Work

AI-Driven Multi-Source Data Analysis of Chinese Students’ Overseas Learning Experience: A Machine Learning Approach to Satisfaction Prediction and Service Optimization

Year: 2026

Type: conference-paper

Abstract: This study proposes an AI-driven Expectation-Satisfaction-Action (ESA) framework to analyze Chinese students’ overseas learning experiences using multi-source data and machine learning. We surveyed 303 Chinese students across 24 destination countries. The analysis combines importance-satisfaction gap analysis, Random Forest and XGBoost models, and K-Means clustering to identify service priorities, predict satisfaction, and group student types. Results show a significant average gap of 0.54 (p < .001) between importance and satisfaction, with digital communication (social media, website timeliness) and financial access as top priorities. Random Forest predicts overall satisfaction at 84.7% accuracy, and XGBoost reaches R² = 0.782 for dimensional forecasting. We identified four student clusters, Digital Optimizers (28.4%), Resource Constrained (24.7%), Experience Seekers (23.6%), and Satisfaction Strugglers (23.3%), each matched with personalized AI-driven service strategies. This work advances the use of educational data mining in service quality research and provides actionable guidance for institutions using AI to improve international student experiences.

Source: International Conference on Artificial Intelligence, Digital Media and Education

DOI: 10.1145/3839592.3839672

Authors: Zongsheng Chen, Kening Da, Yanli Sun, Christos Kelepouris, Chunyizi Yang

Institutions: Shenyang Jianzhu University, Chaoyang Normal University, Benedictine University

Language: English

Topic: International Student and Expatriate Challenges

Subfield: Communication

Field: Social Sciences

Domain: Social Sciences

SDG: Quality education

Open Access status: hybrid

OpenAlex ID: W7220860141

FWCI:

Cites: 21

Cited by:

Related to: 10