Foundation of Data Science

Programme Outline

Learning Objectives

1. Understand what is Big Data Analytics, how it is used and where it can be applied.
2. Prepare, analyse, identify business insights, and apply data science and big data analytics.
3. Develop, evaluate testing methods for statistical model.
4. Demonstrate through a presentation on how they have designed and applied a data analysis project from start to finish and the key results of the analysis and insights gained.

Day 1 – The Data Science Ecosystem

Data Science Introduction
Data Science is one of the hottest topics across industries, understanding what data is and its applications is vital to businesses looking to understand, predict and evaluate business decisions. Be acquainted with the world of Data Science as you immerse yourself through the workflow of Data Science and its Pipeline, even by converting a simple office tool like Excel you can harness its capability and leverage on its potential to embark on your own Data Science Mini Expedition.

Stakeholder Analysis
Often the process of identifying the crucial people before embarking Data Science project is essential. These people may influence or are indirectly impacted by your project downstream. It requires grouping them according to the various paradigm of stakeholders based on various participation levels, interest and influence.

Data Discovery
The collection and analysis of data from various sources to gain insight from hidden pattern and trends, which is a crucial step for critical business decision.

Data Wrangling
Data comes in all shapes and sizes, they can be unstructured or structured and come from all venues. Data Wrangling helps with “cleaning” or mapping of the raw data into more meaningful formats that can then be translated into more valuable data for the consumption of analytics purposes.

Data Modelling
Companies have benefited from leveraging upon the Data Science Models and Algorithms to make a critical business decision. This has upped the maturity of analytics capabilities to greater heights and has placed organisation at a greater advantage compared to the competitors, who are still in its infancy phase or the standard reporting stages. By positioning an organisation towards a more mature analytics phase have not only allowed the companies to reap immense growth and opportunities against its competitor, but also change the way we perceived the Data Science Modelling and Algorithm. This has led to various breakthrough and advancements away from the traditional algorithm approach such as the use of Deep Learning algorithm to fulfil a complex task and exemplifying the way we think about the traditional statistics.

Day 2 – Interactive Data Exploratory phase with preparation

Embarking on a Data Science Journey
The knowledge gained from Day 1, will allow participants to better appreciate the various means to kick off a Data Science project. Participants will be exposed to a methodical way of initiating a problem statement and have a better understanding of an industrial hands-on approach to a case study. The understanding of adopting an empirical approach towards a data science project will better help practitioners leverage on a data-first mindset whenever it comes planning a critical project.

Data Exploratory Process
It entails the use of various interactive data exploratory approach as a means to characterise the existing dataset and better understand the dataset with Interactive Data Exploratory analysis with Excel. In this course, you will be exposed to the power of charts, descriptive statistics, data pivot and powerful yet simplistic functions.

Data Preparation
Data Preparation entails a process of cleaning, treating special types of data and handling its unstructured form into something that is more ingest-able by the algorithm, which is later used as a cleanly formatted dataset within the subsequent lifecycle of Data Science. Often it is infamously hailed as the most painful process by a few Data Scientists, who would eventually try to capitalise on all-in-one tool and package which are off-the-shelf than methodically stripping parts of the wrangling process manually. Indeed it is unavoidable no matter at which phase of an organisation’s maturity in terms of analytics capability or at which stage of advancement within the Data Science Modelling and Algorithm.

Statistical Validation on Dataset
Validating dataset with statistics is often important yet technical enough to turn off some of the data analytics professional. However, the same cannot be said for those who find statistics useful in particular to validate the normality of the dataset or to handle the biasedness is the process of sampling a dataset. This can help experts avoid the pitfalls in the later phase of the data science lifecycle, which as the old adage says “garbage in garbage out”.

Day 3 – Regression Modelling

Regression Modelling
Regression Modelling has been coined as the most common algorithm of all model paradigms. It has been often used in various paradigms from statistical analysis to machine learning and even advanced analytics or as a backtester option for portfolio analysis. It comes in all shapes and sizes from multiple linear regression to its family of regression such as logistic regression or lasso. In this short course, we will explore the intuition of regression and the role it plays from various use cases.

Day 4 – Data Storytelling with Visualisation

Data Visualisation and Storytelling
Being able to communicate data to different stakeholders is critical, coupled with time-poor executives and shortened attention-span of the audience, data visualisation if done right can help the audience understand the data better and quicker.

Data storytelling takes data visualisations and turn them into narratives which helps the audience to quickly understand the insights and increase the impact of your data.

Day 5 – Project Presentation

Project Presentation
Participants will demonstrate through a presentation on how they have designed and applied a data analysis project from start to finish and the key results of the analysis and insights.