COMP606 – Foundations of Information Science
Assignment Part 2 – Fuzzy Logic Application
Due Date: Sunday, May 30th 2021 (midnight)
Late submissions will incur a 10 marks penalty per day.
Weighting: 25% of the final course mark
Submission: You will need to submit the written part of this assignment, together with your Fuzzy Inference Model (.fis file) through Turnitin (a link for this purpose will be set up in Blackboard).
Note: This is an individual assignment. All work submitted must be entirely your own.
COMP606 Assignment Part 2 page 1 of 3
This assignment involves building a Fuzzy Inference System (FIS) capable of solving a real-world problem. You are required to carefully design and implement a model using FIS in MATLAB. You need to assess the performance of the system you have designed and report the results. Data for this assignment will be provided to you.
Task 1
Describe the problem you are trying to solve (400-500 words). [25 marks]
• What is the problem about? What is the output you are trying to predict?
• Choose two attributes of your dataset that are the most useful to solve the problem (Hint: Use the corrplot in MATLAB). Elaborate on the rationale for
choosing them.
• Describe your chosen attributes. Why are they important to generating the output?
Task 2
Using Matlab, design and build a FIS capable of solving your prediction problem.
[40 marks]
• Identify the attributes’ linguistic (fuzzy) values, parameterize the appropriate membership functions, and build the rules of your FIS.
o Use appropriate methods (e.g., boxplot, histogram) to determine the ranges for each membership function. Provide your evidence and explain the outcome.
o Explain the rationale for choosing MFs type and ranges.
o Use corrplot and 3D scatter plot to assist you with creating rules. Limit
the number of rules to maximum 5 rules.
• Use examples (rows), ensuring the selection is representative of your dataset, to refine your rules so that they correctly predict the actual outputs. Save your model as a .fis file (you will submit this file together with your report through Blackboard).
COMP606 Assignment Part 2 page 2 of 3
Task 3
Evaluate your model’s performance on a test set (300-400 words). [30 marks]
• Choose a set from your dataset (different from the ones you already used in task 2) and test your rules on them. Construct a confusion matrix and report the accuracy, precision and recall of your model.
• Compare the performances of your system on task 2 and task 3 and provide an analysis of the results.
• Plot the response surface of your FIS. Describe how the response of your system varies as the inputs vary.
• Explain the rationale for choosing your membership functions.
There will be 5 marks for the presentation of the assignment including spelling
and grammar, layout and formatting, and readability of figures.
Good luck!
COMP606 Assignment Part 2 page 3 of 3