Case-Based Decision Support System in the Field of Tourism

The paper considers the use of case-based reasoning to support decisions for choosing a vacation destination. The use case model and algorithm for their search and extraction are defined. The case model includes a description of the features of tourist attractions, accommodation facilities, public catering establishments, and services. The features of the model have quantitative, qualitative, interval, and semantic similarity values. Scales for measurement features describing cost and distance are individual and defined by a tourist. The similarity between tourists’ reviews and preferences is assessed based on semantic similarity and sentiment calculated using the BERT multilingual neural network model. The similarity of cases is assessed based on the k-d tree (k-dimensional tree) method, which takes into account interval and text data types. The proposed approach is implemented on the Yandex.Cloud cloud platform and is tested in the Baikal natural territory using data collected from open sources. The proposed approach is implemented on the cloud platform Yandex.Cloud, and is being tested on the Baikal nature territory using data collected from open sources.