Fundamentals of Data Science (Technical) - an online course

About this course

This course equips you with the theoretical knowledge and technical skills to apply the powerful insights of data science to your work, as well as providing the foundations for a career in Data Science.

Apply now

Course overview
Aims and learning outcomes
Technology stack
Assessment
Syllabus
Fees
Testimonials
Supplementary tuition

This course has been exclusively designed from the ground up by the world-leading Data Science team at the University of Southampton.

Welcome to the Fundamentals of Data Science – Technical, a course designed to introduce you to theoretical components of data science and give you hands-on experience of producing solutions to data science problems through the use of Python. Assessment for this course is based upon three practical assignments which are presented in the browser based Jupyter notebook environment.

Structure

The course is broken into six weeks.

In week 1, you will meet your tutor and the other participants on this course and find out more about what you will be doing over the next 6 weeks and how we will be supporting you. You will get "hands-on" experience of Jupyter the web-based environment which you will use for the course exercises and assignments. This week also contains a Python Primer activity for those of you who are unfamiliar with the programming language or would like a refresher.

In week 2, you will learn about the fundamental terminology and processes in data science, discovering the technology landscape that has helped fuel the data explosion, and the tools that data scientists use to unlock the hidden value in these vast amounts of data. This week also contains an introduction to using Python for data science.

You will begin gaining hands-on experience of data science in week 3, focusing on collecting, storing and managing data, and you will learn about the different sources of data and how they can be combined in order to increase the potential insights available.

Week 4 will then help you understand how this data is analysed, covering a range of techniques that a data science team would typically use, from statistics to machine learning. You will use Python to apply these analytical techiques to a real-world dataset.

In week 5, you will learn about how the findings from data science work can be reported using different data visualisation techniques. You will discover the various ways in which particular types of data can be displayed in order to highlight a key finding and improve the impact of your reports.

In week 6, you will be looking at the future of data science and we will be focusing on supporting you to finish off your assignments.

Hands-on experience and assignments

Each week contains a mix of taught material, self-study material, activities and practical online exercises.

Week 1 includes an (optional) introduction/'refresher' on Python. This includes online exercises for you to work through at your own pace. This not graded. Week 2 contains further Python practice exercises. You are encouraged to do these as these will help you with your assignments. Again, these practice exercises are not graded.

Weeks 3, 4 and 5 each include online exercises (ungraded) and a related graded coursework assignment.

To find out more, fill out the enquiry form or email us on info@southamptondata.science

Aims and learning outcomes

This course will provide you with the knowledge and expertise to become a proficient data scientist.

Having successfully completed this module, you will be able to:

Understand the key concepts in data science, including their real-world applications and the toolkit used by data scientists;
Explain how data is collected, managed and stored for data science;
Implement data collection and management scripts using MongoDB;
Demonstrate an understanding of statistics and machine learning concepts that are vital for data science;
Produce Python code to statistically analyse a dataset;
Critically evaluate data visualisations based on their design and use for communicating stories from data;
Plan and generate visualisations from data using Python and Bokeh.

Upon successful completion of the course, you will be awarded with a Certificate of Completion and a graded transcript.

Visualising	Bokeh (Python)
Stats / Analysis	NumPy/SciPy/Pandas
Management / Querying	MongoDB (using Python)
Base	Python

Assignment 1 - Libraries
Requests, Json, BS4 – BeautifulSoup, urllib, robotparser – RobotFileParser, Pymongo – MongoClient.

The first assignment asks you to collect data from an API then process the data for a specific output. To complete this assignment, you will have access to the Python libraries listed below though you may wish to use others.

Assignment 2
Numpy, Pandas, Bokeh – charts, Scikit-learn: model_selection – train_test_split, linear_model – LinearRegression, metrics – mean_squared_error.

For the second assignment you are asked to create functions using the following libraries to produce solutions for problems through statistical analysis and creation of machine learning algorithms. The Python libraries suggested for use in producing solutions are listed below.

Assignment 3
Ipywidgets: interact, interactive, Hbox, Label, Slider; Bokeh: palettes - Spectral6, io – output_notebook, output_file, show, push_notebook, curdoc; models: TapTool, CustomJS; models.sources: ColumnDataSource; models.tiles – WMTSTilesSource, tile_providers; Bokeh.io – output_file, output_notebook, show and push_notebook; Ipython: display.

The third and final assignment requires you to create a user friendly and useful interactive visualisation of data which you have collected in your previous work. For this assignment a set of libraries and packages are imported for you though you are free to use other libraries and tools.

With experience of using all of these libraries and such concepts as lists, dictionaries, series, dataframes and .loc you should be able to complete this course with ease within the 10 hours per week of study recommended. If you are aware of the majority of these libraries but have only limited experience of using a few of them the recommended 10 hours study per week should suffice if coupled with weekly attendance to individual and group tutorials.

With limited awareness or understanding of these libraries or the use of Python as a programming language you may still complete the course successfully by choosing one or more of the following suggested options.

If you have limited programming skills there is the opportunity to purchase intensive support in the form of weekly 1-to-1 tutorials in which the tutor will walk you through the course practice material focusing on understanding and application. You may decide to brush up on your programming skills by taking part in one of the many online free Python courses available e.g. Programming with Python and Plotting and Programming with Python offered by the Software Carpentry Foundation.

A final option is to consider the Fundamentals of Data Science (Non-Technical) course with a focus on the data upon which you may be considering performing data science which does not involve any programming.

Week 1: Welcome and course information

Topics

Welcome and introduction
Learning outcomes of the week
What data science is and why it's important
Course syllabus and learning outcomes
Using discussion forums
Introduce yourself
Help and tutoring support
Course assignment details
A 'hands-on' Jupyter familiarisation activity
Python Primer
Glossary of terminology

Week 2: Introduction to core concepts and technologies

Topics

Introduction
Learning outcomes of the week
Data science in a nutshell
Terminology
The data science process
A data science toolkit
Types of data
Example applications
Further reading
Summary

Week 3: Data collection and management

Topics

Introduction
Learning outcomes of the week
Sources of data
Data collection and APIs
Exploring and fixing data
Data storage and management
Using multiple data sources
Further reading
Summary

Week 4: Data analysis

Topics

Introduction
Learning outcomes of the week
Terminology and concepts
Introduction to statistics
- Nature of statistics and introduction
- Central tendencies and distributions
- Variance
- Distribution properties and arithmetic
- Samples/CLT
Basic machine learning algorithms
- Linear regression
- SVM
- Naive Bayes
Further reading
Summary

Week 5: Data visualisation

Topics

Introduction
Learning outcomes of the week
Types of data visualisation
- Exploratory
- Explanatory
Data for visualisation
- Data types
- Data encodings
- Retinal variables
- Mapping variables to encodings
- Visual encodings
Technologies for visualisation
- Bokeh (Python)
Further reading
Summary

Week 6: Future of data science

Topics

Introduction
Learning outcomes for the week
The future of data science

Fees

The Fundamentals of Data Science (Technical) course is £1500 per person, inclusive of VAT.

For corporate packages, please see here.

Extra tuition support is £400, inclusive of VAT.

How to Pay

You can pay by phone, email or Flywire, using the application form here.

Paying Online

To make a payment using this method, fill out the application form and select "Pay by Credit / Debit Card".
A 2% fee is charged for payment by credit card and we do not accept American Express.
Fees paid by this method will be charged in British pounds sterling.

Paying by Phone

To make a payment using this method, fill out the application form and select "Pay by Phone".
A 2% fee is charged for payment by credit card and we do not accept American Express.
Fees paid by this method will be charged in British pounds sterling.

Pay by Email

To arrange a payment using this method, please contact us once you have received confirmation of your place. Contact your course advisor, agent, or email us on payments@southamptondata.science

Flywire

To make a payment using this method, fill out the application form and select "Pay by Flywire".
Best for international participants: accepting over 70 currencies via credit card, debit card, or bank transfer.
Simply visit flywire.com/pay/southamptondata and follow the instructions on the website. Please use the same email address that you used when you applied to the course.

Extra tuition support

All of our courses are tutor-led, designed to develop your knowledge and competency as quickly as possible. The standard fee includes access to weekly guided seminars and support from your tutors throughout the course.

For ambitious students who wish to take the course, but whose knowledge of Python or statistics is less complete, we can offer an additional pack of five intensive one-hour tutorials. These supplementary tutorials will help you work through any issues you may encounter on the course, allowing you to rapidly gain confidence in deploying Python for Data Science.

In this section

Course Facts

Duration: c. 60 hours (across 6 weeks)

Dates:
2nd July - 10th August
3rd September - 12th October
5th November - 14th December

Course lead: Professor Elena Simperl

100% Online Delivery: Personal and group tutorials, guided self-study via core materials, Q&A, peer discussions, and supported code implementation including coursework

Prerequisites: A decent understanding of programming is required; some experience with Python or with a similar language is recommended.

Target group: Anyone wanting to start their career in Data Science, or looking to learn how to apply the powerful techniques of Data Science to their current business challenges

Learning materials: Video tutorials, presentations, online exercises, further reading

Assessment and feedback: Assignments (3 pieces), with expert tutors to provide guidance and formative feedback throughout the course

With these new online courses, we reach out to decision makers, business analysts, IT developers, and other groups of practitioners interested in learning about some of the most promising technologies and application domains in data science, including digital marketing, finances, and healthcare
Dr Elena Simperl, CPD Programme Leader, WSI Associate Professor and Scientific Director of the European Data Science Academy
Coming from a GIS background this course has really broadened my capabilities within Data Science and filled the gaps in my skills that I need to evolve my career.
GeoSpatial Technology & Information Management Consultant, ARUP
The flexible model of the course lets us balance staff training alongside our commitments to our clients.
Data and Systems Architect at ARUP
The flexible nature of the course is a great idea given that many of the clientele are working in full time jobs
Head of Data Science, QBE Insurance
I was very happy to see the more modern languages and tools being used in the course as these are the industry standard. The course has helped me reassess the way we approach things at QBE from both a technological and infrastructure standpoint.
Analyst, QBE Insurance
I really liked the development environment of the course as it was flexible, easy to use and great for experimentation. The course came at the perfect time as we were starting to move from SQL to NoSQL so this element of the course was very interesting and I learnt a lot that will help.
Technical Consultant, Ordnance Survey
I’ve already noticed that after 4 weeks I’m speaking more confidently and know where to look for things. The online platform was easy to use and intuitive with course dates, requirements, deadlines etc very clear and in one place; a great infrastructure!
If I was applying for a Data Science job I’d want this on my CV
Statistician, Office for National Statistics
As our staff are very geographically dispersed, this online option and modular approach with its accessibility was of huge importance.
Assistant Deputy Director- Government Statistical Service Profession at Office for National Statistics
This course has given me a much broader knowledge of what can be done with data and I’m much better prepared to work with data than before.
Data Analyst, Land Registry
It was great that our Data Analyst was still able to his job whilst learning.
Head of Data Capture and Management, Land Registry
Getting access to the tutors is one of the great benefits of this course; not all courses provide this. I’ve seen that this course has helped our Digital Insight Analyst realise that the transitions he wanted to make are very achievable. The flexible method is great. It’d be difficult to commit to a set time on set days; our Digital Insight Analyst has been able to fit around his lifestyle and work commitments without any negative impacts.
Digital Analytics Manager, Shop Direct
As I’m mostly self-taught this course has made me aware of the gaps in my knowledge. I’ve been able to start thinking how I can put these new skills to use in my own work, including bringing in other data sets, using web scraping tools and validating outputs.
Digital Insight Analyst, Shop Direct