Postgraduate Course: Predictive Analytics using Python - Final Project (CMSE11445)
||College||College of Arts, Humanities and Social Sciences
|Credit level (Normal year taken)||SCQF Level 11 (Postgraduate)
|Course type||Online Distance Learning
||Availability||Part-year visiting students only
|Summary||In this 30 SCQF credit course project students have the opportunity to demonstrate the ability to operationalise techniques in predictive analytics with classification and regression on a topic of their interest.
Topics covered include: Introduction to Predictive Analytics using Python, Successfully Evaluating Predictive Models, Statistical Predictive Modelling and Applications,Predictive Analytics using Machine Learning.
In this 30 SCQF credit course project students have the opportunity to demonstrate the ability to operationalise techniques in predictive analytics with classification and regression on a topic of their interest.
1. Introduction to Predictive Analytics using Python
2. Successfully Evaluating Predictive Models
3. Statistical Predictive Modelling and Applications
4. Predictive Analytics using Machine Learning
Content covering the four areas listed above is made available on the edX platform. Besides videos and learning material, the platform provides assessment tools (quizzes, code checkers etc) to self-assess learning progress.
Entry Requirements (not applicable to Visiting Students)
||Other requirements|| The four edX modules completed as prior learning to the course, which require a foundation in linear algebra and programming in Python.
Information for Visiting Students
|Pre-requisites||The four edX modules completed as prior learning to the course, which require a foundation in linear algebra and programming in Python.
|High Demand Course?
Course Delivery Information
|Academic year 2020/21, Part-year visiting students only (VV1)
|Learning and Teaching activities (Further Info)
Dissertation/Project Supervision Hours 3,
Other Study Hours 174,
Programme Level Learning and Teaching Hours 6,
Directed Learning and Independent Learning Hours
|Additional Information (Learning and Teaching)
174 hours edX MicroMasters
|Assessment (Further Info)
|Additional Information (Assessment)
The evaluation is based on two pieces of assessment:
A proctored exam, testing for a holistic understanding of the theory over the four modules covered in 2 open questions of no more than 1 A4 each. The proctored part of the assessment will be done by edX through a third party (50%)
A Jupyter notebook which blends coding and text. This reflects a real-life case study which is typically delivered in the form of a proof-of-concept augmented with interpretation and a visual representation of results. The structure of both code and document can be fixed up front to allow for a streamlined evaluation. (50%)
The notebook (1,700 words) contains the following sections:
1. Interpretation of the problem (300 words)
2. Descriptive statistics and interpretation of the data and its variables (300 words)
3. Data pre-processing/cleansing/transformation (400 words)
4. Modelling and benchmark of at least 2 predictive techniques (400 words)
5. Interpretation of the results and discussion in line with the research questions (300 words)
||The student can get feedback mid-way of the term (week 3 of the module) on sections 1 and 2, which encompasses both code and text.
|No Exam Information
On completion of this course, the student will be able to:
- Identify various business settings in which predictive analytics can be used.
- Show a systematic and critical understanding of the entire predictive analytics process.
- Critically assess, develop and implement predictive models and processes.
- Identify and evaluate social, cultural, global, ethical and environmental responsibilities and issues.
|All material will be written by the course organisers and made available on Edx in the form of four Edx modules each of which provides material associated with one of the learning outcomes below.|
Furthermore, the following books can be considered for background reading:
Applied Predictive Modeling- Max Kuhn, Springer
Introduction to Statistical Learning- Daniela Witten, Gareth James, Robert Tibshirani, and Trevor Hastie, Springer
|Graduate Attributes and Skills
Information technology: numeracy and Big Data
Application of knowledge: knowledge integration and application
|Course organiser||Dr Johannes De Smedt
|Course secretary||Miss Carrie Innes
Tel: (0131 6)51 3757