2016/17 Taught Postgraduate Module Catalogue

GEOG5927M Predictive Analytics

15 creditsClass Size: 80

Module manager: Mark Birkin
Email: m.h.birkin@leeds.ac.uk

Taught: Semester 2 (Jan to Jun) View Timetable

Year running 2016/17

This module is not approved as an Elective

Module summary

This module explores and evaluates a range of modelling techniques which are used in an applied context to simulate and predict consumer behaviour. Combining theory and practical examples, students are introduced to geocomputational techniques such as clustering and segmentation, data mining, machine learning, microsimulation and agent based modelling. Applied research-led and industry case studies are embedded throughout the module enabling students to evaluate how these approaches are used in practice for predictive analytics. Students develop an in-depth understanding of modelling techniques used to simulate or forecast population attributes and behaviours. Through linked and supported practical activities, students apply these techniques in order to optimise service provision, forming the basis of their individual assessed work. Students consider how clustering and segmentation can be used for targeted marketing and predictive response modelling. Students are introduced to data mining and machine learning as tools to support the development of richer ‘agent based’ behavioural models. The module concludes with a focus on model building and evaluation in the commercial sector, including ROI and business justification. Students complete an assessed group work activity which requires them evaluate the potential value of behavioural models, operationalised through the NetLogo software, in a particular business context area. Students will identify the potential opportunities/challenges in their specific organisational context and identify the data requirements and potential behavioural insights which could be gained. Students will present their findings in a boardroom style setting in which they are pitching for funds in order to implement their strategy.

Objectives

This module seeks to:
-Introduce students to the variety of geocomputational modelling techniques used to simulate and predict consumer behaviour.

-Demonstrate how predictive analytics and behavioural modelling are used for consumer analysis and the evaluation of marketing strategies, via research-led and industry case studies.

-Give students practical hands-on experience at modelling consumer behaviours and responses in order to optimise service provision.

-Enable students to evaluate, justify and communicate the benefits, opportunities and challenges when using predictive analytics to address specific business needs

Learning outcomes
On completion of this module students will:
1.Be able to explain and critically evaluate the role of analytics and geocomputational modelling in simulating and predicting consumer behaviour, drawing on applied research-led examples and industry practice.

2.Be able to execute geocomputational modelling, and simulation using appropriate data sources and software packages.

3.Be able to identify, evaluate and justify potential ROI from applications of predictive behavioural modelling techniques to address research needs or business objectives.

4.Be able to devise a strategy for the implementation of sophisticated modelling tools to address a business scenario, presenting and justifying their recommendations in an appropriate context.

Syllabus

Predictive Analytics and applied spatial modelling
Simulating and forecasting behaviour
Clustering and segmentation
Targeted marketing and response modelling
Agent Based Behavioural models
Data mining and Machine Learning
Model building and evaluation – ROI and business justification
Predictive analytics for retail consultancy

Teaching methods

Delivery type	Number	Length hours	Student hours
Workshop	3	4.00	12.00
presentation	1	2.00	2.00
Lecture	8	2.00	16.00
Private study hours			120.00
Total Contact hours			30.00
Total hours (100hr per 10 credits)			150.00

Private study

Private study comprises core reading around lecture content and case study material to support assessed work.
Students will work independently on their service optimisation project outside of timetabled and supported sessions; undertaking wider reading, hands on modelling, interpretation of model outputs, visualisation of model results and preparation of their written academic report.
Students will also work in small groups to complete their group project outside of formal timetabled sessions. This will include independent reading, critical thought and hands-on exploration of modelling tools, alongside group meetings to prepare their recommendations and final presentation.

Opportunities for Formative Feedback

Short question and answer sessions integrated into lectures provide an opportunity for formative assessment of student progress throughout the taught component of this module. The practical sessions related to the service optimisation project enable staff and postgraduate demonstrators to assess progress of the cohort and individual students and to provide additional support and clarification where required. The service optimisation project is handed in during week 20, enabling summative assessment to take place midway through the module. Supported group work sessions provide a further opportunity to monitor student engagement and understanding prior to summative assessment.

Methods of assessment

Coursework

Assessment type	Notes	% of formal assessment
Essay or Dissertation	2,000 word academic report based on service optimisation project	50.00
Group Project	Group presentation – pitching a modelling strategy (2,000 word equivalent)	50.00
Total percentage (Assessment Coursework)		100.00

Normally resits will be assessed by the same methodology as the first attempt, unless otherwise stated

Reading list

There is no reading list for this module

Last updated: 11/03/2016

Disclaimer

Browse Other Catalogues

• Undergraduate module catalogue
• Taught Postgraduate module catalogue
• Undergraduate programme catalogue
• Taught Postgraduate programme catalogue

Errors, omissions, failed links etc should be notified to the Catalogue Team.PROD