Article: Application of visual data analysis system based on artificial intelligence Journal: International Journal of Information and Communication Technology (IJICT) 2025 Vol.26 No.43 pp.1 - 19 Abstract: For the insufficiency of traditional systems in automated data processing and predictive analysis capability, this study explored a visual data analysis system based on advanced artificial intelligence technology, integrating the three core functions of automated data preparation, intelligent recommendation and predictive analysis. Data cleaning was carried out by weighted k-nearest neighbours imputation and isolation forest algorithm. The unstructured data was handled utilising bidirectional encoder representations from transformers (BERT) models, and key patterns, trends and anomalies were discovered by means of association rule learning techniques. Relying on the autoregressive integrated moving average (ARIMA) model, the time series data was precisely forecasted. Distributed deployment supports the hardware and solves the system layout problem. The evaluation outcomes demonstrated that the ARIMA model performed the best in data prediction with an average prediction time of only 1.075 seconds, the lowest RMSE (7.19) and MAE (4.70), and the highest prediction accuracy (96.00%). This paper provides efficient and intelligent data support and solutions to help decision-making and strategic planning in various industries. Inderscience Publishers - linking academia, business and industry through research

Title: Application of visual data analysis system based on artificial intelligence

Authors: Xinyun Cheng; Shijie Zhang; Pengfei Wang; Zhikang Wang; Lincheng Qi

Addresses: Information Transportation Inspection Center, State Grid Jiangsu Electric Power Company, Information and Telecommunication Branch, Nanjing, 210024, Jiangsu, China ' Jiangsu Electric Power Information Technology Co., Ltd., Nanjing, 210024, Jiangsu, China ' Information Transportation Inspection Center, State Grid Jiangsu Electric Power Company, Information and Telecommunication Branch, Nanjing, 210024, Jiangsu, China ' Information Transportation Inspection Center, State Grid Jiangsu Electric Power Company, Information and Telecommunication Branch, Nanjing, 210024, Jiangsu, China ' Jiangsu Electric Power Information Technology Co., Ltd., Nanjing, 210024, Jiangsu, China

Abstract: For the insufficiency of traditional systems in automated data processing and predictive analysis capability, this study explored a visual data analysis system based on advanced artificial intelligence technology, integrating the three core functions of automated data preparation, intelligent recommendation and predictive analysis. Data cleaning was carried out by weighted k-nearest neighbours imputation and isolation forest algorithm. The unstructured data was handled utilising bidirectional encoder representations from transformers (BERT) models, and key patterns, trends and anomalies were discovered by means of association rule learning techniques. Relying on the autoregressive integrated moving average (ARIMA) model, the time series data was precisely forecasted. Distributed deployment supports the hardware and solves the system layout problem. The evaluation outcomes demonstrated that the ARIMA model performed the best in data prediction with an average prediction time of only 1.075 seconds, the lowest RMSE (7.19) and MAE (4.70), and the highest prediction accuracy (96.00%). This paper provides efficient and intelligent data support and solutions to help decision-making and strategic planning in various industries.

Keywords: visual data analysis system; artificial intelligence technology; distributed deployment; predictive analysis; ARIMA model; data automation processing.

DOI: 10.1504/IJICT.2025.150398

International Journal of Information and Communication Technology, 2025 Vol.26 No.43, pp.1 - 19

Received: 18 Jul 2025
Accepted: 09 Sep 2025
Published online: 12 Dec 2025 *

Title: Application of visual data analysis system based on artificial intelligence

Keep up-to-date