Nalphack/Morocco-covid-19--A-machine-learning-and-multi-agent-system-approach

Morocco Covid-19 data analysis, news headlines scrapping and sentimental analysis using machine learning models in a multi agents system architecture. # Morocco covid-19: A machine learning and multi-agent system approach This project is a class project (for learning purposes) as a part of Multi agents systems and Machine learning class at The Faculty of Sciences and Technologies of Tangier ## Objective The main objective of the project is the implementation of an intelligent multi-agent system based on several supervised and unsupervised learning algorithms. The system will be composed of several intelligent and interactive agents, from which each agent performs a specific task. In addition to a Single page Application (Angular based) that will interact with the system. ## Built With The project is built using the following Frameworks * MongoDB - The used to store data * Flask - The web framework used to build the smart layer API * Spring Boot - The web framework used to build the MAS layer API * JADE Framework - The framework used to implement agents in the MAS layer * Angular - The framework used to build the front-end layer For detailed instructions on prerequisites for each layer please read the readme.md file on each layer. ## Architecture The project is built using a combination of restful and multi agent architechture, it consists of 3 separated layers: ### Smart layer This is a Flask based REST api containing all the scripts needed to scraping, pre-processing and loading data into a Mongodb database in addition to all the models we have developed for clustering, predictions, sentimental analysis and visualizations. ### Multi Agent System (MAS) layer This is a REST api based on SpringBoot containing JADE Framework with all agents, and their behaviours and how they communicate with each other and how they serve the SPA frontend layer. ### Single Page Application (SPA) layer This is the front-end part which contains pages for each agent (clustering, predictions, sentimental analysis and data visualisation), each page communicates with it's own appropriate agent using a restful architecture. ## Aut …