Data Science

Topic 1: Problem Identification

This unit covers the initial step of the Data Science Method, which is the problem statement. To begin a data science project, it is important to identify the problem that needs to be solved, define it clearly, and break it down into smaller and manageable pieces. In this unit, you will learn how to choose the right problem to solve and set goals for your project.

• SMART problem statements
• Problem statement worksheet

Topic 2: Python for Data Analysis

In this unit, you will learn how to program in Python while following the best coding practices. Additionally, Python-based tools will be introduced to you. Pandas will be introduced as a powerful tool for cleaning and manipulating data.

Topics Covered:

• Introduction to Python data types and standard libraries
• Statistical concepts for data analysis
• Data manipulation and analysis with Pandas
• Data visualization with Matplotlib and Seaborn

Topic 3: Data Analysis with SQL, Tableau / Power BI

This unit is focused on teaching data visualization techniques using Tableau and Power BI so that you can effectively communicate insights derived from data. You will learn how to collect, organize, and clean data to prepare it for visualization. You’ll learn to connect to various data sources, structure data, resolve formatting issues, and create data profiles to understand underlying trends and patterns. By the end of the unit, you will have the skills and knowledge to use Tableau and Power BI to create engaging and informative data visualizations that can drive decision-making and convey key insights effectively.

Topics Covered:

• Introduction to Databases
• Basic SQL: Syntax
• Retrieving data
• Sorting and Filtering
• Aggregating Data, Grouping, Joining Tables, Subqueries etc
• Introduction to Data visualization tools
• Best Practices for Effective Data Visualisation
• Real-world examples of data Visualization projects using Tableau and Power BI
• Capstone project

4: Data Science Methodology

Throughout this unit, you will gain a comprehensive understanding of the Data Science Methodology. This methodology encompasses all aspects of real-world data science challenges. You will be taught how to collect and manipulate data, engineer features, explore a range of modelling techniques, evaluate, validate and interpret models. To conclude, you will master the art of deploying models into production environments and making well-informed decisions using data.

Topics Covered:

• Problem identification
• Data cleaning, collection and wrangling
• Exploratory data analysis
• Feature Engineering
• Model selection
• Hyperparameter tuning
• Evaluation and validation
• Deployment
• Capstone project

Topic 5: Machine Learning Overview

Machine learning is a powerful tool that extracts valuable insights from complex data sets using computer science and statistics. This unit teaches you fundamental concepts, best practices, and common challenges in machine learning. By the end of this unit, you’ll be able to build predictive models and make accurate recommendations based on data.

Topics Covered:

• Intro to supervised vs. unsupervised learning
• Batch vs. online learning
• Instance-based vs. model-based learning

6: Supervised Learning

In this unit, you will receive a crash course on supervised learning, which is one of the most commonly used forms of machine learning. In supervised learning, you provide the machine with your labelled training data and encode procedures for it to learn how to assign those labels by itself. You will learn about the most popular algorithms, including linear and logistic regression, support vector machines, decision trees, clustering, time series and forecasting, ensemble learning with random forests, and gradient boosting.

Topics Covered:

• Supervised learning and its common applications
• Regression and classification techniques
• Decision trees, random forests, and gradient boosting
• Support Vector Machine (SVM) and kernels
• Evaluation metrics for regression and classification tasks
• Model optimization

Topic 7: Unsupervised Learning

In this unit, you’ll explore unsupervised learning, which involves the machine finding patterns in a dataset without pre-existing labels. You’ll learn the most commonly used clustering techniques, complete exercises on distance metrics and cosine similarity, and delve into a case study on customer segmentation using K-means clustering

Topics Covered:

• Euclidean and Manhattan distances
• K-means clustering
• Agglomerative hierarchical clustering

Topic 8: Natural Language Processing

In this unit, you will learn about unsupervised learning techniques that are specifically designed for Natural Language Processing (NLP). Unlike supervised learning, where models are trained on labelled data, NLP involves extracting patterns and structures from text data without predefined labels. You will explore various clustering techniques that are specifically tailored for analyzing text data.

Topics Covered:

• Introduction to NLP and its applications
• Text pre-processing techniques
• Text representation methods (e.g., Bag of Words, TF-IDF)
• Sentiment analysis and text classification

Topic 9: Neural Network

This unit will cover neural networks -computational models that mimic the human brain’s structure and function. Neural networks can learn complex patterns and relationships from data and outperform traditional algorithms in tasks such as image recognition and natural language processing. We’ll explore various types of neural networks, their architectures, training methodologies, and real-world applications.

Topics Covered:

• Basics of artificial neural networks (ANNs)
• Multilayer perceptron (MLPs) for classification and regression
• Convolutional Neural Networks (CNNs) for image analysis

Topic 10: Building a Data Science Portfolio

Capstone One: Guided Capstone

◘ The first capstone project focuses on data analytics with Power BI or Tableau. You’ll receive step-by-step guidance and valuable insights throughout the project to help you proficiently use these tools. The primary aim is to build a solid understanding of data analytics before moving on to more advanced projects.

Capstone Two:

◘ In your second capstone, you must come up with a project idea and proposal and wrangle the data. Use explanatory data analysis techniques to understand the data. Pre-process and create a training dataset and build a working model. Finally, document and present your work.

Capstone Three:

◘ Your inaugural NLP capstone project will introduce you to its core principles and methodologies through hands-on experience using tailored tools and techniques. You will learn to apply NLP techniques, analyze text data, and extract meaningful insights. The goal of this capstone is to establish a solid foundation in NLP and equip you with practical skills for future projects.