Natural Language Processing Applied to Marketing-Oriented Sentiment Analysis

This thesis deals with sentiment analysis in the context of marketing. Emphasis is placed on natural language processing techniques used in this process.

First, we present a literature review laying the theoretical foundations of sentiment analysis, its stakes and its implementation methods, including machine learning algorithms.
Second, we carry out a case study based on the analysis of an annotated and preprocessed corpus of consumers' tweets by a set of sentiment analysis tools.
The results obtained by these tools and their evaluation demonstrate the value of sentiment analysis in marketing and the limited impact of preprocessing on this process.
This project served as my thesis for my Master's in Information and Communications Technology at ULB.
Python (Tweepy, NLTK)