Bokeh-Dashboard-Debate-Analysis

Project developed in Python 3.5 making use of Bokeh library to display opinion of users from the Debate of June 13, 2016 among the candidates for the presidency of the Government of Spain.

Dashboard elements

In the upper left, it can see 3 widget where can be entered the next information:

• Political: It can select all tweets where the name of a certain candidate appears.
• Politic Party: It can select all tweets where the name of a certain politic party appears.
• Keywords: It can select certain keywords that appear during the debate. It can be choosen from the following words [españ, vota, catal, gana, pais, gobiern, corrup, euro, grecia, europ, cambio, trabaj, venez, mejor, perd, programa, refugiado, perd, independ, finan, violenc, empleo, terror, banc, vergüenza]. The words have undergone a previous stemmer process for getting maximum number of matches with tweet texts.

Below the widgets, it can see 2 figures where the first one represents the mean of all hashtagg and mentions used in each tweet. And finally, it can see 3 graphs with the information of the number of positive, negative or neutral tweets gather every 5 minutes.

Execution example

First you must be inside of proyect folder and then:

bokeh serve --show .