Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

Fuzzy model for the language knowledge evaluation and its computer implementation

Presentation for Third International Students Conference on Informatics "Imagination, Creativity, Design, Development" ICDD 2013
by

Michael Dorokhov

on 19 April 2014

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Fuzzy model for the language knowledge evaluation and its computer implementation

Mykhailo Dorokhov
Fuzzy model for the language knowledge evaluation and its computer implementation
Kharkiv National University of Economics
Modern education and languages
French language in the world
DELF and DALF tests
Problems with test evaluation
The Fuzzy approach
Fuzzy Sets Theory and Fuzzy Logic
DELF A2 Structure
The model in FuzzyTECH
Conclusions
with support of "Agence Universitare de la Francofonie (AUF)"
student:
Kharkiv National University of Economics
La Francophonie
Map of La Francophonie
Modèle flou pour l'évaluation des connaissances
linguistiques et sa mise en œuvre par ordinateur
Etudiant:
Mykhailo Dorokhov
Why learning languages is so important?
NOTHING
Career
Self Development
Travel
&
Discover the World
Learning foreign languages
ici on parle français
french is spoken here
French
Français
French language is used in 60 countries
And has official status in 29 countries
78 000 000 native speakers
270 000 000 speakers
29
60
DELF A1
DELF A2
DELF B1
DELF B2
DALF C1
DALF C2
It can help you check
your language skills
Travelling
Obtaining non-touristics visa
Getting job in France and in Francophonie
Entering French speaking universities
+
Teacher
examiner
+
+
+
Criteria of the test evaluating for teachers
Text description
Fuzzy requirements
Many points of view
Fuzzy approach
Statistics + Fuzzy Logic + Fuzzy Sets Theory
Do a survey for teachers experts
Analyse answers statictically
Make fuzzy rules for
calculation total mark
Fuzzy Sets Theory
Classical sets – either an element belongs to the set or it does not
A = {apples, oranges, cherries, mangoes}
A = {a1,a2,a3 }
A = {2, 4, 6, 8, …}
Examples:
Classical Sets
Formulas:
A = {x | x is an even natural number}
A = {x | x = 2n, n is a natural number}
Membership or characteristic function:
Fuzzy Sets
Fuzzy sets are sets whose elements have degrees of membership
Professor Lotfi A. Zadeh
1965
Fuzzy Logic
The membership function
is usually varying from 0 to 1
Example:
Comfort Temperature Set =
10° | μ(x) = 0
14° | μ(x) = 0.4
26° | μ(x) = 1
32° | μ(x) = 0.8
45° | μ(x) = 0
{ }
fuzzification
fuzzy rules
defuzzification
Fuzzification
Fuzzy Rules
linguistic terms
marks, numbers in any scale
Example
Defuzzification
Task
to develop corresponding model
to use appropriate approach for calculation final grade
to study ways of developing application based on learnt mathematical methods
Problems
Text criteria description
Many points of view are possible
Fuzzy requirements
Multilevel criteria tree
Unability to use standard mathematical tool
DELF A2
it's popular among beginners
it's one of the most popural tests to pass
(information by Institut Français du Kharkiv)
to study the structure of the test and its evaluating guide
to obtain neccesary information from teachers-experts
to develop a fuzzy model based on the corresponding criteria tree
to determine all fuzzy rules based on the survey
Next problems have been solved
selected test:
A2
The structure of DELF A2
DELF
Listening
Reading
Writing
Speaking
Criteria tree
{
25 points
25 points
25 points
25 points
100 total
51 points out of 100
Success:
Speaking
Writing
DELF A2
Structure
Reading - as is
Listening - as is
The model in FuzzyTech
The model view in
FuzzyTech environment
Input and output variables
fuzzification and defuzzification plots
Fuzzy rules for calculation
and making desicions
Output Area
Input Area
Final Mark for the test
The model in progress
Plots of dependence
Manner
Ability to tell and describe
Abitily to give impressions
Extended example of working
Conclusions
use the more complex forms of membership functions;
training and setting the developed fuzzy inference system using neural networks;
use different values of weighting coefficients for different initial, intermediate parameters and the corresponding decision rules;
create a stand-alone program that could be used by the examiner during the test, including usage on mobile or tablet devices.
Intended to explore issues are such as:
linguistic terms
marks, numbers in any scale
Grammar errors
Stylistic errors
Total Mark
Block of Fuzzy rules
Rules
50 points
CoM (center of maximum) method
50%
50%
Fuzzy Logic
CONTENTS
Thanks for your attention!
:)
High
High
Low
Low
Low
Medium
Medium
High
Low
Medium
High
Full transcript