Year: 2026
Type: article
Abstract: This study investigates the process that generates a personalized chatbot response based on the analysis of a user request within one interaction session. The task addressed is to increase the adaptability of the response by taking into account the parameters of the request and the related prompts. The proposed mechanism made it possible to describe the personalization process in terms of the total time spent on performing the main operations of the user and the system. The constructed model establishes a connection between the parameters of the user request, the characteristics of the temporary interaction dictionary, and the probability of receiving a personalized response. A feature of the model is the consideration of the number of possible semantic interpretations of the request and permissible options for combining words to build a logically consistent response. Based on the model, an algorithm has been developed that involves forming a temporary interaction dictionary, selecting relevant words, combining them, substituting them into the response structure, and checking the logical and semantic consistency of the result. The experiments showed that the probability of a positive completion of the personalization process increases with an increase in the number of response refinement steps, and the average time for analyzing a user's request with additional prompts and forming a personalized response is 3.056 s. It was experimentally established that with a decrease in the probability of a positive completion of the personalization process at each individual step from 10-6 to 10-8, values of the model components and the total time costs decrease. The proposed approach could be used in practical tasks of designing personalization modules for educational and consulting dialog systems, as well as for controlled generation of responses taking into account the structure of the request and the parameters of the current interaction session
Language: English
Topic: Speech and dialogue systems
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $242
OpenAlex ID: W7204990930
