NCC Expert Short Course in Data Science

Course Overview

The Expert Short Course in Data Science is designed to provide an in-depth understanding of advanced data science methodologies and techniques. This course aims to equip participants with comprehensive knowledge and skills in data science, including data processing, optimisation, big data concepts, predictive analytics, and machine learning. The course is suitable for individuals aiming to advance their careers in data science or related fields.

Entry Requirements

Participants should have a solid foundation in data science principles and basic programming skills. Previous completion of intermediate-level data science courses is recommended.

Syllabus Content

Data Science Process
- What is Data Science?
- What do Data Scientists do?
- How to be a Data Scientist?
- What should you learn?
- Data Science Process
- Data Science Ethics
- Data Scientific Method
  Learning Outcomes: 1, 2
Optimisation and How to Formulate the Problem
- Optimisation & Operation Research
- Mathematical Optimisation Problem
- Classification of Optimisation Problems
- Integer Programming Problem
- Stochastic Programming Problem
- Linear Programming I: Simplex Method
- Standard Form of a Linear Programming Problem
- Characteristics of a Linear Programming Problem
- Transformation of LP Problems into Standard Form
  Learning Outcomes: 3, 4
Big Data Concepts
- What is Big Data?
- Big Data Characteristics
- Big Data: 6V’s
- Cloud Computing
- Why Study Big Data Technologies?
- Big Data Open Source Tools
- Philosophy to Scale for Big Data Processing
- What is Hadoop?
- Why use Hadoop?
- Core Parts of a Hadoop Distribution
- Common Hadoop Distributions
  Learning Outcomes: 3
Data Mashups and Big Data Infrastructure
- Data Analytics and Mash-up
- Data/Web Mashup
- Architecture of Web Mashup
- Implementation Architecture
- Why Cloud Computing?
- Advantages and Disadvantages of Cloud Computing
- How Cloud Computing Works
- Challenges of Cloud Computing
- Layers of Cloud Computing
- Components of Cloud Computing
- Big Data
- Hadoop Architecture
- Hadoop with Big Data
- Map Reduce
- Why Data Analytics?
- Types of Data Analytics
- Big Data Analytics
  Learning Outcomes: 5
Data Collection with Web Scraping Tools
- What’s Wrong with Scraping?
- Before Writing a Scraper
- Alternatives to Scraping
- Why Scrape Web?
- Harvesting Options
- Watch Out for Spider Traps!
- Web Scraping, the Easy Way
- Document Object Model (DOM)
- Regular Expressions
  Learning Outcomes: 6
Introduction to Predictive Analysis and Prediction to Multiple Regression Model
- Data Mining Foundations
- Problem Types and Model Paradigms
- Model Performance Considerations
- Recognising a Strong Model
- K-Nearest Neighbours Classification
- Decision Tree Models
- K-Means Clustering
- Multiple Regression
- Formal Definition of the Model
- Estimating the Parameters of the Model
- Analysis of the Variance Table
- F-Test for the Overall Fit of the Model
- Interval Estimation
- Selecting the Best Regression Equation
- Example: Sales Forecasting
- Interpreting the Final Model
  Learning Outcomes: 7
Introduction to Logistic Regression
- What is Logistic Regression?
- Questions
- Assumptions
- Background
- Plain Old Regression
- An Alternative – The Ogive Function
- The Logistics Function
- The Logit
- Conversion
  Learning Outcomes: 7
Time Series & Predictive Modelling Techniques
- Objective of Time Series Analysis
- Classical Decomposition: An Example
- Trend
- Residuals
- Trend and Residual Variation
- Time Series Models
- Gaussian White Noise
- Random Walk
- Time Series Modelling
- Nonlinear Transformations
- Differencing and Trend
- Differencing and Seasonal Variation
- What are Predictive Analytics?
- Why Predictive Analytics?
- What Do Predictive Analytics Do?
- Predictive Modelling Techniques
  Learning Outcomes: 7
Data Mining, Data Exploration, Cleaning, and Visualisation
- What is Data Mining?
- Data Mining Definitions
- Why Mine Data? Commercial Viewpoint
- Why Mine Data? Scientific Viewpoint
- Database Processing vs. Data Mining Processing
- Data Mining: Classification Schemes
- Decisions in Data Mining
- Data Mining Tasks
- Data Mining Models and Tasks
- Classification: Definition
- Classification: Applications
- Classification Techniques
- Why Do We Need Data Mining?
- The Data Analysis Pipeline
- Data Quality
- Sampling
- Types of Sampling
- A Data Mining Challenge
- Data Collection
- Mining Task
- Exploratory Analysis of Data
- Post Processing
  Learning Outcomes: 8
Data Transformation and Reduction

Data Transformation
Data Transformation Challenges
Data Transformation Strategies
Normalisation
Normalisation Methods
Data Reduction Strategies
The Curse of Dimensionality
The Curse of Dimensionality: Principal Components Analysis
The Curse of Dimensionality: Factor Analysis
Data Sampling
Binning and Reduction of Cardinality
Learning Outcomes: 9

Cluster Analysis

What is Cluster Analysis?
Types of Data in Cluster Analysis
A Categorisation of Major Clustering Methods
Partitioning Methods
Hierarchical Methods
Grid-Based Methods
Model-Based Clustering Methods
Outlier Analysis
Learning Outcomes: 8

Decision Tree and Model Evaluation & Deployment

Decision Tree and Classification Task
Building Decision Tree
Built Decision Tree Algorithm
Node Splitting in BuildDT Algorithm
Entropy and Its Meaning
Decision Tree Induction Techniques
CRISP-DM Phases
Modelling
- Select Modelling Technique
- General Test Design
- Build Models
- Access Models
Evaluation
- Evaluate Results
- Review Process
- Determine Next Steps
Deployment
- Plan Deployment
- Plan Monitoring and Maintenance
- Produce Final Report
- Review Project
  Learning Outcomes: 9

Assessment Strategy

Assessment includes:

Coursework
Various practical assignments and projects.

Examinations
Formal exams to test understanding of theoretical concepts.

Formative Assessments
Continuous feedback throughout the course for improvement.

Summative Assessments
Contribute to the final grade and assess overall knowledge and skills gained.

Career and Professional Development

Graduates can pursue roles such as Data Scientist, Machine Learning Engineer, Big Data Analyst, and Data Mining Specialist. This course also provides a strong foundation for further academic research or professional certifications in data science.

Support for Student Learning:

Students will have access to:

Personal Tutors
Available for guidance and support throughout the course.

Learning Resources
Online libraries, tutorials, and study materials.

Discussion Forums and Live Sessions
For interaction with peers and instructors.

Total Qualification Time:

Approximately 200 hours (including guided learning and self-study).

Conclusion

The NCC Expert Short Course in Data Science equips learners with advanced methodologies, tools, and skills to excel in data-driven environments. With comprehensive knowledge in predictive analytics, big data, and machine learning, graduates are well-prepared to tackle complex data challenges, pursue advanced roles in data science, or continue with further academic research, making a significant impact in the rapidly evolving data science landscape.

Entry Requirements

Participants should have a solid foundation in data science principles and basic programming skills. Previous completion of intermediate-level data science courses is recommended.

Menu

Menu