Deep Learning Based Intelligent System for Data Automation in a Private Security Company

Abstract

Nowadays, the advance of digitalization and the growing need to optimize processes have driven companies to look for innovative solutions to improve their operational efficiency. In this context, the following study aimed to implement an intelligent system based on Deep Learning to automate data processes within the company, focusing on improving efficiency, accuracy and optimization of operational tasks. The research used a pre-experimental design, with quantitative approach and applied character, analyzing a population of 1,400 records of human management data, such as attendance, task and customers. Data collection was carried out through direct observation and documentary analysis, while the system development was based on the Crystal-Clear agile methodology, allowing specific adaptations to the company's operational needs. The results showed significant improvements: information access time was reduced from 29.90 seconds to 15.77 seconds, accuracy in identifying sensitive data increased from 58.70% to 91.43%, and automation efficiency increased from 28.67% to 95.67%. Statistical tests confirmed the relevance of these results. The system optimized processes and reduced manual intervention and highlighted the importance of customized intelligent solutions to improve organizational performance. These findings highlight the potential of Deep Learning-based systems to transform data-sensitive sectors such as private security.