Design and Development of a Lyrics Emotion Analysis System for Creative Industries

José María Izquierdo-Mora. (2018). Design and Development of a Lyrics Emotion Analysis System for Creative Industries. Final Career Project (TFG). Universidad Politécnica de Madrid, ETSI Telecomunicación, Madrid.

Abstract:
The way to achieve this, it has been developing a visualization environment to present the data obtained from the lyrics of the songs , and analyze them with Senpy. This visualization environment has been developed with Polymer Web Components and D3.js. The data flow has been between musixmatch, Senpy and ElasticSearch. Luigi has been used in this process as it helps to build complex flows of clustered tasks, Luigi has gone from analyzing the lyrics to storing them in ElasticSearch. Next, D3.js has been used to create interactive widgets that make data easily accessible, these widgets will allow the user to interact with them and filter the data that is most interesting to them. To make the interface according to the Material Designing Google and displaying dynamic data in widgets, the Polymer Web Components library has been used. As a resolution of the project, we can make a broad analysis of the lyrics, depending on the songs and artists and thus know the emotions and feelings they generate.