Scholarlink

Depression can be considered a disorder of the state of mind of a person, where some can experience feelings of sadness, anger, and many others. According to Pan American Health Organization (PAHO/WHO), 50 million people in America suffer from depression disorder, and it is the main cause of mental health and disability worldwide. The World Health Organization (WHO) consider that 300 million people live with depression, a 18% increase from 2005 to 2015 [1]. This study analyzes the depression that suffered people through COVID-19 in Barranquilla, Colombia. With this objective, a dataset was built using the Beck’s Depression Inventory (BDI) which has been previously used in several studies such as [5–7]. The dataset was collected during pandemic in 2021 between April and may (moment 1), and post-pandemic in 2023 between February and march (moment 2), and both moments (1 and 2) were consolidated, to better identify the levels of depression using classification algorithms such as Decision Tree (DT), Support Vector Machine (SVM), Artificial Neuronal Networks (RNA) and Naïve Bayes (NB), these techniques were used to discover patterns useful to practitioners and patients, and after comparison, the results obtained demonstrated all algorithms were successful to recognize the different levels of depression contained in the dataset nevertheless, the DT method achieved the best result with TPR (98.1%), FPR (1.8%), Precision (98.1%) and Recall (98.1%).

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Bi-class Classification System Using Supervised Techniques for Depression Level Detection During and After Covid-19

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Fabio Mendoza Palechor,
  • Alexis De la Hoz Manotas,
  • Dionicio Neira-Rodado

摘要

Depression can be considered a disorder of the state of mind of a person, where some can experience feelings of sadness, anger, and many others. According to Pan American Health Organization (PAHO/WHO), 50 million people in America suffer from depression disorder, and it is the main cause of mental health and disability worldwide. The World Health Organization (WHO) consider that 300 million people live with depression, a 18% increase from 2005 to 2015 [1]. This study analyzes the depression that suffered people through COVID-19 in Barranquilla, Colombia. With this objective, a dataset was built using the Beck’s Depression Inventory (BDI) which has been previously used in several studies such as [5–7]. The dataset was collected during pandemic in 2021 between April and may (moment 1), and post-pandemic in 2023 between February and march (moment 2), and both moments (1 and 2) were consolidated, to better identify the levels of depression using classification algorithms such as Decision Tree (DT), Support Vector Machine (SVM), Artificial Neuronal Networks (RNA) and Naïve Bayes (NB), these techniques were used to discover patterns useful to practitioners and patients, and after comparison, the results obtained demonstrated all algorithms were successful to recognize the different levels of depression contained in the dataset nevertheless, the DT method achieved the best result with TPR (98.1%), FPR (1.8%), Precision (98.1%) and Recall (98.1%).