Analysis of the Effectiveness of Data Warehousing in Management Information Systems Using the Neural Networks Method

Abstract

Purpose: The purpose of this research is to investigate the effectiveness of data warehousing and the application of Neural Networks methods in analyzing bicycle travel app user data, with a focus on enhancing the annual membership of app users in North America.
Design/methodology/approach: This study utilizes a dataset that includes membership and usage data from relevant bicycle travel apps. It involves comparing the performance of different Neural Networks architectures, such as Feedforward Neural Networks, Convolutional Neural Networks (CNN), and other suitable models, to evaluate their effectiveness in predicting user membership.
Findings/result: The analysis results demonstrate that the implementation of Neural Networks can improve prediction accuracy, with the most effective model achieving 76.03% accuracy. The research also highlights the importance of preprocessing steps, such as data normalization and transformation, in contributing significantly to model performance. However, challenges such as overfitting were identified, suggesting the need for further testing with model and parameter variations.
Originality/value/state of the art: This research provides valuable insights for application developers and policy makers, helping them create data-driven strategies to improve the bicycle travel management information system. It also supports efforts to sustainably grow user membership. The study contributes to the field by exploring the practical application of Neural Networks for data analysis in the context of bicycle travel management, filling a gap in current research on effective predictive models for user membership growth.