Prediction of NACE codes based on Machine Learning and Automated Processing

This study aims to predict businesses' NACE codes from textual data using a machine learning algorithm developed based on Nature Language Processing. The data used in the study is derived from textual information from various businesses. The dataset was prepared using preprocessing techniques and trained and tested with Naive Bayes, Random Forest, Support Vector Machines. The performance of the models were evaluated using accuracy and other metrics. The results show that the Random Forest algorithm is slightly more accurate and has better overall performance in terms of precision, recall, and F1-scores across more categories. The findings of this study will contribute to businesses by improving their sectoral classifications, allowing them to make more efficient and accurate decisions.