The "DSD" repository has been created to share the projects developed in each of the modules that make up the Diploma in Data Science, taken during the year 2021 at the National Technological University - Regional Faculty Cordoba.
The modules are as follows:
Module 1: Data Science and Exploratory Analysis (30 hours) Brief description: This module covers the basics of Data Science, roles within the discipline, project structuring, and common work methodologies. During the labs, an exploratory analysis of a proposed dataset will be carried out and a Churn use case will be built. Use case: Churn model.
Module 2: Fundamentals of Machine Learning (30 hours) Brief description: This module covers the basics of Machine Learning, focusing on supervised learning. It covers the less complex methods (regressions and decision trees) and the variable engineering process. A predictive model will be developed in the lab. Use case: Prediction model.
Module 3: Unsupervised Machine Learning - Clustering (30 hours) Brief description: This module delves deeper into unsupervised machine learning, including the variable treatment used for segmentation. Different segmentation and dimensionality reduction techniques will be evaluated. Finally, the cluster description process and the best ways to convey information obtained from that segmentation will be addressed. During the labs, a Customer Segmentation model will be developed. Use case: Clustering model.
Module 4: Neural Networks (30 hours) Brief description: This module covers the fundamentals of neural networks, working on the concept of neurons, activation types, and layers within a network. The concept of Cloud Computing will also be addressed, as well as its advantages and disadvantages. Some useful use cases will be reviewed and a prediction lab will be carried out through neural networks.
Module 5: Deep Learning and Computer Vision (30 hours) Brief description: The necessary concepts to understand Deep Learning will be provided, including the different types of use cases that can be used. Fine tuning mechanism will be used to leverage already trained learning and knowledge transfer will be carried out. Key computer vision concepts and technologies used will be addressed. During the lab, an object/person classification project in images will be developed. Use case: Object Classification through Deep Learning.
Module 6: Natural Language Processing (30 hours) Brief description: This module covers the fundamentals of Natural Language Processing along with the most common challenges. The concepts of Language and the semantic value of words will be addressed. The necessary tools for sentiment analysis and chatbots will be provided. In the lab, a sentiment analysis project on social media data will be carried out. Use case: Natural Language Processing in Sentiment Analysis.
Module 7: NLP in Recommendation Systems (30 hours) Brief description: This module is entirely optional and those who choose to take it will do so completely virtually, with faculty support. It covers what a recommendation system is, the most common use cases, as well as the fundamentals and tools necessary for its construction. During the lab, a simple movie or music recommendation system will be built. Use case: Building a Recommendation System.