Module Details
The information contained in this module specification was correct at the time of publication but may be subject to change, either during the session because of unforeseen circumstances, or following review of the module at the end of the session. Queries about the module should be directed to the member of staff with responsibility for the module.
Title PREDICATIVE ANALYTICS FOR DECISION MAKING
Code CKIT526
Coordinator Prof FP Coenen
Computer Science
Coenen@liverpool.ac.uk
Year CATS Level Semester CATS Value
Session 2020-21 Level 7 FHEQ Whole Session 15

Aims

·          To provide students with a systematic understanding of key predictive analytics techniques. ·          To provide students with a comprehensive understanding of the types of business problems predictive analytics can solve. ·          To provide students with a pratical understanding of the process of applying the tools and techniques of predictive analytics in commercial environments. ·          To provide students with an opportunity to apply the tools and techniques learnt on the module to typical business problems.

Learning Outcomes

(LO1) A critical knowledge of key predictive analytics techniques.

(LO2) A deep and critical understanding of how to match appropriate predictive analytics technique to a given business related problem.

(LO3) A significant ability to translate results from a predictive analytics process into business actions.

(LO4) An ability to deploy the end-to-end process of predictive analytics in commercial settings.

(S1) Organisational skills

(S2) Communication skills

(S3) IT skills

(S4) Communication and collaboration online participating in digital networks for learning and research

(S5) Learning skills online studying and learning effectively in technology-rich environments, formal and informal

(S6) Team (group) working respecting others, co-operating, negotiating / persuading, awareness of interdependence with others

Syllabus
 

Week 1 Common predictive analytics techniques (e.g. logistic regression, classification trees, collaborative filtering) and model development best practices.     Week 2 The End-to-end process of predictive analytics in the context of commercial environments: (i) problem identification, (ii) data assessment, (iii) data preparation, (iv) model development, (v) model deployment and (vi) making insights available to decision makers.   Week 3 Application of predictive analytics scoring models I: customer acquisition. Use case on optimising direct marketing campaigns utilising lists of scored prospects.     Week 4 Application of predictive analytics scoring models II: customer retention. Use case on decreasing customer churn.   Week 5 Application of predictive analytics scoring models III: operational efficiency. Use case of optimising collection agent’s time by prioritising outreach to "high risk customers".   Week 6 Application of pre dictive analytics scoring models IV: recommender systems. Use case on product recommendations in an online store.     Week 7 End-to-end application of predictive analytics techniques in business environments I.   Week 8 End-to-end application of predictive analytics techniques in business environments II (continued from week 7).

Teaching and Learning Strategies

Teaching Method 1 - Online Learning
Description: Weekly seminar supported by asynchronous discussion in a virtual classroom environment facilitated by an online instructor.
Attendance Recorded: Yes
Notes: Number of hours per week that students are expected to attend the virtual classroom so as to participate in discussion, dedicated to group work and individual assessment is 7.5.

Teaching Schedule
  Lectures Seminars Tutorials Lab Practicals Fieldwork Placement Other TOTAL
Study Hours           60

 60
Timetable (if known)              
Private Study 90
TOTAL HOURS 150

Assessment
EXAM Duration Timing
(Semester)		 % of
final
mark		 Resit/resubmission
opportunity		 Penalty for late
submission		 Notes
             
CONTINUOUS Duration Timing
(Semester)		 % of
final
mark		 Resit/resubmission
opportunity		 Penalty for late
submission		 Notes
eight discussion questions Standard UoL penalty applies for late submission. This is not an anonymous assessment. Assessment Schedule (When) :Whole session  Weekly Discussion Qu    40       
Import and prepare for modelling a data set using RapidMiner Standard UoL penalty applies for late submission. This is not an anonymous assessment. Assessment Schedule (When) :Week 1  750-1000 word report    8       
Explore a data set using RapidMiner’s statistical and visualisation features Standard UoL penalty applies for late submission. This is not an anonymous assessment. Assessment Schedule (When) :W  750-1000 word report    8       
Design, build and validate an ensemble classification model Standard UoL penalty applies for late submission. This is not an anonymous assessment. Assessment Schedule (When) :Week 4  1250-1800 word repor    14       
Improve the performance of a logistic regression model Standard UoL penalty applies for late submission. This is not an anonymous assessment. Assessment Schedule (When) :Week 5  750-1000 word report    8       
Design, build and validate a product recommender model Standard UoL penalty applies for late submission. This is not an anonymous assessment. Assessment Schedule (When) :Week 6  750-1000 words repor    8       
Design, build, and validate a customer segmentation (clustering) model Standard UoL penalty applies for late submission. This is not an anonymous assessment. Assessment Schedule (When) :Week 8  1500 word report des    14       

Recommended Texts
Reading lists are managed at readinglists.liverpool.ac.uk. Click here to access the reading lists for this module.