Postgraduate Diploma in Data Science

Purpose

The Post Graduate Diploma in Data Science qualification is designed to develop data analysis and computational abilities in learners. The status of the business environment is being disrupted by technological developments that simplify the storage and processing of data for effective decision-making. There is a wide range of data used and generated in all walks of life. Hence, it is essential to understand and grasp the 3Vs of data, that is volume, variety, and velocity and data science techniques that enable the development of automated procedures for such incomprehensible data, including enhanced productivity, quality of work, and greater accuracy.

The primary goal of the qualification is to provide learners with a specialised grasp of data analytical skills using data science techniques in their respective domains and across a wide range of disciplines and industries.

The qualification aims to enhance the knowledge of prospective learners, give them an awareness of the data-driven world, and make them employable. This qualification is a multidisciplinary qualification which includes, learning how to use data in a business context and how massive data sets can be managed and analysed to make appropriate decisions.

Upon completion of the qualification, qualifying learners will be able to

• Understand and identify different components of the data ecosystem.
• Build interactive dashboards to analyse data retrieved from various sources and extract business insights.
• Explain Statistical, Probability and Linear Algebraic techniques for better understanding of the data.
• Analyse and synthesize raw data to draw conclusions to make data-driven decisions using Python language-based tools.
• Develop exploratory data analysis (EDA) mini project based on modules 1 to 4, using Python programming language.

Rationale

Technology has become one of the means to manage business challenges, including maintaining growth, efficiency, and effectiveness. The institution recognises the importance of professional technology education and has taken the initiative to provide a Postgraduate Diploma in Data Science at the National Qualifications Framework (NQF) level 8.

This qualification is designed to enable data scientists to manage complex data and provide solutions in the business environment globally. Having earned this credential, data scientists will be able to handle and analyse volumes of data utilising cutting-edge data science methods, including those that quickly transform and link information to artificial intelligence (AI).

The qualification includes a foundation in the data ecosystem, mathematical and statistical techniques, and the use of a variety of tools. As learners progress through, they will learn how to develop end-to-end data science applications or solutions using several data analysis and data visualisation tools based on statistical techniques. Further, they will also develop predictive models using classical machine learning algorithms and neural network-based learning algorithms. These algorithms can be applied to both structured and unstructured data to solve various business problems.

The qualification aims to assist South African youth in becoming employable across varied industries. Professionals will learn and gain advanced knowledge in implementing data analysis in business, mitigating ways within an ethical framework, and the capacity to choose the most suitable technique from which every industry will benefit as the paradigm shifts to include forward-thinking and flexibility. There is a strong demand for those with the ability to manage data and provide a company with a viable strategy for attaining its objectives.

• Learners who do not meet the minimum entrance requirements or the required qualification that is at the same NQF level as the qualification required for admission may be considered for admission through RPL.
• To be considered for admission in the qualification based on RPL, applicants should provide evidence in the form of a portfolio that demonstrates that they have acquired the relevant knowledge, skills, and competencies through formal, non-formal and/or informal learning to cope with the qualification expectations.

RPL for exemption of modules

• Learners may apply for RPL to be exempted from modules that form part of the qualification. For a learner to be exempted from a module, the learner needs to provide sufficient evidence in the form of a portfolio that demonstrates that competency was achieved for the learning outcomes that are equivalent to the learning outcomes of the module.

RPL for credit

• Learners may also apply for RPL for credit for or towards the qualification, in which they must provide evidence in the form of a portfolio that demonstrates prior learning through formal, non-formal and/or informal learning to obtain credits towards the qualification.
• Credit shall be appropriate to the context in which it is awarded and accepted.

Entry Requirements

The minimum entry requirement for this qualification is

• Bachelor of Computer Science, NQF Level 7.

Or

• Bachelor of Science in Data Science, NQF Level 7.

Or

• Bachelor of Information Technology, NQF Level 7.

Or

• Advanced Diploma in Information Technology, NQF Level 7.

This qualification consists of the following compulsory modules at National Qualifications Framework Level 8 totalling 130 Credits.

Compulsory Modules, NQF Level 8, 130 Credits.

• The Data Ecosystem, 10 Credits.
• Data storytelling using Power BI, 15 Credits.
• Data Science with Python, 25 Credits
• Statistics for Data Science, 10 Credits.
• Mini Project (Module 1 - 4), 10 Credits.
• Predictive modelling with machine learning, 25 Credits.
• Deep Learning with artificial neural networks, 20 Credits.
• Major Project, 20 Credits.

Exit Level Outcomes

1. Understand and identify different components of the data ecosystem.
2. Build interactive dashboards to analyse data retrieved from various sources and extract business insights.
3. Explain Statistical, Probability and Linear Algebraic techniques for a better understanding of the data.
4. Analyse and synthesize raw data to draw conclusions to make data-driven decisions using Python language-based tools.
5. Develop exploratory data analysis mini project based on modules 1 to 4, using Python programming language.
6. Build, train and evaluate machine learning models for regression, classification, and clustering problems.
7. Understand the structure and functioning of artificial neural network models for unstructured data like images and text.
8. Develop end-to-end data science projects encompassing all stages of data science such as data collection, data analysis, model building, evaluation and deployment.

Associated Assessment Criteria for Exit Level Outcomes 1.

• Determine the broad nature of the data and explain the 5Vs of data and data values.
• Discuss the dimensions of data and explain the data labelling concept.
• Extract data based on the occurrence of events and identify input and output data.
• Apply knowledge to analyse and visualize data.
• Illustrate and interpret data into meaningful information and assess findings.

Associated Assessment Criteria for Exit Level Outcomes 2.

• Discuss features and components of Power BI and use them to connect to different data sources.
• Explain different data types supported by Power BI.
• Use Power BI Functions for mathematical and statistical calculations.
• Select and develop appropriate charts and apply filters to get a subset of data based on certain conditions.
• Apply data transformation techniques to convert raw data into meaningful values and build data queries to fetch data from multiple data sources.

Associated Assessment Criteria for Exit Level Outcomes 3.

• Utilize statistical knowledge by having familiarity with a variety of statistical techniques.
• Determine the accuracy of statistical estimations by performing in-depth analyses of the probability of various outcomes.
• Construct statistical analysis (descriptive and inferential) to find the solutions to various business functions through the use of hypothetical testing.
• Summarize and forecast analysis using graphical methods to assist in decision-making.
• Analyse data and draw assumptions and conclusions from the respective data.
• Build and develop machine learning models using linear algebra-based vectors and matrices.
• Devise machine learning model algorithm and evaluate the machine learning models by slicing and dicing the data for new opportunities in business prospects.

Associated Assessment Criteria for Exit Level Outcomes 4.

• Discuss the principles, characteristics, and features of Python programming as well as the use of application software.
• Identify program units, data and data structures required to implement a given design.
• Design a programming solution for a given problem and select and implement control structures to meet the design algorithm.
• Apply several feature engineering techniques which include selection, manipulation, and transformation of raw data into desired features using a statistical approach.
• Apply dimensionality reduction technique on the dataset to reduce the number of columns thereby reducing the computational cost of modelling.
• Perform exploratory data analysis on the dataset to discover trends, and patterns, or to check assumptions with the help of statistical summary and graphical representations.

Associated Assessment Criteria for Exit Level Outcomes 5.

• Perform data pre-processing, data analysis and data visualization and discuss the results.
• Create a report to provide project details, results, discussion and conclusion.
• Present the topic, EDA steps, results, and conclusion.

Associated Assessment Criteria for Exit Level Outcomes 6.

• Discuss what is machine learning, the machine learning process, applications of machine learning and its types.
• Illustrate understanding of the machine learning models and formulate the problem.
• Identify the steps to perform EDA and the appropriate ML algorithm to perform model building using SKLearn.
• Select and apply appropriate model evaluation metrics to evaluate the built model and interpret the result of model evaluation.

Associated Assessment Criteria for Exit Level Outcomes 7.

• Illustrate knowledge of and list similarities between artificial neurons and biological neurons.
• Explain mathematical calculations involved in artificial neuron - summation and activation function.
• Explain different activation functions, construct or experiment with artificial neural network model by stacking layers of artificial neurons for classification and regression problems as well as functional and sequential APIs of TensorFlow.
• Experiment with hyperparameters to improve artificial neural network model accuracy.
• Apply pooling techniques to the output of the intermediate convolution layer to further extract.

Associated Assessment Criteria for Exit Level Outcomes 8.

• Perform data pre-processing, data analysis and data visualization and discuss the results.
• Create machine learning models, test, and evaluate them for data.
• Create a report to provide project details, results, discussion, and conclusion.
• Present the topic, implementation of all phases of the data science project life cycle, results, and conclusion.