Machine Learning and Data Analytics using Python

Programme Modules

• 1. Analytics: From Data to Insights (Week 1 – 6)

  This section teaches you how to collect, organise, and analyse data to
  a. provide insights on how a business is functioning
  b. predict trends
  c. identify areas for improvement
  
  Week 1: Introduction to Analytics
  
  

  • The role of analytics in business intelligence and artificial intelligence
  • Data Processing Chain
  • A simple analytics example

  
  Week 2: Database: Data Source and Data Queries
  
  

  • Understanding databases structure
  • Exploring a database in SQLite
  • Getting started with queries using SQLite
  • Turning data into information
  • Working with multiple tables
  • Using functions

  
  Week 3: Data Warehouse: Load and Transform Data and Create a Data Model
  
  

  • Data preparation and data wrangling
  • Introduction to Exploratory Data Analysis (EDA)
  • Exploring and importing data into Power Pivot
  • Data munging with Power Query
  • Star schema data model
  • Creating a data model in Power Pivot

  
  Week 4: Data Warehouse: DAX and Time-based Analysis
  
  

  • DAX calculation types and functions
  • Getting data from related tables
  • Data context
  • Creating calculated columns and measures in Power Pivot
  • Time-based analysis and period-based evaluations
  • Utilising EDA to present time-based data

  
  Week 5: Data Visualisation: Pivot Tables and Charts and Power BI Basics
  
  

  • Working with Pivot tables and charts to create visualisations
  • Creating the BI interface in Excel
  • Case study using Power Pivot: Reseller Sales Analysis
  • Creating data Model, visualisations and dashboards in Power BI
  • Case study using Power BI: Sales Quota Analysis
  • Common visualisations used in EDA

  
  Week 6: Data Mining Basics
  
  

  • The fundamentals of data mining and CRISP-DM
  • Demonstrating unsupervised learning technique, specifically, K-Means clustering in Orange using practical examples
  • Demonstrating supervised learning techniques such as decision tree and linear regression models in Orange to solve real-life problems
• 2. Programming with Python (Week 8 – 16)

  This section provides hands-on experience on the most popular programming language in the data science world, Python Programming, to apply Python functions, operators, files, and packages for:
  a. data evaluation
  b. data visualisation
  c. application development
  
  Week 8: Getting Started with Python
  
  

  • Introduction to Python as a programming language
  • Identifying types of errors in a Python program
  • Examining the basic building blocks in a Python program
  • Creating a simple Python program

  
  Week 9: Collections, Strings and Comments
  
  

  • Using appropriate containers or collection data types to store data
  • Applying properties of lists, tuples, sets and dictionaries to organise data
  • Manipulating data and containers using functions and built-in methods on collections
  • Identifying the characteristics of strings
  • Implementing functions, operators and built-in methods on strings
  • Applying comments to describe written Python programs

  
  Week 10: Operators and Program Flow Controls
  
  

  • Explaining concept of various operators
  • Using operators under appropriate conditions
  • Applying operator precedence and associativity for expression resolution
  • Examining the different types of program flow control
  • Utilising appropriate program flow control mechanisms
  • Executing sequence, selection and iteration program flow controls

  
  Week 11: Functions, Import Statements, Inputs, Outputs, Exception Handling and File Handling
  
  

  • Exploring different types of functions
  • Differentiating functions from methods
  • Applying import keywords to Python modules
  • Applying the correct input syntax
  • Using exception statements to handle code errors
  • Displaying outputs with the correct syntax and format
  • Exploring methods in handling text and binary files

  
  Week 12: The Numpy Library
  
  

  • Applying concepts in indexing, slicing and iterating arrays
  • Understanding the process of reading and writing arrays into files
  • Exploring the methods to manipulate arrays

  
  Week 13: The Pandas Library
  
  

  • Creating and manipulating series and dataframes in the pandas library
  • Applying operations and functions to manipulate and manage values

  
  Week 14: Scikit Learn and Keras
  
  

  • Overview on Scikit learn
  • Classification, regression, clustering, model selection, pre-processing and dimensionality reduction in Scikit Learn
  • Importing the library along with various functions required for model evaluation- confusion matrix, test_train_split etc.,
  • Overview on deep learning
  • Types of models supported by Keras
  • Importing the library and building models in Keras and its applications

  
  Week 15: Data Visualisation
  
  

  • Creating data visualisations in matplotlib, pandas plot and Seaborn library
  • Applying concepts in enhancing and manipulating data visualisations

  
  Week 16: Python Application Development
  
  

  • Applying concepts in Python to create an Inventory Management System
  • Building a Python application using external data
• 3. Machine Learning for Practical Applications (Week 18 – 30)

  This final section on Machine Learning covers the theoretical learning while emphasising on practical know-how needed to apply the techniques to generate impactful results.
  
  While focusing on Advance models and implementation challenges, the section stands on 3 building blocks of Machine Learning namely
  1. supervised learning
  2. reinforcement learning
  3. unsupervised learning
  
  Week 18: Introduction to Machine Learning and Learning
  
  

  • Defining artificial intelligence, machine learning and data science and their correlation
  • The history and applications of AI and machine learning
  • Future trends and approaches in the AI field
  • Learning patterns of machines
  • The machine learning process and approaches
  • Concepts of learning difficulties and lack of performance
  • Underfitting and overfitting

  
  Week 19: Supervised Learning: Classification
  
  

  • Applications of classification in business scenario
  • Explanation of dependent variable in classification
  • Continuous dependent variable versus categorical variable
  • ML techniques to predict the classes in classification
  • Demonstration using logistic regression in Python
  • Discussion of model statistics to evaluate the model

  
  Week 20: Supervised Learning: Regression
  
  

  • Regressive algorithms for implementing supervised learning with examples
  • Case study discussion: The use of supervised learning, specifically linear regression in NBA teams to improve their ranking and win a cup
  • Regressive models such as function estimation as linear regression and classification using logistic regression
  • The role of regressive models in function estimation and classification

  
  Week 21: Supervised Learning: Decision Trees
  
  

  • The theory, terminologies and applications of decision trees
  • The random forest ensemble system for high-performance decision-making and its applications
  • Machine learning concepts such as the entropy, GINI index and imbalanced classes issues

  
  Week 22: Support Vector Machine Algorithms
  
  

  • Defining support vector machine (SVM) and explaining the underlying algorithm
  • Discussing parameter tuning in SVM
  • Demonstrating SVM for classification and regression
  • Advantages and disadvantages of using SVM

  
  Week 23: Unsupervised Learning: Clustering techniques in K-Means
  
  

  • Describing the K-means clustering algorithm
  • Developing and using the K-means algorithm using Scikit Learn
  • Optimising a clustering system by changing the clustering parameters
  • VIsualising the results of clustering

  
  Week 24: Unsupervised Learning: Hierarchical Clustering Method
  
  

  • Different applications of hierarchical clustering
  • The importance of metrics and linkage criteria in hierarchical clustering
  • Top-down and bottom-up approaches of hierarchical clustering
  • The implementation of hierarchical clustering and interpretation of results

  
  Week 25: Unsupervised Learning: Probabilistic and Association Rule
  
  

  • Overview of probabilistic clustering and comparison with K-means
  • Exploring the Gaussian Mixture Model and Expectation Maximization algorithm
  • Overview of association rule and its application
  • Applying pre-processing and Apriori algorithm
  • Demonstration of association rule for market basket analysis

  
  Week 26: Neural Network
  
  

  • The theory, terminology and definition of artificial neural networks
  • Training algorithms of neural networks
  • Neural network parameters and ways to prevent underfitting and overfitting
  • Identifying a multi-layer perceptron as a solution to the problems of nonlinearity
  • The implementation of some machine learning applications in the system
  • Developing a model of biological brain as an artificial neural network
  • Examples of how to use neural networks as a function estimator
  • Setting optimal parameters to improve performance

  
  Week 27: Reinforcement Learning
  
  

  • Definitions and applications of reinforcement learning
  • Types of reinforcement learning
  • Reinforcement learning examples, such as traffic light control and personalised recommendation systems
  • Robot navigation using reinforcement learning

  
  Week 28: Deep Learning Using Keras and Tensorflow
  
  

  • Defining deep learning and its applications
  • Examining how deep neural networks deal with semi-structured and unstructured data
  • Explain types of networks and its applications
  • Exploring backward and reverse propagation
  • Describing layers in sequential neural net-hidden layers, input and output layers
  • Describing hyperparameters in neural net
  • Demonstration and application of deep learning techniques

  
  Week 29: Recommendation Systems
  
  

  • Defining recommendation systems and its uses
  • Exploring types of recommendation systems and their advantages and disadvantages
  • Exploring types of recommendation systems and their advantages and disadvantages

  
  Week 30: Challenges
  
  

  • The effects of super intelligence on society and its future
  • Examples of AI used in the real world
  • The role of AI in developing effective surveillance systems
  • Technical advantages and future trends
  • The role of quantum computing in the field of ML and big data