Analysis of Drug Addicts Charasteristics Using a Hybrid Learning Approach
Authors
-
Hoda Mashayekhi 1
1. School of Computer Engineering, Shahrood University of Technology, Shahrood, Iran.
http://orcid.org/0000-0003-0080-2743
- Mehdi Khaksari 2 2. School of Medicine, Shahroud University of Medical Sciences, Shahroud, Iran.
DOI:
https://doi.org/10.22100/jkh.v13i2.1942Keywords:
Analysis of drug addicts charasteristics, Hybrid learning, Clustering, Classification, Correlation analysis, Addiction prevention and rehabilitation.Abstract
Introduction: Refinement and targeting of addiction prevention and rehabilitation programs, will increase the efficiency of executing such programs. However, the role of learning and data mining methods are rarely studied in this area. The aim of this paper is to assess the performance of data mining methods in analysis, refinement and grouping of the community for targeting the programs.
Methods: In this paper, first an analysis framework is introduced. The goal of this framework is indicating the features influencing behavior of the target community, and also deriving different groups from the target community along with specific characteristics of each group. This way, targeted programs can be executed in each group. The information of addicts referring to a number of Methadone maintenance treatment (MMT) centers in Shahrood, Iran are analyzed with a range of data mining techniques as guided by the proposed framework. To indicate the influential features, first the feature selection step (using a hybrid of correlation analysis, association rule mining and decision tree analysis) is performed. Next, the data is clustered considering the selected features. The result of clustering is extraction of different groups from the target community. Finally, a hybrid of association rule mining and decision tree algorithms are used to analyze data in each cluster and determine their characteristics.
Results: The results show the effectiveness of the proposed approach in analysing and grouping the community. Specifically, we focus on the reason of referring to a MMT center and indicate influential features in this context. Next, leveraging the extracted features, we derived various groups from the target community and determine their characteristics by deriving rules. The confidence of the extracted rules is at least 84%.
Conclusion: The results of this research insist on the significant role of the data mining techniques in improved execution of prevention and rehabilitation programs. Specifically, the relation of the addicts’ family and social conditions with the reason of referring to MMT centers, which are explored in the paper, can be effective in the treatment programs of drug addicts and prevention of similar people becoming a consumer.
References
2016 World Drug Report. The United Nations Office on Drugs and Crime (UNODC) Web site. http://www.unodc.org/wdr2016/. Accessed December 23, 2017.
Aman Abad I, Salimi AmanAbad M. A Relationship between the Youth Life Style and Attitude to Artificial Drugs Consumption in Shirvan. Sociological Studies Of Youth 2012;3(6):57-70.[Persian].
García EG, Blasco BC, López RJ, Pol AP. Study of the factors associated with substance use in adolescence using Association Rules. Adicciones 2010;22(4):293-9.
Nikkhou MR, Avadis Yans H. The handbook of clinical adult psychology. Tehran:Bikaran;1998.[Persian].
Tavakoli Ghouchani H, Armat MR, Mortazavi H. Factors related to treating addiction in those referring to addiction treatment clinics. Journal of Zanjan University of Medical Sciences 2002;10:32-37.[Persian].
Steensma C, Boivin JF, Blais L, Roy É. Cessation of injecting drug use among street-based youth. J Urban Health 2005;82(4):622-37. doi:10.1093/jurban/jti121
Han J, Pei J, Kamber M. Data mining concepts and techniques. 3rd ed. Morgan Kaufmann;2011.
Mashayekhi H, Nourollah Z. Proposing an educating and notifying targeted framework for improvement of Police services. Police Organizational Development 2016;12(52):109-26.[Persian].
Yun L, Xiang-sheng L. The data mining and knowledge discovery in biomedicine. Proceeding of 5th International Conference on Computer Science and Education (ICCSE);2010 August 24-27;Hefei,China.IEEE;2010.p.1050-52. doi:10.1109/ICCSE.2010.5593411
Harper G, Pickett SD. Methods for mining HTS data. Drug Discovery Today 2006;11(15-16):694-99. doi:10.1016/j.drudis.2006.06.006
Gervilla E, Cajal B, Palmer A. Quantification of the influence of friends and antisocial behaviour in adolescent consumption of cannabis using the ZINB model and data mining. Addict Behav 2011;36(4):368-74. doi:10.1016/j.addbeh.2010.12.007
Sun J, Bi J, Chan G, Oslin D, Farrer L, Gelernter J, Kranzler HR. Improved methods to identify stable, highly heritable subtypes of opioid use and related behaviors. Addictive Behaviors 2012;37(10):1138-44. doi:10.1016/j.addbeh.2012.05.010
Pal S, Chaurasia V. Performance Analysis of Students Consuming Alcohol Using Data Mining Techniques. International Journal of Advance Research in Science and Engineering 2017;6(2):238-50.
Kim SJ, Marsch LA, Hancock JT, Das AK. Scaling up research on drug abuse and addiction through social media big data. J Med Internet Res 2017;19(10):e353. doi:10.2196/jmir.6426
García-Martínez R, Martins S, Bianco S, Navas H. Discovery of psychoactive substance addiction patterns based on information mining engineering. Studies in health technology and informatics 2017;245:1282.
Zahedi F, Zare-Mirakabad MR. Employing data mining to explore association rules in drug addicts. Journal of AI and Data Mining 2014;2(2),135-9. doi:10.22044/JADM.2014.308
Liao SH, Chu PH, Hsiao PY. Data mining techniques and applications–A decade review from 2000 to 2011. Expert Systems with Applications 2012;39(12):11303-11. doi:10.1016/j.eswa.2012.02.063
Hall MA. Correlation-based feature subset selection for machine learning. Hamilton:University of Waikato;1998.
Guyon I, Elisseeff A. An introduction to variable and feature selection. Journal of machine learning research 2003;3:1157-82.
Quinlan R. C4.5: Programs for Machine Learning. San Mateo,California: Morgan Kaufmann; 1993. 74 p.
Agrawal R, Srikant R. Mining sequential patterns. Proceeding of 11th International Conference on Data Engineering;1995 March 06-10;Washington,DC,USA.IEEE;1995.p.3-14. doi:10.1109/ICDE.1995.380415
Ester M, Kriegel HP, Sander J, Xu X. A density-based algorithm for discovering clusters in large spatial databases with noise. Proceeding of 2nd International Conference on Knowledge Discovery and Data Mining;1996 August 2-4;Portland,Oregon,USA.AAAI Press;1996.p.226-31.
Dempster AP, Laird NM, Rubin DB. Maximum likelihood from incomplete data via the EM algorithm. Journal of the royal statistical society 1977;Series B (methodological):1-38.
Hall M, Frank E, Holmes G, Pfahringer B, Reutemann P, Witten IH. The WEKA data mining software: an update. ACM SIGKDD explorations newsletter 2009;11(1):10-18. doi:10.1145/1656274.1656278
Fathtabar S, Asadollahi A, Haji Azizi S, Arabpour A, Havasi A. Study of social and cultural characteristics of drug offenders in ahwaz municipal areas using geographic information system (GIS). Quarterly Journal of Research on Addiction 2016;9(36):125-40.[Persian].
Shahmoradi A, Mohammadifar MA, Fakhri MK. The effect of methadone maintenance treatment on health psyche and having hope in drug addicted persons. Quarterly Journal of Research on Addiction 2011;4(16):73-84.[Persian].
Abdollahpour M, Golzari M. Investigation of the effect of life skill training on attitude towards substance abuse in sirjan city high school students. Quarterly Journal of Research on Addiction 2009;2(8):33-54.[Persian].
Hashemi T, Gasem Baklu Y, Mohammadzadeghan R, Vakili S, Irani M. On the comparison of public health and social support in addicts and non-addicts. Research on Addiction 2014;8(31):119-33.[Persian]
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
