COML FRAMEWORK

This chapter proposes a method for acquiring the competency level of learners considering the performance indicators resulting from the competency algorithm, as shown in Figure 2.2. The performance markers are

Author

Technology Used

Remarks

Cognitive Theory

Brain-Computer

Interface

CoML

(Alexandr a & Brad, 2017)

C ompetency Assessment using Key Perfonnance

NA

NA

Applied for measur ing IT professional Competency

(Mwakondo, 2016)

CRESSTMode for Learning. Cognitive Tlieoiy For Training, Evaluation Kraiger’s model

NA

Al, Neural Networks, Statistical techniques, SYMs

Pr edictive Mapping of Graduate’s Skills to Industry Roles

(Akangah et al., 2018)

NA

NA

CART Decision Tree. Random Forest,

Predicting Academic Achievement in Fundamentals of Thermodynamics

(Oloruntoba, 2017)

NA

NA

SYM

Student Academic Perfonnance Pr ediction Using S"M

(Colt et al., 2010)

competeucy-based Metrics

NA

NA

Use to Determine the Effectiveness of a Postgr aduate Thoracoscopy Course

(Bhargava, 2015)

Assessment based on Orientation Style, Personality, Attitude. Interest

NA

NA

Stream and subject Selection. Career Planning

Klimesch, Wolfgang. 1999

NA

EEG oscillations in alpha and beta

NA

Memory perfonnance concerning age

Foresiglit, 2003

The interplay of interests between natural and artificial cognitive systems

NA

NA

Cognitive systems

Author

Technology Used

Remarks

Cognitive Theory

Brain-Computer

Interface

CoML

Anupama, H.S. Cauvery, N.K. Lingaraju, G.M.

NA

Emotiv Instrument used for signals

NA

A tool to make disabled people communicate

Sunday, Joseph Heiny. Nwani

NA

В Cl uses in clinical applications

NA

Recording brain wave with an action performed

Proposed CoML framework

Cognizance Factors for self-assessment

YES

Supervised Learning: Decision Tree Algorithms. SYM. Naive Bayes Unsupervised Learning: A'-Means Clustering, Density-Based Cluster

Competency in course Endorsement

estimated considering the cognitive levels of a learner in the IT profession while endorsing a course. Following the cognitive ability of the learner, the course content is designed to identify the competency level of a learner.

Modular view of the CoML framework

FIGURE 2.2 Modular view of the CoML framework.

The proposed framework is based on four-layered architecture, as shown in Figure 2.3. The first layer is data acquisition. This layer acquires cognitive data from the BCI device and the performance evaluation interface. The acquired data are stored in excel file format, that is, CSV. The second layer is data processing. This layer performs preprocessing data tasks. The preprocessing tasks are data transformation, normalization, cleaning, encoding, and feature engineering. Feature engineering skills play a vital role in preparing data for further execution. The third layer is ML experimentation. This layer evaluates and executes supervised and unsupervised ML algorithms. For the implementation of the ML algorithm, dynamic programming skill is the requirement. Python, Java, and R-language support dynamic programming. The output of ML experimentation is used to generate a predictive analytical report of the learner in the last layer of the architecture.

The performance indicator difference is very minimal. The learners had similar features and characteristics irrespective of their performance levels. Based on this analysis, the computed performance indicators can be integrated

Architectural view of the CoML framework

FIGURE 2.3 Architectural view of the CoML framework.

into an endorsement of a course facilitating in measuring the performance of their technical subjects. The result is analyzed by the facilitator as well as the learner in the form of neurofeedback.

This model is to design the framework in classification and computation of the performance indicators. These indicators were a priori defined, adapted by the competency model. The self-cognizance factors and course endorsement competency factors are considered for two courses, namely, AI and DM, as shown in Table 2.2.

TABLE 2.2 Cognizance Factors for AI and DM

Main Factors

Secondary Factors

7j (Al-Cognizance Factors)

/j i—I am interested in enrolling and learning more of AI course

2—I have a stronghold on mathematics 7j3—I have good knowledge of programming language like C, C++, Java

I 4—I have ability to writing algorithm for finding patterns and learning

  • 15— I have strong data analytics skills.
  • 16— I have good knowledge of Discrete Mathematics. 1 —I have excellent logic building skills
  • 1 s—I understand Ann Controller Boards 19—I work on Ann Controller Boards —I have a strong will to leam ML languages

/2 (DM- Cognizance Factors)

I2—I have excellent communication skills I22—I have excellent writing skills /2з—I am a reasonable observer.

/24—I can identify the trends /25—I can develop new approaches.

/2<5—I can quickly adopt the new idea.

/27—I know web applications.

/28—I work on social media.

/29—I have good imagination power.

/30—I can create innovative designs

The course endorsement competency factors have four categories based on learning ability, knowledge applicability, hardware, and software compatibility, as shown in Table 2.3.

TABLE 2.3 Course Endorsement Competency Factors for AI and DM

Course Endorsement Competency—AI

Main Factors

Secondary Factors (P)

CEC

PI. Learning Ability

P2. Hardware Compatibility

P3. Software Compatibility

P4. Applicability

Course Endorsement Competency—DM

Main Factors

Secondary Factors

CEC

PI. Learning Ability

P2. Exemplary Study

P3. Case Studies

P4. Applicability

 
Source
< Prev   CONTENTS   Source   Next >