16
views
0
recommends
+1 Recommend
1 collections
    0
    shares

      To submit your manuscript to JMIR, please click here

      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      A Machine Learning Approach for the Detection and Characterization of Illicit Drug Dealers on Instagram: Model Evaluation Study

      research-article

      Read this article at

      ScienceOpenPublisherPMC
      Bookmark
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          Background

          Social media use is now ubiquitous, but the growth in social media communications has also made it a convenient digital platform for drug dealers selling controlled substances, opioids, and other illicit drugs. Previous studies and news investigations have reported the use of popular social media platforms as conduits for opioid sales. This study uses deep learning to detect illicit drug dealing on the image and video sharing platform Instagram.

          Objective

          The aim of this study was to develop and evaluate a machine learning approach to detect Instagram posts related to illegal internet drug dealing.

          Methods

          In this paper, we describe an approach to detect drug dealers by using a deep learning model on Instagram. We collected Instagram posts using a Web scraper between July 2018 and October 2018 and then compared our deep learning model against 3 different machine learning models (eg, random forest, decision tree, and support vector machine) to assess the performance and accuracy of the model. For our deep learning model, we used the long short-term memory unit in the recurrent neural network to learn the pattern of the text of drug dealing posts. We also manually annotated all posts collected to evaluate our model performance and to characterize drug selling conversations.

          Results

          From the 12,857 posts we collected, we detected 1228 drug dealer posts comprising 267 unique users. We used cross-validation to evaluate the 4 models, with our deep learning model reaching 95% on F1 score and performing better than the other 3 models. We also found that by removing the hashtags in the text, the model had better performance. Detected posts contained hashtags related to several drugs, including the controlled substance Xanax (1078/1228, 87.78%), oxycodone/OxyContin (321/1228, 26.14%), and illicit drugs lysergic acid diethylamide (213/1228, 17.34%) and 3,4-methylenedioxy-methamphetamine (94/1228, 7.65%). We also observed the use of communication applications for suspected drug trading through user comments.

          Conclusions

          Our approach using a combination of Web scraping and deep learning was able to detect illegal online drug sellers on Instagram, with high accuracy. Despite increased scrutiny by regulators and policymakers, the Instagram platform continues to host posts from drug dealers, in violation of federal law. Further action needs to be taken to ensure the safety of social media communities and help put an end to this illicit digital channel of sourcing.

          Related collections

          Most cited references23

          • Record: found
          • Abstract: not found
          • Conference Proceedings: not found

          Recurrent neural network based language model

            Bookmark
            • Record: found
            • Abstract: not found
            • Article: not found

            Accelerating the convergence of the back-propagation method

              Bookmark
              • Record: found
              • Abstract: found
              • Article: found
              Is Open Access

              Quality of Online Pharmacies and Websites Selling Prescription Drugs: A Systematic Review

              Background Online pharmacies are companies that sell pharmaceutical preparations, including prescription-only drugs, on the Internet. Very little is known about this phenomenon because many online pharmacies operate from remote countries, where legal bases and business practices are largely inaccessible to international research. Objective The aim of the study was to perform an up-to-date and comprehensive review of the scientific literature focusing on the broader picture of online pharmacies by scanning several scientific and institutional databases, with no publication time limits. Methods We searched 4 electronic databases up to January 2011 and the gray literature on the Internet using the Google search engine and its tool Google Scholar. We also investigated the official websites of institutional agencies (World Health Organization, and US and European centers for disease control and drug regulation authorities). We focused specifically on online pharmacies offering prescription-only drugs. We decided to analyze and report only articles with original data, in order to review all the available data regarding online pharmacies and their usage. Results We selected 193 relevant articles: 76 articles with original data, and 117 articles without original data (editorials, regulation articles, or the like) including 5 reviews. The articles with original data cover samples of online pharmacies in 47 cases, online drug purchases in 13, consumer characteristics in 15, and case reports on adverse effects of online drugs in 12. The studies show that random samples with no specific limits to prescription requirements found that at least some websites sold drugs without a prescription and that an online questionnaire was a frequent tool to replace prescription. Data about geographical characteristics show that this information can be concealed in many websites. The analysis of drug offer showed that online a consumer can get virtually everything. Regarding quality of drugs, researchers very often found inappropriate packaging and labeling, whereas the chemical composition usually was not as expected in a minority of the studies’ samples. Regarding consumers, the majority of studies found that not more than 6% of the samples had bought drugs online. Conclusions Online pharmacies are an important phenomenon that is continuing to spread, despite partial regulation, due to intrinsic difficulties linked to the impalpable and evanescent nature of the Web and its global dimension. To enhance the benefits and minimize the risks of online pharmacies, a 2-level approach could be adopted. The first level should focus on policy, with laws regulating the phenomenon at an international level. The second level needs to focus on the individual. This approach should aim to increase health literacy, required for making appropriate health choices, recognizing risks and making the most of the multitude of opportunities offered by the world of medicine 2.0.
                Bookmark

                Author and article information

                Contributors
                Journal
                J Med Internet Res
                J. Med. Internet Res
                JMIR
                Journal of Medical Internet Research
                JMIR Publications (Toronto, Canada )
                1439-4456
                1438-8871
                June 2019
                15 June 2019
                : 21
                : 6
                : e13803
                Affiliations
                [1 ] Department of Healthcare Research and Policy University of California - San Diego, Extension La Jolla, CA United States
                [2 ] Global Health Policy Institute La Jolla, CA United States
                [3 ] Department of Anesthesiology University of California - San Diego, School of Medicine La Jolla, CA United States
                [4 ] Division of Infectious Disease and Global Public Health University of California - San Diego, School of Medicine La Jolla, CA United States
                Author notes
                Corresponding Author: Tim K Mackey tmackey@ 123456ucsd.edu
                Author information
                http://orcid.org/0000-0001-9801-4715
                http://orcid.org/0000-0002-4507-1094
                http://orcid.org/0000-0001-8670-6124
                http://orcid.org/0000-0002-2191-7833
                Article
                v21i6e13803
                10.2196/13803
                6598421
                31199298
                b10a2973-5727-43c1-ace9-665d280ba195
                ©Jiawei Li, Qing Xu, Neal Shah, Tim K Mackey. Originally published in the Journal of Medical Internet Research (http://www.jmir.org), 15.06.2019.

                This is an open-access article distributed under the terms of the Creative Commons Attribution License ( https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/.as well as this copyright and license information must be included.

                History
                : 22 February 2019
                : 20 March 2019
                : 12 May 2019
                : 27 May 2019
                Categories
                Original Paper
                Original Paper

                Medicine
                opioids,social media,narcotics,substance abuse,machine learning,internet,prescription drug abuse,artificial intelligence

                Comments

                Comment on this article