Scholarlink

Many people struggle with mental health concerns of some kind at some point in their lives, triggered by negative ideas brought on by specific life events such as divorce, grieving, work-related stress, relationships, and many more. There are a plethora of factors that contribute to mental health disorders and depression. People may occasionally harm both themselves and others with their actions. Counsellors or professionals cannot offer more effective treatment for changing clients’ perspectives if they cannot pinpoint the cause of their negative behaviours. Thus, this research addresses this issue by giving counsellors access to the cognitive distortion classification model. This research work proposes a novel framework for the automatic identification of 11 common cognitive distortions from patient’s negative thoughts (sentences) using TF-IDF, Word2Vec and Textstat features with a weighted average of Support vector machine, Random forest and LightGBM which achieves 94% accuracy. We believe this study aids Counsellors and Researchers in classifying distortions in patients.

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

Identification of Cognitive Distortions Using Weighted Ensemble Approach

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Vandana Rajput,
  • Garvin Bhati,
  • Amita Jain,
  • Shubham Tiwari,
  • Minni Jain,
  • Mahima Kapoor

摘要

Many people struggle with mental health concerns of some kind at some point in their lives, triggered by negative ideas brought on by specific life events such as divorce, grieving, work-related stress, relationships, and many more. There are a plethora of factors that contribute to mental health disorders and depression. People may occasionally harm both themselves and others with their actions. Counsellors or professionals cannot offer more effective treatment for changing clients’ perspectives if they cannot pinpoint the cause of their negative behaviours. Thus, this research addresses this issue by giving counsellors access to the cognitive distortion classification model. This research work proposes a novel framework for the automatic identification of 11 common cognitive distortions from patient’s negative thoughts (sentences) using TF-IDF, Word2Vec and Textstat features with a weighted average of Support vector machine, Random forest and LightGBM which achieves 94% accuracy. We believe this study aids Counsellors and Researchers in classifying distortions in patients.