<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>In recent years, the burgeoning interest of enterprises in the e-commerce sector has underscored the necessity for effective analytical tools that monitor user engagement on agricultural products’ e-commerce platforms and provide precise market forecasts. To address this need, the current study proposes an analytical framework and develops a comprehensive market prediction system for agricultural products. Initially, behavioral analysis indices on the e-commerce platform are employed to delineate user behavioral patterns, which include traffic and purchase path analyses. Based on a case study, enhancements are made to the conventional RFM (Recency, Frequency, Monetary) customer behavior model, both strategically and structurally, thus enriching the analysis of user behavior on such platforms. Subsequently, a four-tiered e-commerce market prediction system is established, incorporating a multilayered deep learning network designed to anticipate market trend shifts. The analysis reveals that the application of deviation standardization processing significantly refines the representation of user behavior to the platform’s merchants. Moreover, the implementation of this market prediction system not only reduces processing time from 8 seconds to 4 seconds but also decreases the average relative error by 0.31%. These improvements highlight the system’s enhanced predictive accuracy, confirming its utility in navigating the complexities of the agricultural products market in the e-commerce domain.</p>
</abstract>

In recent years, the burgeoning interest of enterprises in the e-commerce sector has underscored the necessity for effective analytical tools that monitor user engagement on agricultural products’ e-commerce platforms and provide precise market forecasts. To address this need, the current study proposes an analytical framework and develops a comprehensive market prediction system for agricultural products. Initially, behavioral analysis indices on the e-commerce platform are employed to delineate user behavioral patterns, which include traffic and purchase path analyses. Based on a case study, enhancements are made to the conventional RFM (Recency, Frequency, Monetary) customer behavior model, both strategically and structurally, thus enriching the analysis of user behavior on such platforms. Subsequently, a four-tiered e-commerce market prediction system is established, incorporating a multilayered deep learning network designed to anticipate market trend shifts. The analysis reveals that the application of deviation standardization processing significantly refines the representation of user behavior to the platform’s merchants. Moreover, the implementation of this market prediction system not only reduces processing time from 8 seconds to 4 seconds but also decreases the average relative error by 0.31%. These improvements highlight the system’s enhanced predictive accuracy, confirming its utility in navigating the complexities of the agricultural products market in the e-commerce domain.

User Behavior Analysis and Market Forecasting of Agricultural E-commerce Platforms

College of Business Administration, Lishui Vocational and Technical College

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

In recent years, the burgeoning interest of enterprises in the e-commerce sector has underscored the necessity for effective analytical tools that monitor user...

User Behavior Analysis and Market Forecasting of Agricultural E-commerce Platforms

Publié en ligne: 05 août 2024

Pages: -

Reçu: 21 mars 2024

Accepté: 17 juin 2024

DOI: https://doi.org/10.2478/amns-2024-2013

Mots clés
User behavior analysis, RFM model, Deep learning, Market prediction

© 2024 Hongjing Zhu., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

User Behavior Analysis and Market Forecasting of Agricultural E-commerce Platforms

Publié en ligne: 05 août 2024

Pages: -

Reçu: 21 mars 2024

Accepté: 17 juin 2024

DOI: https://doi.org/10.2478/amns-2024-2013

Mots clésUser behavior analysis, RFM model, Deep learning, Market prediction

© 2024 Hongjing Zhu., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Mots clés
User behavior analysis, RFM model, Deep learning, Market prediction