Required Data Analysis,Data Science,Natural Language Processing,Knowledge Representation,Natural Language Generation freelancer for Sentiment Analysis and Latent Dirichlet allocation (LDA) job

We have collected approximately 7 million tweets. We have them stored in approximately 60 different CSV files. The job would be the following:

1. To order and clean the dataset. Information relative to the number of tweets duplicated will be requested (among other details).
2. To apply a Sentiment Analysis to these tweets to be divided into Positive, Negative, and Neutral sentiments using TextBlob in Python.
3. To apply an LDA model. You will have to apply this model to three databases, one with Positive, Negative, and Neutral sentiments (this is the result of point 2). Therefore, the results will show positive topics, negatives topics, and neutral topics.

Skills & Expertise Required

Data Analysis Data Science Natural Language Processing Knowledge Representation Natural Language Generation