Análisis de intereses turísticos utilizando redes sociales en un entorno de big data

Abstract

Twitter, the popular microblogging, has more than 320 million users registered nowadays. A big challenge in this project was to use the information collected in Paris, London and New York, to perform a data analysis. The data that was analyzed are: tastes, preferences and opinions about tourist places in the cities mentioned above. Another challenge is to verify this information is accurate for example: coordinates of the tourist places or tweets that mention a tourist place. Since many user’s don´t agree to provide all their personal information in twitter. The applied methodology in this project was “big data harvesting”. The first step of the methodology in question involved giving a correct format to the data. The second step, I collected 16,064,840 tweets from the most populated cities from 20 may to 12 august on 2018. In the third and fourth step, the detection and classification of feelings were performed respectively. The analyzed tweets were classified as: positive and negative. This classification was made with libraries based on NLP (natural language processing). As a final step, the data was analyzed, with the aim of providing information to people or institutions through a web service of the best tourist places in the cities mentioned above. The results obtained represented the recommendations of other users who shared their experiences on twitter, when visiting a tourist place.