Automatic multi-class classification of polish complaint reports about municipal waste management
A Dąbrowska, R Giel… - … on Dependability and …, 2021 - Springer
A Dąbrowska, R Giel, S Werbińska-Wojciechowska
International Conference on Dependability and Complex Systems, 2021•SpringerOne of the many application areas for machine learning tools is waste management, where
mainly classification of different waste materials and prediction of waste generation can be
found. The conducted literature review indicated that there is a lack of papers concerning
waste-related text classification. In this research, we investigated the automatic text
classification of complaint reports written in Polish that were sent to the municipal waste
management system operating in one of the biggest Polish cities, Wroclaw. The analyzed …
mainly classification of different waste materials and prediction of waste generation can be
found. The conducted literature review indicated that there is a lack of papers concerning
waste-related text classification. In this research, we investigated the automatic text
classification of complaint reports written in Polish that were sent to the municipal waste
management system operating in one of the biggest Polish cities, Wroclaw. The analyzed …
Abstract
One of the many application areas for machine learning tools is waste management, where mainly classification of different waste materials and prediction of waste generation can be found. The conducted literature review indicated that there is a lack of papers concerning waste-related text classification.
In this research, we investigated the automatic text classification of complaint reports written in Polish that were sent to the municipal waste management system operating in one of the biggest Polish cities, Wroclaw. The analyzed problem regards a multi-class Machine Learning classification. Waste-related vocabulary, Polish language, and unlabeled dataset are the main difficulties in the considered problem.
The results showed that the automatic classification using Machine Learning algorithms achieved higher accuracy than the standard procedure based on people’s choices currently applied in the reporting system.
Springer