In this repository I've include a Python Jupyter Notebook that explain how to classify a collection of Tweets in order to build a recommendation engine.
The goal of this work is to build predictive models that can automatically infer people’s needs from user- generated content. Specifically, we want to analyze text data from social media in order to build a recommender system. The engine can be able to recognize a specific need or pattern inside the data and suggest a product to sell. For example, we suppose to analyze data from twitter and try to identify people who want to travel in the future (or who are a "traveller") and suggest them a travel-related products.
In this work I use my personal python package for ML tasks. You can install it by download the source folder from my GitHub repository https://github.com/Tostox/mltools or use the pip commad:
pip install git+https://github.com/Tostox/mltools.git