MACHINE LEARNING FOR EFFICIENT ASSESMENT

AND PREDICTION OF HUMAN PERFORMANCE IN

COLLABORATIVE LEARNING ENVIRONMENT

TEAM GUIDE NAME

SOWMYA.L(16C01A0533) SHAIK MOHAMMED SHAFIULLA NANDINI.K(15C01A0529)

Assistant Professor

SPANDANA.B(16C01A0509) Dept of CSE

SAI PRAKASH.G(16C01A0519)
 CONTENTS

1. Abstract
2. Introduction
3. Existing System and Its Disadvantages
4. Proposed System and Its Disadvantages
5. System Requirements
6. System design
7. Implementation
8. System Testing
9. Output Screens
10. Conclusion
 abstract

 The objective of this work is to develop a machine learning based methodology.

 A Collaborative Learning environment(CLE) is used to find patterns of learning, interaction and

effective assessment of a complex system involving massive data.

 Collaborative learning environment may takes place between larger team members to find the

solutions for real time events and to discuss concepts during situational judgement tasks(SJT).
 INTRODUCTION

 In our project we used a three stage method to explore and study the collaborative group behaviors
 Our system identifies behavioral components at a micro level.
 To gather and analyze data in different forms scalable computing is used.
 Convolutional Neural Networks (CNN) are used for feature extraction.
 Collaborative learning may take place between dyads or larger team members to find solutions for
real time events or problems, and to discuss concepts or interactions during situational judgment
tasks (SJT).
 Modeling a collaborative, networked system that involves multimodal data presents many
challenges.
 EXISTING SYSTEM

 In existing scenario collaborative learning methods have been implemented with active human
involvement in micro group communication which is critical for effective learning in many
organizations.
 Modelling human behaviour in collaborative learning environment is highly a complex task.
 Cipresso introduced computational psychometric-based method for modelling the characteristics
of human behaviour which took lots of time to extract data.
 By using that method failed to achieve accuracy and valid assessment of micro level interactions
in CLE.
DISADVANTAGES:
 One of the biggest challenge in data analytics is to extract information from huge data which
takes lots of time.
 Extraction of accurate data is also one of the deficiency in the current methodology.
 PROPOSED SYSTEM

 The first stage integrates and processes multimodal data obtained in a collaborative learning
environment (CLE).
 The second stage performs feature extraction and cloud computation using computational
psychometrics (CP) and convolutional neural network (CNN).
 The third stage uses the parameters measured in the previous two stages to understand and
model group interactions, competencies, and collaborative behavior at a micro-level.

ADVANTAGES:
 Massive data-intensive, high-performance, scalable computing is transforming our capabilities to
gather and analyze data in different forms.
 SYSTEM REQUIREMENTS

HARDWARE REQUIREMENTS:
 Processor : Core I3.
 Hard Disk : 500 GB.
 RAM : 4 GB.
 Camera : 1 mp
 Audio : Microphone
SOFTWARE REQUIREMENTS:
 Operating system : Windows 10
 Coding Language : Python
 Tool : PyCharm
 SYSTEM DESIGN

Cont…
 DATA FLOW DIAGRAM

 Data Flow Diagram

Cont..
 UML DIAGRAMS

 Use Case for Authentication  Use Case for Detecting face, smile, eyes

detecting face
Collect Training data

detecting smile
user

Authentication
check authentication

detecting eyes

Cont…
 UML DIAGRAMS

 Use Case for Emotion Detection

 Class Diagram

facial expression detection

users

Authentication +Checking Authentication

voice detection +Test data +Detecting face()

user +Detecting smile()
+traing data() +Detecting eye()
+checking user() +Emotion detection()
+Audio Converstion()
description

Cont…
 SEQUENCE DIAGRAM

 Sequence Diagram

Authentication User detecting smile,face,eye emotion detection voice detection

1 : collect Training data()

2 : check authentication()

3 : Detection start()

4 : detecting face,smile,eye()

5 : detecting emotions()

6 : detecting voice()

Cont…
 ACTIVITY DIAGRAM

 Activity Diagram

user Authonitication

false

detecting face, smile, eye

detecting emotions

Voice recognitation
 IMPLEMENTATION

MODULES:
 Authentication
 Detection of Face, Eye and Smile.
 Emotion detection
 Audio Conversion
 SYSTEM TESTING

 Testing is the process of trying to discover every conceivable fault or weakness in a work product.
 It is the process of exercising software with the intent of ensuring that the Software system meets its requirements
and user expectations and does not fail in an unacceptable manner.
TYPES OF TESTING
 Unit testing involves the design of test cases that validate that the internal program logic is functioning properly.
 Integration tests are designed to test integrated software components to determine if they actually run as one
program.
 Functional tests provide systematic demonstrations that functions tested are available as specified by the business
and technical requirements.
 System testing ensures that the entire integrated software system meets requirements.
 White Box Testing is a testing in which the software tester has knowledge of the inner workings, structure and
language of the software, or at least its purpose.
 Black Box Testing is testing the software without any knowledge of the inner workings, structure or language of
the module being tested.
 TEST CASES

S.no Test Case Expected Result Result Remarks (IF Fails)

1 Checking user If user is valid pass If user is not valid

2 Detecting Face, smile eye It shows the indications like face, Pass It doesn’t show the indications
smile, eye like face, smile, eye

3 Emotion detection It detects the emotions like happy sad, pass It doesn’t detect the emotions like
angry. Etc happy sad, angry. Etc

4 Emotion behaviour If detect the emotion it shows the pass It not detects the emotions
behaviour based on your expressions

5 Voice detection It detects the voice and give the report pass It doesn’t detect the voice.
base on detection  
 OUTPUT SCREENS

 Training User Images

Cont…
IMAGE DETECT Cont…
IMAGE STORED Cont…
RUNNING MODEL Cont…
USER AUTHENTICATION Cont…
USER NOT DETECTED Cont…
USER AUTHENTICATED Cont…
EYES DETECTED Cont…
EMOTION DETECTED Cont…
AUDIO RECORDING SESSION Cont…
PLAYING RECORD Cont…
ESTIMATING USER BEHAVIOR
 CONCLUSION

 In this paper, we presented a machine learning (ML)-based system architecture to identify evidence about
teamwork skills from the behavior, group dynamics, and interactions in the CLE.
 We developed a three-stage robust architecture for data intensive computing and efficient assessment of teamwork
CPS skills.
 The knowledge gained in developing this baseline model will represent significant guidance for proceeding phases
and potential studies to follow.
THANK YOU