Telescopia is a web observatory for the stock market. It was made for the Databases class lectured by Dr. Leonardo Rocha in 2019/2 period of Computing Science at the Federal University of São João del-Rei.

The main goal of the project was to deal with large amounts of data, therefore non-relational databases should be used.

It consists in two main parts:

• REST API
• Frontend application

As the observatory functions, there are 4:

• Notícias pela variação de ação (Articles by stock variation)
• Notícias por palavra chave (Articles by keywords)
• Análise de tópicos (Topic analysis)
• Notícias por ações citadas (Articles by cited stock)

This function will return, once given a stock name (already in the database), a variation number and a preset comparison operator, all the dates where the selected stock had the (exact, bigger than, less than, range) given variation. Once the dates are retrieved, the API returns all the articles that were published in those retrieved dates. Also, a graph is displayed showing all the dates and it's variation of the selected stock.

This function WAS NOT implemented, but was intended to be made using tf-idf method, returning articles ranked by the presence of the given keyword.

Once given a stock, that was previously registered in the database, all the articles that have cited the stock name are listed, showing the article title, publish date, portal name and a link to the original article.

For the development of the API, it was used:

• Node.js 10.16.2
• MongoDB 4.2.1
• Mongoose 5.7.12
• Express 4.17.1
• Cors 2.8.5
• Body-parser 1.19.0

For the frontend application:

• Material Design for Bootstrap 4
• Vue.js 2
• Axios
• Chart.js
• Font Awesome
• Coming Soon Free MDBoostrap Template (index page)

Before running the following instructions, make sure you got installed in you computer Node.js and NPM.

First you need to install the dependencies:

npm install

Then, to run in development mode:

npm start

Disclaimer: To run the API in production mode, you can use the PM2 package.

The frontend application is made out of pure HTML5, CSS and JavaScript libraries, therefore, you won't need nothing but a Code Editor and a Browser to develop it.

To run it in production mode, any HTTP server will serve well.