2. Intelligent Student Profiling
using Fuzzy Logic
Submitted as Final Year Project
Towards completion of BS(CS)-Hons Degree
The Project is Approved By National ICT R&D Fund.
Ministry of Information Technology
Aniqa Bano
Reg# 11-BCS-F-HU/HR-12
Sonia khan
Reg# 11-BCS-F-HU/HR-36
Project Supervisor:
Mr. Muhammad Aamir Khan[
Department of Information Technology
3. Contents
• Introduction
• Fuzzy Logic
• Fuzzy set
• Fuzzy Rules
• Membership functions
• Neural network
• Existing System
• Proposed System
• Conducted Survey
• Dataset
• Results of survey
• Anfis
• Fuzzy c-means (FCM) Clustering
• Conclusion
• Future work
• References
4. Introduction
We present an intelligent agent-assisted environment for
– active learning to better support the student-centered,
– self-paced, and
– highly interactive learning approach.
The objective of proposed system is to
– understand the students’ perception,
– our study makes great improvement on personalization of learning
and
– achieves learning effectiveness.
5. Introduction to Basic Terms
Fuzzy Logic
• Fuzzy logic is an approach to computing, based on "degrees of truth" rather than the usual "true
or false" (1 or 0). [1]
Fuzzy set
• A set whose elements have degrees of membership. Fuzzy set allows elements to be partially in a
set. [2]
Fuzzy rules
• Rules that applies on fuzzy sets are known as fuzzy rules. IF – THEN are fuzzy rules and in the
form: IF ‘x’ is A THEN ‘y’ is B, Here A and B are I/P and O/P fuzzy sets respectively. [3]
Membership function
• MF is an association between the values of an element and its degree of membership in a set.
Membership value can range from 0 to 1. [4]
Neural Network
• A computing system made up of a number of simple, highly interconnected processing elements,
which process information by their dynamic state response to external inputs. [5]
6. Existing System
The Traditional system of evaluation are of two types:
1) Awarding numbers or 2) Grades.
Following are some problems associated with evaluation:
• Current evaluation doesn’t tell on what basis students get marks
• Multiple Evaluators
• System is not Transparent
• Cognitive Science – Based on mood
• Checking Type
– Lenient type
– Strict type
– Normal type
• Personal factors: like fatigue, stress, etc.
• Lack of details of criteria
To overcome these problems a fuzzy epistemic logic
has been built to present
– The student's knowledge state,
– Learning skills, and
– The ways they handle their problems.
7. Proposed System
We proposed the system that make use of Fuzzy C Mean (FCM) Clustering and the
Adaptive Neuro Fuzzy Inference System (ANFIS) to draw inferences about the
learning capability of a student in the system.
Accordingly this observation helps in making rules based on a more natural order
of learning concepts by the students.
FCM clustering is applied on the dataset that we got from the field survey that we
have conducted from students of BCS, Department of Information Technology.
8. Conducted Survey
A field survey has been conducted at University of Haripur in Department of
Information Technology from students of BCS.
Thirty students took part in the survey. Survey consisted of two sessions, session-1
based on subjective type questions and session-2 contained objective type
questions.
Each student was given 30 questions, questions were randomly arranged to
approach cognitive and learning state of the students.
The results from the survey indicate that the study makes great improvement on
personalization of learning and achieves learning effectiveness.
9. Dataset
Dataset taken from the conducted
survey
We got dataset from the field survey from 30 number of students and number of
questions asked were also 30.
We come to know that mostly students perceive
less complexity and expects more results.
The students who were free of test anxiety and have
strong learning abilities they choose complicated
questions. But most students choose easy and less
complicated questions.
Std.id Session 1 Session 2 Percentage
1 8 9 56.60
2 7 12 63.30
3 10 13 76.60
4 6 10 53.30
5 7 8 50.00
6 7 7 46.60
7 5 6 36.60
8 7 5 40.00
9 6 10 53.30
10 8 9 56.50
11 5 8 43.30
12 4 12 53.30
13 8 13 70.00
14 9 7 53.30
15 7 11 60.00
16 5 9 46.60
17 4 6 33.30
18 10 14 80.00
19 8 11 63.33
20 10 13 76.60
21 6 11 56.60
22 7 15 73.30
23 5 11 53.30
24 9 14 76.67
25 11 13 80.00
26 13 15 93.33
27 10 11 70.00
28 7 12 63.33
29 9 15 80.00
30 8 10 60.00
11. ANFIS(Adaptive Neuro-Fuzzy Inference System)
We used ANFIS (Adaptive Neuro-Fuzzy Inference System) to create rules of our
dataset that we got from the field survey. We have used Fuzzy Logic Toolbox Version
2.2.13 (R2011a).
Using a given input data set, the toolbox function anfis constructs a fuzzy inference
system (FIS).
Loading Training Data into ANFIS from workspace
14. FCM Clustering
Fuzzy C-means (FCM) is a data clustering technique in which given dataset is grouped into ‘n’
clusters with each data point in the dataset that belongs to every cluster to a certain degree.
The figure shows FCM clustering that applied on the dataset that we got from the field
survey.
GUI for
Fuzzy C-Means Clustering
on field survey dataset
15. Plotting Membership Function
When the FCM clustering is done, we selected one of the clusters by clicking on it and
view membership function surface by the “Plot MF” button.
MF Plot for Cluster 2MF Plot for Cluster 1
16. Conclusion
The Fuzzy C-Means Clustering technique is more flexible and produces efficient
results in evaluation of students’ academic performance, the proposed System is
more efficient model in comparison to existing fuzzy expert system.
It also enhances the decision making by academic planners semester by semester
by improving on the future academic results in the subsequence academic session.
It is also helpful for making future question papers.
Project has been Approved by
17. Future work
The conducted survey is sample for getting dataset to test the proposed
system.
We will work on more number of students with different age level and
besides of students we will conduct the survey with distinguish groups of
people in different educational institutions and implement the rules on
datasets for better results.
In future work we will use the combine technique of fuzzy C-Means and
Artificial Neural Networks called Neuro-Dynamic Fuzzy Expert system to
evaluate teacher and student academic performance and also develop
adaptive learning system and Intelligent Tutoring System for Internet
based education like LMS.
18. References
[1] http://whatis.techtarget.com/definition/fuzzy-logic
[2] Fuzzy sets.html
[3]citeseerx.ist.psu.edu/viewdoc/download?
doi=10.1.1.54.2479&rep=rep1&type=pdf
[4] Fuzzy Logic: Intelligence, Control, and Information, J. Yen and R. Langari,
Prentice Hall
[5] Neural Network Primer: Part I" by Maureen Caudill, AI Expert, Feb. 1989