@techreport{oai:ipsj.ixsq.nii.ac.jp:00241564,
 author = {Ruijia, Yan and Ruijia, Yan},
 issue = {3},
 month = {Dec},
 note = {In recent years, mental health issues like depression have become a global concern, making accurate evaluation of depression severity essential for effective treatment. Traditional scales rely on single numeric values, which may not capture the uncertainty caused by the subjective and fluctuating nature of mental health symptoms, leading to inconsistencies. This paper introduces a novel method enabling clinicians to record symptoms using heterogeneous data, including quantitative (crisp numbers, intervals) and qualitative data (linguistic terms). Additionally, a linguistic scale algorithm is employed to model the linguistic term sets, converting each symptom into a cloud model characterized by three numeric features. Aggregating these models allows for a comprehensive evaluation of both depression severity and the associated uncertainty levels. This approach enhances diagnostic accuracy and reliability in mental health care., In recent years, mental health issues like depression have become a global concern, making accurate evaluation of depression severity essential for effective treatment. Traditional scales rely on single numeric values, which may not capture the uncertainty caused by the subjective and fluctuating nature of mental health symptoms, leading to inconsistencies. This paper introduces a novel method enabling clinicians to record symptoms using heterogeneous data, including quantitative (crisp numbers, intervals) and qualitative data (linguistic terms). Additionally, a linguistic scale algorithm is employed to model the linguistic term sets, converting each symptom into a cloud model characterized by three numeric features. Aggregating these models allows for a comprehensive evaluation of both depression severity and the associated uncertainty levels. This approach enhances diagnostic accuracy and reliability in mental health care.},
 title = {Enhancing Depression Assessment with a Comprehensive Method for Heterogeneous Data},
 year = {2024}
}