Exploring Boost Efficiency in Text Analysis by Using AI Techniques in Port Companies

This study presents how integrating natural language processing (NLP) and machine learning (ML) optimizes strategic management in the port sector. Using hybrid NLP-ML models, the accuracy of classification and prediction of strategic information is significantly improved by analyzing large sets of textual data, both unstructured and semi-structured. The methodological approach is developed in three phases: first, a strategic analysis of port systems is performed using NLP; then, ML is integrated with NLP for text classification using advanced tools such as BERT and Word2Vec; finally, advanced models, including Decision Trees and Recurrent Neural Networks are evaluated. Applied to 55 companies in three countries, this method extracts key strategic data such as mission, vision, values and corporate objectives from their websites to obtain strategic terms related to innovation and sustainability. The study improves the ability to interpret textual data, enabling more informed and agile decision-making, which is essential in a highly competitive and dynamic environment.