Accès libre

Reasearch on Cross-National E-commerce User Behavior Analysis and Conversion Rate Improvement Based on the Improved XLSTM Algorithm

 et   
17 mars 2025
À propos de cet article

Citez
Télécharger la couverture

Figure 1.

XLSTM Network Architecture
XLSTM Network Architecture

Figure 2-3.

Sample Data Distribution by Product Category and the impact of Product Rating on Conversion Rate
Sample Data Distribution by Product Category and the impact of Product Rating on Conversion Rate

Figure. 4.

Heatmaps comparing the performance of clustering algorithms across months (January–June 2024).
Heatmaps comparing the performance of clustering algorithms across months (January–June 2024).

Figure. 5.

A comparative analysis of feature importance derived from LSTM and XLSTM models. Coupon
A comparative analysis of feature importance derived from LSTM and XLSTM models. Coupon

Device and Platform Distribution

Device Type Frequency Conversion Rate (%)
Mobile 44,392 5.2
Desktop 25,670 7.8
Tablet 11,582 4.1

Performance Comparison of Clustering Algorithms

Algorithm Silhouette Score Davies-Bouldin Index (DBI) Adjusted Rand Index (ARI)
K-means Clustering 0.62 1.09 0.71
DBSCAN 0.58 1.32 0.67
Agglomerative Hierarchical 0.55 1.45 0.64
XLSTM + K-means Clustering 0.77 0.92 0.81

User Purchase Behavior

User Segment Average Purchase Frequency Avg. Purchase Value Conversion Rate (%)
Frequent Shoppers 5/month $157 25%
Occasional Shoppers 2/month $75 8%
Window Shoppers 0.5/month $21 1%