[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-121391-en":3,"doc-seo-121391-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},121391,1374391974468,"Eden","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","A Machine Learning Approach to Classifying Harm Reduction Street Substance Categories - Harm Reduction-focused substance classification using ML","This study uses harm-reduction variables to categorize street drug substances, improving education, research, and overdose prevention. Statistical modeling and machine learning are applied using chemical and pharmacological properties, contextual usage characteristics, prevalence, and co-occurrence patterns from a collected sample set. Feature selection with Latent Class Analysis and Latent Profile Analysis, plus dimension reduction via UMAP and K-Means clustering, are evaluated for class separation. Interpretability remains limited by low sample size, and the approach is designed as a low-maintenance alternative to manual recategorization.","A Machine Learning Approach to Classifying Harm Reduction Street Substance Categories  \nAnuja Panthari  \nUniversity of North Carolina, Chapel Hill  \nMaster of Public Health Data Science  \nAbstract: This study aimed to use harm reduction related variables to categorize street substances in order for greater ease in education, research and prevention of overdoses. The categorization process was achieved through statistical modeling and machine learning, and included variables relating to the chemical and pharmacological properties, contextual usage characteristics, prevalence, and  \nco-occurrence of substances in our sample set. The statistical modeling approach included two featureselection methods accompanied by Latent Class Analysis (LCA) and Latent Profile Analysis (LPA) . The machine learning approach experimented with dimension reduction techniques including Uniform Manifold Approximation and Projection (UMAP) to reduce dimensions, and K-Means clustering to achieve the distinct drug classes. Each aforementioned approach revealed some level of class separation, although interpretability of classes was a challenge due to low sample size. As the street drug supply is in constant flux, manual categorization is no longer an efficient or comprehensive method to understand the supply. Our approach to modeling drug categories using machine learning, sets the stage for an on-going, low maintenance method to model categories. Follow up studies can be done using more comprehensive chemical data for the creation of more accurate substance classes using minimal harm reduction information.  \nKeywords: Harm Reduction, Street Substances, Machine Learning, Statistical Modeling, Clustering, Substance Categorization, Novel Approach to Substance Categorization.  \nIntroduction  \nThe street drug supply includes any range of substances that are unregulated by state or federal government agencies. The contents ofthe street drug supply have been known to fluctuate due to the unregulated nature of the black market, its dependency on the availability of raw materials, and the ever-changing law enforcement tactics aimed at controlling the supply. These factors pose a challenge for harm reductionists to effectively support community needs, which include overdose prevention, overdose-reversal treatments, and safe drug use practices. Rapid identification of and education about substances found within the supply is an example of a pertinent community need. In order to implement this need, an accurate yet standardized street drug classification system is required; however, no standard methodology currently exists in the literature.  \nDrug overdose is a leading cause of injury mortality in the United States, with nearly 108,000 people dying from an illicit or prescription drug-involved overdose in 2022 (CDC, 2023) . Between 2009 to 2019, the age-adjusted death rate for drug overdose increased from 11.9 per 100,000 to 21.6 (CDC, 2022). Although the CDC found that there has been a drop of drug related deaths from 114,000 in August 2023 to just under 87,000 in September 2024, the street drug supply remains vastly unknown and unpredictable (CDC, 2025). Furthermore, marginalized groups, such as Black, Indigenous, People of Color, particularly men within these groups continue to have overdose trends that have not seen a similar drop (CDC, 2025) . Trends within these groups are particularly troubling: it was seen that in just 2024, overdose death rates increased 44% for Black people and 39% for American Indian and Alaska Natives (CDC, 2025) . For these groups, it is especially pertinent to build tools like a comprehensive street drug classification system, which we will be focusing on building in this paper.  \nDrugs have been categorized for a variety of reasons historically: within political systems, to guide policies; within health services, to guide prescriptions and prevent harm; and within survey research, to assess and compare patterns (Lee, 2012) . Often, ","cbCailArVhzI3pEE","https://ap.wps.com/l/cbCailArVhzI3pEE","pdf",551373,1,13,"English","en",105,"# Introduction\n## Background and need for standardized classification\n## Drug overdose burden and equity considerations\n## Prior classification approaches and limitations\n## Study objective\n# Methods\n## Street Drug Analysis Lab data and GC/MS drug checking\n## Modeling and machine learning workflow\n# Results and Implications\n## Class separation and interpretability constraints\n## Toward low-maintenance, adaptive categorization","[{\"question\":\"What is the main goal of this study on street substances?\",\"answer\":\"To use harm-reduction related variables to categorize street drug substances, supporting education, research, and overdose prevention.\"},{\"question\":\"Which variables and techniques are used to build the substance categories?\",\"answer\":\"The study incorporates chemical and pharmacological properties, contextual usage characteristics, prevalence, and co-occurrence patterns, then applies statistical modeling (LCA/LPA) and machine learning (UMAP and K-Means).\"},{\"question\":\"Why is interpretability of the resulting classes challenging?\",\"answer\":\"Interpretability is limited by the low sample size, even though the methods show some level of class separation.\"}]","A Machine Learning Approach to Classifying Harm Reduction Street Substance Categories - Harm Reduction-focused substance classification using ML | PDF",1785735442,33,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"a-machine-learning-approach-to-classifying-harm-reduction-street-substance-categories-harm-reduction-focused-substance-classification-using-ml","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/a-machine-learning-approach-to-classifying-harm-reduction-street-substance-categories-harm-reduction-focused-substance-classification-using-ml/121391/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What is the main goal of this study on street substances?","Question",{"text":75,"@type":76},"To use harm-reduction related variables to categorize street drug substances, supporting education, research, and overdose prevention.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Which variables and techniques are used to build the substance categories?",{"text":80,"@type":76},"The study incorporates chemical and pharmacological properties, contextual usage characteristics, prevalence, and co-occurrence patterns, then applies statistical modeling (LCA/LPA) and machine learning (UMAP and K-Means).",{"name":82,"@type":73,"acceptedAnswer":83},"Why is interpretability of the resulting classes challenging?",{"text":84,"@type":76},"Interpretability is limited by the low sample size, even though the methods show some level of class separation.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]