Development of a Web-Based Expert System for the Diagnosis of Depressive Disorders

Abstract

The diagnosis of depression is a critical task in modern clinical psychology. However, traditional methods using paper questionnaires and manual scoring are labor-intensive and do not provide immediate feedback to patients. Existing web platforms for psychological testing are primarily focused on data collection and do not offer formalized algorithms for the automated interpretation of results according to clinically validated methods. This paper presents the development of a web-oriented expert system for diagnosing depression based on the Beck Depression Inventory. A mathematical model is proposed that formalizes the structure of the knowledge base as a set of questions with diagnostic coefficients, a user response vector, and a classification function for unambiguously mapping results to a set of diagnostic conclusions. Based on this formal model, a three-component system architecture was developed using React.js, which includes a root state management component, a component containing the expert system logic, and a static knowledge base module. The solution is implemented as a web application with an intuitive interface built using HTML5 and CSS3, along with data validation mechanisms and automated calculation of the diagnostic conclusion based on the clinically validated Beck Scale threshold values. Experimental testing on four response scenarios corresponding to different diagnostic categories confirmed the correctness of the algorithms for all intervals of the Beck Scale, including boundary values. The proposed approach provides immediate feedback to the user, increases the accessibility of screening diagnostics, and creates the prerequisites for timely seeking professional help through integration with the contact information of a specialist.