Implementación de herramienta de inteligencia artificial de forma local y una interfaz grafica web: implementación de servicios de cómputo y de seguridad.

Abstract

This project primarily aims to implement a Linux operating system, specifically Ubuntu, to locally run artificial intelligence models using Ollama. The initiative seeks to develop a robust and efficient environment that enables the processing and execution of language models, facilitating the completion of questionnaires and other AI-related tasks. The project's theoretical framework addresses fundamental topics such as the Linux operating system, artificial intelligence, and the tools used like Docker and DevOps. These concepts are essential to understand the foundation upon which the entire system is built. The methodology employed in the project includes the initial configuration of the Ubuntu virtual machine, the download and setup of Ollama, and the implementation of Docker and DevOps. Detailed descriptions are provided on how to create docker-compose.yaml and main.py files to define and build container services, as well as how to access the system locally. The obtained results demonstrate the effectiveness of the implemented system. Users can access the system via local web links and ask questions on various topics, leveraging the processing of language models to receive precise and quick answers. The project concludes with an evaluation of the implementation's success, highlighting the advantages and benefits of using a Linux environment for running artificial intelligence models. Additionally, recommendations for future work and potential system improvements are offered, ensuring its evolution and adaptation to new needs and technologies.