COMPUTATIONAL INTELLIGENCE FOR DATA ANALYTICS

CSE4CI

2017

Credit points: 15

Subject outline

Quantitive analysis plays an important role in business analytics and knowledge engineering, thus it is very useful to develop computing skills for data regression and classification. This subject covers some fundamentals of computational intelligence techniques, including fuzzy inference systems, neural networks and hybrid neuro-fuzzy systems. The subject is designed with a focus on solving time-series forecasting problems using fuzzy inference systems, where fuzzy inference mechanisms and fuzzy rule extraction from numerical data are addressed. Some advanced learning techniques for training neural networks will also be highlighted. In labs and assignment students will work with business datasets for time-series prediction using a fuzzy system and neural networks with advanced learning algorithms, which help to consolidate the knowledge taught in the lectures and gain a hand-on experience on computational intelligence applications in business.

SchoolSchool Engineering&Mathematical Sciences

Credit points15

Subject Co-ordinatorJustin Wang

Available to Study Abroad StudentsYes

Subject year levelYear Level 4 - UG/Hons/1st Yr PG

Exchange StudentsYes

Subject particulars

Subject rules

Prerequisites CSE2AIF or CSE2DBF OR Admission in one of the following courses: SMIT, SMICT,SMITCN SMCSC, SGIT or SGCS.

Co-requisitesN/A

Incompatible subjects CSE3CI AND Students in the following courses are not permitted to enrol: SBCS, SBIT, SBCSGT, SVCSE, SZCSC, SBITP and SBBIY.

Equivalent subjectsN/A

Special conditionsN/A

Graduate capabilities & intended learning outcomes

01. Explain the technologies and applications of computational intelligence systems (CIS).

Activities:
Lecture 1 is on the introduction of computational intelligence systems and its applications.
Related graduate capabilities and elements:
Critical Thinking (Critical Thinking)
Writing (Writing)
Inquiry/ Research (Inquiry/ Research)

02. Explain the major components and issues in developing computational intelligence systems, such as forecasting and classification systems using fuzzy inference and neural networks.

Activities:
Lectures 2, 3, 4 are on fuzzy logic and fuzzy inference systems. Lectures 5, 6, 7 are on some basics of neural networks, including neuron models, and supervised and unsupervised learning algorithms. Lecture 8 is on Genetic Algorithms for solving model or parameter optimization problems.
Related graduate capabilities and elements:
Critical Thinking (Critical Thinking)
Creative Problem-solving (Creative Problem-solving)
Quantitative Literacy/ Numeracy (Quantitative Literacy/ Numeracy)
Inquiry/ Research (Inquiry/ Research)

03. Evaluate fusion technology, i.e. hybrid intelligent systems and the links between CIS and knowledge engineering.

Activities:
Lectures 9 and 10 are on hybrid intelligent system design. Lecture 11 is on knowledge engineering, where students will find the links between computational intelligence techniques and data mining and knowledge engineering for their further studies.
Related graduate capabilities and elements:
Critical Thinking (Critical Thinking)
Creative Problem-solving (Creative Problem-solving)

04. Implement a fuzzy expert system and a neural network for time-series forecasting with real world data from industry.

Activities:
Lab 1 to Lab 6 are on fuzzy expert system implementation, where students will learn how to use Matlab tools to implement a computational intelligence system for resolving a time-series forecasting problem. Lab 7 and Lab 8 are on neural network modeling and applications.
Related graduate capabilities and elements:
Quantitative Literacy/ Numeracy (Quantitative Literacy/ Numeracy)
Ethical Awareness (Ethical Awareness)
Critical Thinking (Critical Thinking)
Teamwork (Teamwork)
Creative Problem-solving (Creative Problem-solving)
Inquiry/ Research (Inquiry/ Research)
Discipline-specific GCs (Discipline-specific GCs)

Subject options

Select to view your study options…

Start date between: and    Key dates

Melbourne, 2017, Semester 1, Day

Overview

Online enrolmentNo

Maximum enrolment sizeN/A

Enrolment information

Subject Instance Co-ordinatorJustin Wang

Class requirements

Laboratory Class Week: 11 - 22
One 2.0 hours laboratory class per week on weekdays during the day from week 11 to week 22 and delivered via face-to-face.

Lecture Week: 10 - 22
One 2.0 hours lecture per week on weekdays during the day from week 10 to week 22 and delivered via face-to-face.

Assessments

Assessment elementComments% ILO*
Assignment 1 - Fuzzy expert system design (750 word equivalent)20 04
Assignment 2 - Neural networks modeling (Group project) ( 750 word equivalent)This group-based assignment requires students to present a case study. A presentation slide with about 20 pages should be submitted.20 04
Exam (one three-hour examination)60 01, 02, 03