The emergence of a large number of e-learning platforms and courses does not solve the problem of improving the quality of education. This is primarily due to insufficient implementation or lack of mechanisms for adaptation to the individual parameters of the student. The level of adaptation in modern e-learning systems to the individual characteristics of the student makes the organization of human-computer interaction relevant. As the solution of the problem, a mathematical model of the organization of human-computer interaction was proposed in this work. It is based on the principle of two-level adaptation that determines the choice of the most comfortable module for studying at the first level. The formation of an individual learning path is performed at the second level. The problem of choosing an e-module is solved using a fuzzy logic. The problem of forming a learning path is reduced to the problem of linear programming. The input data are the characteristics of the quality of student activity in the education system. Based on the proposed model the computer technology to support student activities in modular e-learning systems is developed. This technology allows increasing the level of student’s cognitive comfort and optimizing the learning time. The most important benefit of the proposed approach is to increase the average score and increase student satisfaction with learning.

In the field of education, every institution stores a significant amount of data in digital form on the academic performance of students. If this data is correctly analysed to discover any pattern related to student learning, it can assist the institution in achieving a favorable outcome in the future. Because of this, the use of data mining techniques makes it much simpler to unearth previously concealed information or detect patterns in student data. We use a variety of data mining methods, such as Naive Bayes, Random Forest, Decision Tree, Multilayer Perceptron, and Decision Table, to predict the academic performance of individual students. In the real world, a dataset may contain many features, yet the mining process may only place significance on some of those aspects. The correlation attribute evaluator, the information gain attribute evaluator, and the gain ratio attribute evaluator are some of the feature selection methods that are used in data mining to remove features that are not important for the mining process. Other feature selection methods include the gain ratio attribute evaluator and the gain ratio attribute evaluator. In conclusion, each classification algorithm that is designed using some feature selection methods enhances the overall predictive performance of the algorithms, which in turn improves the performance of the algorithms overall.

The ISP is responsible for providing programmable cables or DSL modems. The ISP will send a technician to run the cabling and activate the service to the home or office. There are several obstacles from observations made at PT Telkom Akses Ujung Berung in evaluating the performance of Technicians. Management has difficulty evaluating the performance of the desired assessment technician. The evaluation process still uses an assessment based on subjective perceptions from the team leader, this is due to the absence of appropriate methods to be applied in the process of evaluating the performance of technicians at PT Telkom Akses Ujung Berung. One way to overcome the problem is the existence of a method for making appropriate decisions to assess or evaluate the performance of the Technician. The purpose of this research is to implement a decision support system that is implemented in an application performance evaluation technician with the Android-based Weight Product method that can solve problems by multiplying to connect the attribute rating with the corresponding weight attribute. In this method, there are 5 criteria used and 12 alternative ratings for PT. Telkom Access Ujung Berung. The results of this research are the highest results from the criteria, which are sorted from the highest to the lowest technician scores so that it can facilitate the management in evaluating the performance of technicians at PT Telkom Akses Ujung Berung.

Software testing is the significant part of the software development process to guarantee software quality with testing a program for discovering the software bugs. But, the software testing has a long execution time by using huge number of test suites in the software development process. In order to overcome the issue, a novel technique called Hybridized Buffalo and Truncation Cyclic Gene Optimization-based Densely Connected Deep Neural Network (HBTCGO-DCDNN) introduced to improve the software testing accuracy with minimal time consumption. At first, the numbers of test cases are given to the input layer of the deep neural network layer. In the first hidden layer, the test suite generation process is carried out by applying the improved buffalo optimization technique with different objective functions namely time and cost. The improved buffalo optimization selects optimal test cases and generates the test suites. After the generation, the redundant test cases from the test suite are eliminated in the reduction process in the second hidden layer. The Truncative Cyclic Uniformed Gene Optimization technique is applied for the test suite reduction process based on thefault coverage rate. Finally, the reduced test suites are obtained at the output layer of the deep neural network The experimental evaluation of the HBTCGO-DCDNN and existing methods are discussed using the test suite generation time, test suite reduction rate as well as fault coverage rate. The comparative results of proposed HBTCGO-DCDNN technique provide lesser the generation time by 48% and higher test suit reduction rate by 19% as well as fault coverage rate 18% than the other well-known methods.

Finite Element Method Magnetics (FEMM) is an open source software package for solving electromagnetic problems based on the finite element method. The application can numerically solve linear electrostatic problems and magnetostatic 2D problems, respectively low frequency magnetic, linear harmonic and nonlinear. FEMM is a product much used in science and engineering that, in the last 15 years, has begun to be used more and more in the academic environment. Despite the fact that FEMM can be used to solve complex problems in science and engineering, electrostatic FEMM cannot work directly with discrete electric charge distributions, that is, point electric charge. This work presents a FEMM model for simulating point electric charge that can be used in case of electrostatic problems with discrete charge distributions. The numerical solution for the electrostatic field is compared with the analytical solution. This model can be used in the case of an assembly of point electric charges with axial symmetry.

Emotion recognition is a significant research topic for interactive intelligence system with the wide range of applications in different tasks, like education, social media analysis, and customer service. It is the process of perceiving user's emotional response automatically to the multimedia information by means of implicit explanation. With initiation of speech recognition and the computer vision, research on emotion recognition with speech and facial expression modality has gained more popularity in recent decades. Due to non-linear polarity of signals, emotion recognition results a challenging task. To achieve facial emotion recognition using multimodal signals, an effective Bat Rider Optimization Algorithm (BROA)-based deep learning method is proposed in this research. However, the proposed optimization algorithm named BROA is derived by integrating Bat Algorithm (BA) with Rider Optimization Algorithm (ROA), respectively. Here, the multimodal signals include face image, EEG signals, and physiological signals such that the features extracted from these modalities are employed for the process of emotion recognition. The proposed method achieves better performance against exiting methods by acquiring maximum accuracy of 0.8794, and minimum FAR and minimum FRR of 0.1757 and 0.1806.