Design and Implementation of Automatic Classification and Processing System for Library Complaints Based on Machine Learning Algorithm

Abstract

The efficient management of library complaints and feedback is essential for maintaining high-quality services and enhancing user satisfaction in library settings. In this study, they present the design and implementation of an Automatic Classification and Processing System for library complaints based on machine learning algorithms. Specifically, Support Vector Machines (SVM) and Random Forest algorithms are employed for complaint categorization. The methodology involves data collection, preprocessing, feature extraction, model selection, training, evaluation, and integration into existing library infrastructure. A diverse dataset of library complaints is utilized to train and evaluate the SVM and Random Forest models, with performance metrics including accuracy, precision, recall, and F1-score analyzed. The results demonstrate the effectiveness of both algorithms in accurately classifying library complaints, with the Random Forest algorithm exhibiting slightly superior performance in recall and F1-score values. The implications for practical deployment and considerations for algorithm selection are discussed, emphasizing the need for a balanced assessment of computational resources, interpretability, and application requirements. The Automatic Classification and Processing System offers a promising solution for streamlining complaint management processes in libraries, with the potential for further enhancements through future research endeavours.