A multi-criteria decision-making method based on discrete Z-numbers and Aczel-Alsina aggregation operators and its application on early diagnosis of depression

Abstract

In mental health diagnostics, the questionnaire is an effective and cost-effective method. However, the traditional questionnaire test methods for depression and anxiety have great ambiguity. The discrete Z-numbers (DZs) provide solutions for describing and resolving complex fuzzy issues in the intelligent multi-criteria decision-making (MCDM) process. However, large-scale datasets are not suited for the present MCDM techniques due to their extremely high computational cost. Additionally, these techniques are less stable and flexible. To address the above issues, a novel MCDM method is introduced, which is based on the DZs theory and the Aczel-Alsina (AA) aggregation operator (AO) for large-scale datasets. To begin with, centroid points are calculated for DZs, and a series of novel AOs are introduced. And then a score function with a parameter is introduced to balance the influence between the possibility restriction and the fuzzy restriction of DZs. Thirdly, a new MCDM method under DZs is presented based on the proposed AA AOs and score function. Finally, to support the early diagnosis of depression and anxiety, we apply our method to the real-life online Depression, Anxiety, and Stress Scale (DASS) which can be transformed into DZs by our proposed preprocessing method. According to experimental results, our method is applicable to large-scale datasets and has much lower complexity as well as higher flexibility and stability.