyberbullying through online social media, in Proc. International Conference on Advances in Computing and Communications, 2021, pp. 1−8.
Crossref
[59]

C. P. Barlett, Anonymously hurting others online: The effect of anonymity on cyberbullying frequency, Psychol. Pop. Media. Cult., vol. 4, no. 2, pp. 70–79, 2015.

Crossref Google Scholar
[60]

V. Balakrishnan, Cyberbullying among young adults in Malaysia: The roles of gender, age and Internet frequency, Comput. Hum. Behav., vol. 46, pp. 149–157, 2015.

Crossref Google Scholar
[61]
L. Cheng, J. Li, Y. N. Silva, D. L. Hall and H. Liu, Xbully: Cyberbullying detection within a multi-modal context, in Proc. 12th ACM International Conference on Web Search and Data Mining, 2019, pp. 339−347.
Crossref
[62]

M. Arif, A systematic review of machine learning algorithms in cyberbullying detection: Future directions and challenges, J. Inf. Secur. Cybercrimes Res., vol. 4, no. 1, pp. 1–26, 2021.

Crossref Google Scholar
[63]

M. F. López-Vizcaíno, F. J. Nóvoa, V. Carneiro and F. Cacheda, Early detection of cyberbullying on social media networks, Future Gener. Comput. Syst., vol. 118, pp. 219–229, 2021.

Crossref Google Scholar
[64]
R. I. Rafiq, H. Hosseinmardi, R. Han, Q. Lv and S. Mishra, Scalable and timely detection of cyberbullying in online social networks, in Proc. 33rd Annual ACM Symposium on Applied Computing, 2018, pp. 1738−1747.
Crossref
[65]
D. Arthur and S. Vassilvitskii, K-means++ the advantages of careful seeding, in Proc. 18th Annual ACM-SIAM Symposium on Discrete Algorithms, 2007, pp. 1027−1035.
[66]

R. J. Kuo, Y. Zheng and T. P. Q. Nguyen, Metaheuristic-based possibilistic fuzzy k-modes algorithms for categorical data clustering, Inf. Sci., vol. 557, pp. 1–15, 2021.

Crossref Google Scholar
[67]

C. Tîrnăucă, D. Gómez-Pérez, J. L. Balcázar and J. L. Montaña, Global optimality in k-means clustering, Inf. Sci., vol. 439, pp. 79–94, 2018.

Crossref Google Scholar
[68]

R. Storn and K. Price, Differential evolution-a simple and efficient heuristic for global optimization over continuous spaces, J. Glob. Optim., vol. 11, no. 4, pp. 341–359, 1997.

Crossref Google Scholar
[69]

Y. Zhou, H. Wu, Q. Luo, and M. Abdel-Baset, Automatic data clustering using nature-inspired symbiotic organism search algorithm, Knowl. Based. Syst., vol. 163, pp. 546–557, 2019.

Crossref Google Scholar
[70]

J. C. Gower, A general coefficient of similarity and some of its properties, Biometrics, vol. 27, no. 4, pp. 857–871, 1971.

Crossref Google Scholar
[71]

H. Y. Du and W. J. Wang, A clustering ensemble framework with integration of data characteristics and structure information: A graph neural networks approach, Mathematics, vol. 10, no. 11, p. 1834, 2022.

Crossref Google Scholar
[72]
J. MacQueen, Some methods for classification and analysis of multivariate observations, in Proc. 5th Berkeley Symposium on Mathematical Statistics and Probability, 1967, pp. 281−297.
[73]

J. C. Bezdek, R. Ehrlich and W. Full, FCM: The fuzzy c-means clustering algorithm, Comput. Geosci., vol. 10, nos. 2&3, pp. 191–203, 1984.

Crossref Google Scholar
[74]
F. Nie, C. L. Wang and X. Li, K-multiple-means: A multiple-means clustering method with specified k clusters, in Proc. of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2019, pp. 959−967.
Crossref
[75]

B. Schölkopf, A. Smola and K. R. Müller, Nonlinear component analysis as a kernel eigenvalue problem, Neural. Comput., vol. 10, no. 5, pp. 1299–1319, 1998.

Crossref Google Scholar
[76]
G. J. McLachlan and K. E. Basford, Mixture Models : Inference and Applications to Clustering. Marcel Dekker, 1988.
[77]

Murtagh and P. Contreras, Algorithms for hierarchical clustering: an overview, Wiley Interdiscip. Rev. Data Min. Knowl. Discov., vol. 2, no. 1, pp. 86–97, 2012.

Crossref Google Scholar
[78]
M. Kelly, R. Longjohn and K. Nottingham, UCI Machine Learning Repository, http://archive.ics.uci.edu/ml, 2023.
[79]

J. Derrac, S. Garcia, L. Sanchez and F. Herrera, Keel data-mining software tool: Data set repository, integration of algorithms and experimental analysis framework, J. Mult. Valued Log. Soft Comput., vol. 17, pp. 255–287, 2015.

Google Scholar
[80]

D. Olweus and S. P. Limber, Bullying in school: Evaluation and dissemination of the Olweus bullying prevention program, Am. J. Orthopsychiatry., vol. 80, no. 1, pp. 124–134, 2010.

Crossref Google Scholar
[81]

R. M. Kowalski and S. P. Limber, Electronic bullying among middle school students, J. Adolesc. Health., vol. 41, no. 6, pp. S22–S30, 2007.

Crossref Google Scholar
[82]

E. Tiiri, T. Luntamo, K. Mishina, L. Sillanmäki, A. B. Klomek and A. Sourander, Did bullying victimization decrease after nationwide school-based antibullying program? A time-trend study, J. Am. Acad. Child Adolesc. Psychiatry., vol. 59, no. 4, pp. 531–540, 2020.

Crossref Google Scholar
[83]

I. S. Mahmood, Are cyberbullying interventions and criminal law prevention effective? (A review of cyberbullying legislation in Iraq), J. Arch. Egyptol., vol. 17, no. 7, pp. 16983–16998, 2020.

Google Scholar
[84]

J. Bailey, Canadian legal approaches to ‘cyberbullying’ and cyberviolence: An overview, Ottawa Faculty of Law Working Paper, no. 2016-37, 2016.

Crossref Google Scholar
[85]
H. W. Bierhoff, R. L. Cohen and J. Greenberg, Justice in Social Relations. Springer Science & Business Media, 2013.
[86]
M. L. Hoffman, Empathy and Moral Development : Implications for Caring and Justice. Cambridge University Press, 2001.
Crossref
[87]
S. Chen, Consensus research in network collective action, (in Chinese), masters’s dissertation, Harbin Institute of Technology, Harbin, China, 2015.
[88]

N. Eisenberg, Emotion, regulation, and moral development, Annu. Rev. Psychol., vol. 51, no. 1, pp. 665–697, 2000.

Crossref Google Scholar
[89]

H. Landmann and U. Hess, What elicits third-party anger? The effects of moral violation and others’ outcome on anger and compassion, Cogn. Emot., vol. 31, no. 6, pp. 1097–1111, 2017.

Crossref Google Scholar
[90]
X. Fan, The communication interpretation of the collapse of persona, (in Chinese), http://media.people.com.cn/n1/2018/0504/c419419-29965157.html, 2018.
Tsinghua Science and Technology
Volume 30 Issue 4,
August 2025
Pages 1659-1684
DOI: 10.26599/TST.2024.9010066
Cite this article:
Wang R, Lu T, Zhang P, et al. Role Identification Based Method for Cyberbullying Analysis in Social Edge Computing. Tsinghua Science and Technology, 2025, 30(4): 1659-1684. https://doi.org/10.26599/TST.2024.9010066
Metrics & Citations  
Article History
Copyright
Rights and Permissions
Return

Algorithm 1　Framework of DEK
1: Initialization phase:
　 D={d1,d2,…,dn}, the original clustering dataset; K, the 　 clustering number; MaxGs, the maximum number of 　 iterations; Np, the population size; F, the scaling factor; Cr, 　 the crossover probability.
2: G=1;
3: Data continuous operation;
4: Generate initial population P1={x11,x21,…,xNp1} 　 randomly;
5: repeat
6:　 for xiG in PG do
7:　　 Find the optimal individual xbestG;
8:　　 Use mutation operators to generate mutation vectors viG;
9:　　 Perform the crossover operation on the target vector xiG 　　　 and the mutation vector viG using a crossover strategy to 　　　 produce a trial vector uiG;
10:　　 Perform the dimensionality reduction on the target 　　　 vector xiG and the trial vector uiG;
11:　　 Evaluate the objective function values of the target 　　　 vector xiG and the trial vector uiG by the objective 　　　 function;
12:　　 Compare the objective function values of the target 　　　 vector xiG and the trial vector uiG, and the better vector 　　　 is stored in the next generation population;
13:　　 Update the optimal vector;
14: 　end for
15: until Gs⩽MaxGs
16: Gs=Gs+1;
Output: Different clusters and their vectors.

Parameter settings for DEK.

ParameterValue
Np, the population size60
F, the scaling factor0.7
Cr, the crossover probability0.8
MaxGs, the maximum number of iterations1500

Descriptive statistics of the general datasets.

DatasetNumber Dcon Dcat# choices in Dcat
Australian69068[2 3 14 9 2 2 2 3]
Diagnosis12015[2 2 2 2 2]
Heart27085[2 4 3 2 3]
German1000713[4 5 10 5 5 4 3 4 3 3 4 2 2]
SouthGermanCredit1000713[4 5 11 5 5 4 3 4 3 3 4 2 2]

Comparison of DEK, K-means (KM), FCM, K-means++ (KM++), K-Multiple-Means (KMM), Kernel K-means (KKM), GMM, and hierarchical clustering (HC) on the general dataset. The bold values represent the best performances.

DatasetMetricsDEKKMFCMKM++KMMKKMGMMHC
AustralianDBI1.43361.98012.15501.99731.65482.02542.07391.8834
SC0.21070.18800.19220.18740.07760.16080.13440.1951
DVI0.11230.15240.08740.16520.16320.12750.1470.3495
DiagnosisDBI0.39471.24651.45771.23361.08091.24000.86440.8667
SC0.67960.42210.51910.42920.50020.43770.62720.5764
DVI0.41670.00990.00960.00930.13790.10700.33980.3600
HeartDBI0.89732.34202.34292.34321.75212.50811.97690.9387
SC0.26530.05090.05110.05070.25110.00700.23570.0920
DVI0.26800.02560.02320.02560.08230.01340.13290.2258
GermanDBI1.46692.02672.10472.03412.05312.01162.02391.9939
SC0.37100.12990.11580.12900.12600.12720.13380.1940
DVI0.28840.10380.08750.10170.09680.09730.14610.2426
SouthGermanCreditDBI0.97432.05182.11922.02592.09422.03462.04140.9754
SC0.25290.12010.11740.12730.12590.12490.10880.2229
DVI0.21600.08150.06920.08420.08420.08330.10640.3694

Descriptive statistics of the real-world datasets from ten scenarios.

Dataset# User# Comment Dcon Dcat# Choices in Dcat
Scenario 1272167179174[3 7 2 2]
Scenario 2354210682174[3 7 2 2]
Scenario 3664306067174[3 7 2 2]
Scenario 426064157174[3 7 2 2]
Scenario 526653549174[3 7 2 2]
Scenario 6405218451174[3 7 2 2]
Scenario 729457509174[3 7 2 2]
Scenario 8286152250174[3 7 2 2]
Scenario 9224139586174[3 7 2 2]
Scenario 10206103591174[3 7 2 2]
Total32311473021174[3 7 2 2]

Comparison of DEK, K-means (KM), FCM, K-means++ (KM++), K-Multiple-Means (KMM), Kernel K-means (KKM), GMM, and Hierarchical Clustering (HC) on the real-world datasets. The bold values represent the best performances.

MethodMetricScenario 1Scenario 2Scenario 3Scenario 4Scenario 5Scenario 6Scenario 7Scenario 8Scenario 9Scenario 10
DEKDBI1.06001.26331.13751.23091.13861.90291.41881.49131.09671.2019
SC0.45590.31530.50570.38430.43930.37030.42280.36330.43330.4285
DVI0.03590.02650.03600.04820.02150.01900.08070.03730.02660.0556
KMDBI1.87041.13661.47151.65541.49311.42841.68091.76331.50381.6704
SC−0.06960.26200.0166−0.04860.0781−0.01560.0394−0.0721−0.0946−0.0046
DVI0.00100.04140.00500.01290.00420.00020.01670.00250.00170.0041
FCMDBI1.88701.54041.80341.67951.61021.40082.10852.13711.56002.7799
SC−0.02660.1091−0.0127−0.09100.0174−0.03050.0158−0.0413−0.0586−0.0309
DVI0.00070.00680.00100.01140.00150.00020.00730.00180.00130.0044
KM++DBI1.98201.31081.70691.78801.61841.70451.92172.00081.60171.7079
SC−0.05060.2397−0.0123−0.03140.0456−0.01580.0492−0.0407−0.0180−0.0062
DVI0.00080.03020.00260.01130.00340.00020.01310.00180.00120.0040
KMMDBI2.16562.08602.04332.13122.13562.14111.95242.12052.09392.1349
SC−0.0790−0.0316−0.0348−0.0613−0.0470−0.07390.0809−0.0884−0.0649−0.0138
DVI0.00070.00120.00070.00510.00050.00000.01600.00160.00040.0029
KKMDBI2.43652.39682.30302.43602.34292.28852.34282.40872.35162.3728
SC−0.0672−0.0512−0.0595−0.0635−0.0534−0.0725−0.0404−0.0625−0.0712−0.0627
DVI0.00000.00080.00030.00350.00020.00000.00150.00140.00010.0017
GMMDBI1.47141.67731.69981.55241.61771.56441.69251.75011.60481.6212
SC0.30350.19510.23560.21370.20480.21320.28660.21710.21790.2327
DVI0.01500.00750.01540.01810.00660.00640.02950.02430.00740.0254
HCDBI1.25280.88381.15461.41441.20281.17421.05721.36551.19141.5105
SC−0.36570.11800.2137−0.3512−0.00070.0493−0.0634−0.1113−0.2659−0.0810
DVI0.01210.05320.03460.03010.01030.01110.05900.01910.00730.0338

Comparison of algorithm performance by metric.

MethodMetric# Best performance according to the metrics
DEKDBI11
SC15
DVI12
HCDBI4
SC0
DVI3