Development of a chatbot for a car service

• Main Authors: Rubleva, Olga, Emelyanova, Elena, Mozharova, Tatiana, Polshakova, Natalya, Gavrilovskaya, Nadezhda
• Format: Conference Proceeding
• Language: English
• Subjects: Artificial neural networks; Automation; Chatbots; Data exchange; Digitization; Internet resources; Modernization; Neural networks; Process management; Recognition; Recurrent neural networks; Virtual environments
• ISSN: 0094-243X; 1551-7616
• DOI: 10.1063/5.0125264

There is an objective need to control and manage processes of data exchange between users in the virtual environment where there is electronic interaction. Therefore, automation of processing and recognition of messages is a logical and necessary stage in the further digitalization of the work of any enterprise. The purpose of the study is to develop an intelligent chat bot for automating information exchange in a car service. In this work a detailed analysis of online resources for automation of information exchange between participants of network interaction in the service market is carried out. The problem of text recognition is considered and the main stages of the solution are highlighted. In addition, neural networks for text recognition were considered: a convolutional neural network, a recurrent neural network as well as its modifications. As a result, the authors formulated a set of requirements for an intelligent chat bot and chose the language and development environment. The neural network was trained. The chatbot was tested on real data. An option for modernizing the chatbot is its testing in real conditions of information interaction in the digital environment of a car service.