COMP47580 Recommender Systems & Collective Intelligence

Academic Year 2024/2025

Recommendation technologies have become an important part of our online experiences, helping us to discover books, movies, and music that are relevant to our likes and preferences. So much so, in fact, that recommender systems are now a fundamental component of most e-commerce platforms, streaming services, and other content sites. At their core recommender systems operate by learning about the likes and dislikes of individuals and groups of users so that they may proactively tailor content for these users.

In this course we will cover the fundamentals of recommender systems technologies including the main approaches to building and evaluating recommender systems (content-based vs collaborative filtering vs hybrid approaches) as well as a variety of more advanced topics.

This module will be assessed by continuous assessment only which will take the form of practical projects, report writing and an in-class test related to the development of recommender systems technologies.

Please note that proficiency in the Java Programming Language is required. There is a significant software engineering effort required and so students must be comfortable and proficient in developing complex programs using advanced tools and techniques.

Show/hide contentOpenClose All

Curricular information is subject to change

Learning Outcomes:

On successful completion of this module the learner will be able to:

- Understand the typical recommender system architecture and recommendation tasks.
- Understand core algorithms driving common recommender systems including the pros and cons of each.
- Learn about different approaches to evaluating recommender systems, using a variety of metrics and methodologies.
- Learn about more contemporary recommender systems research covering a variety of more advanced topics.

Student Effort Hours: 
Student Effort Type Hours
Lectures

24
Practical

14
Autonomous Student Learning

82
Total

120
Approaches to Teaching and Learning:
This module will involve a combination of lectures and active/task-based learning.
 
Requirements, Exclusions and Recommendations
Learning Requirements:

Please note that proficiency in the Java Programming Language is required. There is a significant software engineering effort required and so students must be comfortable and proficient in developing complex programs using advanced tools and techniques.


Module Requisites and Incompatibles
Incompatibles:
COMP30490 - Collective Intelligence, COMP40320 - Recommender Systems, COMP41440 - Collective Intelligence


 
Assessment Strategy  
Description Timing Open Book Exam Component Scale Must Pass Component % of Final Grade
Practical Skills Assessment: Focuses on the implementation and evaluation of various recommender systems algorithms. n/a Alternative linear conversion grade scale 40% No

30
Report(s): Focuses on the analysis and discussion of results of experiments carried out on various recommender systems algorithms. n/a Alternative linear conversion grade scale 40% No

30
Exam (In-person): In-class examination assessing all topics covered in the module. n/a Alternative linear conversion grade scale 40% No

40

Carry forward of passed components
No
 
Resit In Terminal Exam
Summer No
Please see Student Jargon Buster for more information about remediation types and timing. 
Feedback Strategy/Strategies

• Feedback individually to students, post-assessment
• Online automated feedback

How will my Feedback be Delivered?

CA component grades will be communicated online to students during the trimester. During practical sessions, a teaching assistant and demonstrators will be available to provide assistance and feedback to students on their work. Individual students may make appointment for face-to-face post-assessment feedback with module coordinator.

Name Role
Huan Chen Tutor