Technical gazette, Vol. 32 No. 5, 2025.
Original scientific paper
https://doi.org/10.17559/TV-20240829001951
Design and Performance Analysis of a Web Crawler-Based System for Emotional Feature Extraction from Social Media Data
Xinyue Feng
; School of Electronic Information, Foshan Polytechnic, Foshan, China, 528137 Foshan Polytechnic, No. 3, Vocational Education Road, Leping Town, Sanshui District, Foshan, Guangdong, China, 528137
*
Niwat Angkawisittpan
; Research Unit for Electrical and Computer Engineering Technology (RECENT), Mahasarakham University, Kantarawichai, MahaSarakham, Thailand, 44150 No.41/20, Kantarawichai District, MahaSarakham, 44150, Thailand
Jianhui Li
; School of Electronic Information Foshan Polytechnic, Foshan, China,528137 Foshan Polytechnic, No. 3, Vocational Education Road, Leping Town, Sanshui District, Foshan City, Guangdong Province, China, 528137
* Corresponding author.
Abstract
This paper presents a multi-component web crawler approach for extracting salient emotional features from Weibo data, emphasizing temporal dynamics and data sequence complexity. By implementing a weighted node strategy and extreme point extraction technique, this method ensures high accuracy in data collection and emotional feature identification. The sliding window approach optimizes similarity measurements between acquired data and target emotional content. Experiments demonstrate consistent crawling accuracy above 95%, underscoring the method's stability and scalability. This approach provides a robust tool for social media sentiment analysis, offering enhanced accuracy and completeness in real-time emotional feature extraction from Weibo data.
Keywords
emotional feature extraction; information retrieval; time series analysis; web crawler; Weibo data
Hrčak ID:
335060
URI
Publication date:
30.8.2025.
Visits: 311 *