Projects proposed by the Atlântico Academy Bootcamp to improve knowledge in the areas of Machine Learning, Natural Language Processing and Computer Vision. The repository is subdivided into codes and datasets. The data, for the NLP projects, were taken from messages on the social network Twitter.
- The dataset used in this practice has more than 10,000 tweets, which were selected as they had a connection with search results for disastrous and non-disastrous events. The purpose of this practice is to classify whether the tweets are disastrous or not. Assessments of this type serve to warn safety and health professionals about the disaster more quickly and accurately.
2️⃣ Sentimental Analysis with NLTK
-
The data used in this practice were taken from the NLTK library, using 10,000 tweets divided equally into positive and negative. In the example, data cleaning, processing and classification methods were used. The trained data will be used to classify whether sentences have negative or positive feelings. This type of analysis serves to assist algorithms in verifying users who have a quieter or more aggressive profile.
-
The file nlp_test.py is the main code.
-
The file package.py has the fuction of downloading the packages necessary for the implementation of the example.
- Tutorials about OpenCV and Scikit_image.