Here you will find python notebook or R scripts where I developed exercises or studies aimed at data science cases and machine learning.
The subjects include a summary of concepts, techniques and methods applied in order to support future references to similar cases.
I will include new materials as new cases have been developed. Competition materials and other developments continued to be placed in other repositories.
Below is a brief summary of each case, more details can be found in the README.MD of each solution, as well as inside the materials themselves.
Wine Type and Quality Classification 
A python Notebbok to analyze the physicochemical attributes of wine and understand their relationships and significance with wine quality and types classifications. To do this, We will proceed according to the standard Machine Learning and data mining workflow models like the CRISP-DM model, mainly for:
- Predict if each wine sample is a red or white wine.
- Predict the quality of each wine sample, which can be low, medium, or high.
The dataset are related to red and white variants of the "Vinho Verde" wine. Vinho verde is a unique product from the Minho (northwest) region of Portugal. Medium in alcohol, is it particularly appreciated due to its freshness (specially in the summer). This dataset is public available for research purposes only, for more information, read Cortez et al., 2009. . Due to privacy and logistic issues, only physicochemical (inputs) and sensory (the output) variables are available (e.g. there is no data about grape types, wine brand, wine selling price, etc.).
CNN Exercise - Deep Learning for Computer Vision
This python notebook explore convolutional neural networks through the task of image classification using publicly dataset CIFAR-10. We will utilize our understanding of CNNs to then take on the task of style transfer and understand how neural networks can be used to understand high-level features. Through this notebook, we cover the following topics:
- Image classification use CNNs from scratch
- Transfer learning: image classification using pretrained models
This notebook we using publicly dataset On-line Retail to explore customer segmentation through the interesting task of unsupervised learning method. Then we go further and apply association rule mining approach to find interesting rules and patterns in this transaction database. These customer segmentation, rules and patterns can be used to make interesting and useful decisions as far as user interest is concerned.
The Online Retail a transnational data set which contains all the transactions occurring between 01/12/2010 and 09/12/2011 for a UK-based and registered non-store online retail.The company mainly sells unique all-occasion gifts. Many customers of the company are wholesalers.
This notebook uses the user-song play count dataset to uncover different ways in which we can recommend new tracks to different users. We will start with a very basic system and try to evolve linearly into a sophisticated recommendation system.
In the first part, we cover a wide variety of techniques for analyzing sentiment, which include the following.
- Unsupervised lexicon-based models
- Traditional supervised Machine Learning models
On the second We cover a wide variety of techniques of deep leaning for analyzing sentiment, which include the following.
- Newer supervised Deep Learning models
- Advanced supervised Deep Learning models
Besides looking at various approaches and models, we also focus on important aspects in the Machine Learning pipeline including text pre-processing, normalization, and in-depth analysis of models, including model interpretation and topic models. The key idea here is to understand how we tackle a problem like sentiment analysis on unstructured text, learn various techniques, models and understand how to interpret the results. This will enable you to use these methodologies in the future on your own datasets.