Modules

Core Module Information
Module title: Computational Intelligence
SCQF level: 11:
SCQF credit value: 20.00
ECTS credit value: 10
Module code: SET11127
Module leader: Kevin Sim
School School of Computing, Engineering and the Built Environment
Subject area group: Computer Science
Prerequisites

There are no pre-requisites for this module to be added
Description of module content:

In this module you will:- develop an awareness of the use of modern computational intelligence techniques;- develop an understanding of a number of techniques such as evolutionary algorithms, neural networks, co-evolution, classifier systems and fuzzy logic;- gain practical experience of implementing and evaluating computational intelligence algorithms.
Learning Outcomes for module:

Upon completion of this module you will be able to

LO1: Critically evaluate the use of soft computational intelligence techniques.

LO2: Critically assess and contrast the use of differing techniques to achieve particular functionalities.

LO3: Specify, implement and customise a computational intelligence algorithm.

LO4: Apply appropriate evaluation techniques in order to analyse an algorithm’s effectiveness.

LO5: Present the results in an appropriate form and draw appropriate conclusions.
Full Details of Teaching and Assessment
2024/5, Trimester 2, In Person,
VIEW FULL DETAILS
Occurrence: 001
Primary mode of delivery: In Person
Location of delivery: MERCHISTON
Partner:
Member of staff responsible for delivering module: Kevin Sim
Module Organiser:


Student Activity (Notional Equivalent Study Hours (NESH))
Mode of activityLearning & Teaching ActivityNESH (Study Hours)NESH Description
Face To Face Lecture 20 Weekly lectures running from week 2 covering the topics Evolutionary Algorithms Neural Networks Fuzzy Logic Decision Trees Statistical Methods
Face To Face Practical classes and workshops 20 Practical classes supporting the material covered in the lectures. A range of different programming languages and frameworks are used to provide practical experience of the computational intelligence techniques covered on the module.
Online Guided independent study 160 Guided independent study. You are expected to conduct independant research covering the module topics outside the timetabled classes.
Face To Face Centrally Time Tabled Examination 2 The exam is online and open book via the University Exam Server in either week 14 or 15. Students are asked to answer 3 out of 5 questions which could cover any of the topics included in the module. Questions are broken down into multiple sections requiring either longer subjective narrative answers or on occasion short factual answers. The Exam covers learning outcomes 1 & 2.
Total Study Hours202
Expected Total Study Hours for Module202


Assessment
Type of Assessment Weighting % LOs covered Week due Length in Hours/Words Description
Centrally Time Tabled Examination 40 1~2 Exam Period HOURS= 2 The exam is online and open book via the University Exam Server in either week 14 or 15. Students are asked to answer 3 out of 5 questions which could cover any of the topics included in the module. Questions are broken down into multiple sections requiring either longer subjective narrative answers or on occasion short factual answers. The Exam covers learning outcomes 1 & 2.
Report 60 1~2~3~4~5 Week 12 , WORDS= 20 hours You are supplied with a basic framework in Java that implements some of the basic functionality required to set up a neural network and implements a skeleton EA. At minimum you must: • Add missing evolutionary operators (selection, crossover, replacement etc.)• Evaluate the performance of your algorithm(s) on the training and test scenarios given. • Report the fitness on both the training and test sets averaged over at least 10 runs of your algorithm. In addition, you may wish to: • Conduct a parameter exploration to tune the algorithm you have designed • Investigate different activation functions for the neural network • Investigate the number of nodes that should be in the hidden layer • Redesign the evolutionary algorithm • Investigate the role of different operatorsYou are required to write a scientific report using the template provided (both Word and Latex templates are provided in ACM format). The report should detail any research you undertook while investigating the task and should provide details of your implementation. You should include a scientific analysis of your approach by conducting multiple runs on both training and test sets to show that your implementation can provide consistent results.Students can get formative feedback on their progress during the weekly lab sessions.
Component 1 subtotal: 60
Component 2 subtotal: 40
Module subtotal: 100

Indicative References and Reading List - URL:
Contact your module leader
 

Related information