Title: A preschool education social media-monitoring system based on optimised-sentiment analysis
Authors: Jie Qiu
Addresses: School of Education, Xi'an Fanyi University, Xi'an, 710105, China
Abstract: This paper suggests a sentiment monitoring system in the preschool industry to monitor the sentiment of the people regarding early childhood learning. The system gathers and pre-processes social media posts with help of transformer-based language models and entropy scoring, sentiment classification, and unpredictability measurement. The information is collected and presented in real-time on a dynamic dashboard. Findings indicate that there is no consistency between the magnitude and the sentiment change of post volume and that entropy-based metrics provide a more precise analysis of the volume. The system is capable of identifying any abrupt shifts in the mood and thus organisations can be able to respond to current issues at the earliest opportunity. In preschool learning, this method increases parent involvement, organisational sensitivity, and relationship development by using AI-based sentiment analysis.
Keywords: sentiment analysis; social media monitoring; emotional entropy; transformer models.
DOI: 10.1504/IJICT.2025.150408
International Journal of Information and Communication Technology, 2025 Vol.26 No.44, pp.18 - 38
Received: 07 Sep 2025
Accepted: 21 Oct 2025
Published online: 12 Dec 2025 *


