DATA SCIENCE
DATA SCIENCE
STUDY GUIDE
STUDY GUIDE
LSS Data Science Stream: Study Guide
Round 1: Data in the Real World
This introductory session bridges the gap between raw numbers and real-world
impact. We will explore how data science drives decision-making across industries
—from optimizing sports performance and predicting weather patterns to
curating your favorite entertainment feeds. The second half of the session
establishes the essential mathematical and statistical foundations you will use in
every subsequent module throughout the rest of the workshops.
Round 2: Research Papers
Data science does not exist in a vacuum; it builds upon existing knowledge. This
session focuses on research literacy, equipping you with the skills to effectively
read, break down, and extract methodologies from academic papers to inform
your own work. Instead of just teaching you how to write a passive summary, we
will work through practical examples together. You will learn how to seamlessly
integrate prior literature into your own thesis, ensuring your analysis is grounded
in proven methodologies and clear academic context.
Round 3: Coding with Python
Roll up your sleeves for a hands-on coding lab. This session transitions you from
theory to implementation using Python, the industry-standard language for data
science. Working within a shared Google Colab notebook, you will learn basic
syntax and dive straight into data preprocessing. We will teach you how to clean
data so it can be used properly. We will also teach basic Python syntax, then move
on to essential Python libraries used in data science. You will also learn how to
responsibly leverage AI as a co-pilot to debug and understand unfamiliar code
without developing a reliance on it.
Thesis Introduction: At the end of this session, the final thesis project will be
unveiled. You will be introduced to three distinct datasets, allowing you to choose
your path and begin sourcing relevant research papers.
Round 4: Identifying Trends
It’s time to merge your coding skills with your statistical toolkit. In this highly
interactive session, you will work with a live dataset to determine which variables
hold actual predictive power and how they correlate with one another. You will
visually compare data plotted with and without correlation in mind to witness
firsthand how easily data can be misinterpreted. Throughout the session, your
understanding will be tested. You will be asked to generate specific plots on the fly
and critically evaluate whether your visualizations make logical sense. We will
conclude the interactive rounds with a fast-paced review game, testing your
retention of the math, coding, and trend identification concepts covered
throughout the stream.
The Thesis Project
The culmination of the Data Science stream is a short, open-ended research
thesis. You will select one of three real-world datasets, each paired with a guiding
research question. While each dataset provides a starting question, the angle,
interpretation, and depth of your investigation are entirely up to you. You will be
expected to clean the data, execute exploratory code, apply your statistical toolkit,
and draw meaningful, well-supported conclusions.
Grading Criteria
Your thesis will be evaluated across four core dimensions. These will be explained
in detail during the workshops.
Final Result (40% of your total grade): Examiners will look closely at the clarity
and logic of your final conclusion. Your grade depends on whether your
findings make sense, how thoroughly they are supported by your data, and
your ability to articulate the broader, real-world takeaway.
Data Handling (30% of your total grade): This component evaluates the
structural integrity of your project. You will be graded on the cleanliness of
your dataset, the efficiency and correctness of your Python implementation,
and your ability to clearly explain your programming choices.
Research Question (20% of your total grade): This section assesses your initial
conceptual approach to the project. Marks are awarded based on how deeply
you interpreted the prompt, the analytical framework you established, and the
validity of your initial hypotheses.
Mathematics (10% of your total grade): The final portion evaluates your
technical application. Examiners will look for the accurate, intentional, and
insightful integration of statistical and mathematical concepts to extract
meaningful insights from the data.
* May be subject to change on the day of the event
Prior Knowledge
Delegates are recommended to familiarize themselves with the basics of Python,
NumPy, and Matplotlib before the workshops. Some supplementary links attached
below.
[Link]
[Link]
END OF STUDY GUDE.