An Adaptive Data Preprocessing Framework for Improved Learning: A Case Study of Tangier Container Terminal

• DOI: 10.3844/jcssp.2024.265.275
• Publication Date: 2024-02-15T08:20:10Z
• Authors: Mostafa Al Uahabi; Hicham Attariuas; Mohammed Saleh; Mohamed Chentouf
• Keywords: Artificial intelligence; Social Sciences; FOS: Mechanical engineering; Business, Management and Accounting; Terminal (telecommunication); Computer science; Mechanical engineering; Management Information Systems; Modeling and Control of Multidimensional Systems; Engineering; Container (type theory); Control and Systems Engineering; Physical Sciences; Telecommunications; Predictive Analytics; Network Traffic Analysis; Data mining; Impact of Big Data Analytics on Business Performance; Preprocessor

Container terminals are critical nodes within the maritime transportation system that have a vital function in global merchandise trade, handling significant volume of cargo through use various equipment and personnel. Thus, efficiency container terminal operations relies heavily on ability to collect, analyze, utilize operational data. However, such data can be corrupted by noise, missing points, outliers, incomplete or inconsistent information, making subsequent analysis modeling challenging. This study proposes an adaptive preprocessing framework tailored context operations, using from tangier as case study, leading port Mediterranean Africa, also ranked 4th CPPI 2022. includes techniques for integration, cleaning, transformation, encoding acquire high-quality In addition, RFE feature selection method is employed identify most discriminative subset. Finally, proposed approach, assessed extra tree regressor model, demonstrates strong prediction capabilities with R-squared score 95.4% based selected features predicting duration vessels at port, highlighting its integration into operating improve management efficiency.