This work presents an air quality prediction in the office environment using deep neural networks. The experiment was conducted and the proposed approach showed that it gives good performance at the present time. The deep learning model is used with a combination of Transformer and GRU/LSTM models. Multivariate combination solutions, quantitative regression, and feature selection are researched and used.

Ле Ч. (науч. рук. Крылова А.А.) Improve office air quality prediction performance with Transformer and GRU/LSTM // Сборник тезисов докладов конгресса молодых ученых. Электронное издание. – СПб: Университет ИТМО, [2022]. URL: https://kmu.itmo.ru/digests/article/7938