Detecting substance-related problems in narrative investigation summaries of child abuse and neglect using text mining and machine learning |
| |
| Institution: | 1. Child and Adolescent Data Lab, University of Michigan, School of Social Work, 1080 S University Ave, Ann Arbor, MI, 48109, United States;2. Indiana University School of Social Work, 902 West New York Street Indianapolis, Indiana, 46202, United States;3. University of Michigan, School of Information, 105 S State St, Ann Arbor, MI, 48109, United States;1. The Ohio State University College of Medicine, 370 West 9th Avenue, Columbus, OH 43210, USA;2. Center for Pediatric Trauma Research, Abigail Wexner Research Institute at Nationwide Children’s Hospital, 700 Children’s Drive, Columbus, OH 43205, USA;3. Center for Injury Research and Policy, Abigail Wexner Research Institute at Nationwide Children’s Hospital, 700 Children’s Drive, Columbus, OH 43205, USA;4. The Center for Family Safety and Healing, Nationwide Children''s Hospital, 700 Children’s Drive, Columbus, OH 43205, USA;5. Trauma Program, Nationwide Children’s Hospital, 700 Children’s Drive, Columbus, OH 43205, USA;1. School of Psychology, Trinity College Dublin, Ireland;2. Health Service Executive, Cork, Ireland;1. Department of Nursing, National Cheng Kung University Hospital, College of Medicine, National Cheng Kung University, Tainan, Taiwan;2. Department of Nursing, College of Medicine, National Cheng Kung University, Tainan, Taiwan;3. International Doctoral Program in Nursing, Department of Nursing, College of Medicine, National Cheng Kung University, Tainan, Taiwan;1. Department of Clinical Child and Family Studies, VU University Amsterdam, the Netherlands;2. Department of Innovation and Research, Children’s and Youth Trauma Center, KJTC Kenter Youth Care, Haarlem, the Netherlands;3. Department of Epidemiology, Health Promotion, and Healthcare Innovations, Public Health Service, Amsterdam, the Netherlands;4. Hogrefe Publishers, Scientific Publisher for Psychology, Psychiatry, and Mental Health, Amsterdam, the Netherlands;1. Faculty of Social Sciences, University of Helsinki, P.O. Box 18, FI-00014, Helsinki, Finland;2. National Institute for Health and Welfare, THL, P.O. Box 310, FI-90101, Oulu, Finland;3. Finnish Youth Research Network, Asemapäällikönkatu 1, FI-00520, Helsinki, Finland;4. Institute of Clinical Medicine, University of Eastern Finland, P.O. Box 1627 (Mediteknia building), FI-70211, Kuopio, Finland |
| |
| Abstract: | BackgroundState child welfare agencies collect, store, and manage vast amounts of data. However, they often do not have the right data, or the data is problematic or difficult to inform strategies to improve services and system processes. Considerable resources are required to read and code these text data. Data science and text mining offer potentially efficient and cost-effective strategies for maximizing the value of these data.ObjectiveThe current study tests the feasibility of using text mining for extracting information from unstructured text to better understand substance-related problems among families investigated for abuse or neglect.MethodA state child welfare agency provided written summaries from investigations of child abuse and neglect. Expert human reviewers coded 2956 investigation summaries based on whether the caseworker observed a substance-related problem. These coded documents were used to develop, train, and validate computer models that could perform the coding on an automated basis.ResultsA set of computer models achieved greater than 90% accuracy when judged against expert human reviewers. Fleiss kappa estimates among computer models and expert human reviewers exceeded .80, indicating that expert human reviewer ratings are exchangeable with the computer models.ConclusionThese results provide compelling evidence that text mining procedures can be a cost-effective and efficient solution for extracting meaningful insights from unstructured text data. Additional research is necessary to understand how to extract the actionable insights from these under-utilized stores of data in child welfare. |
| |
| Keywords: | Text mining Machine learning Data science Text classification Child welfare Substance misuse |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
