Search

Due to the COVID-19 outbreak that hits everyone globally, every person is affected, including students. Starting with this, no one can refuse the importance of a smart online learning system being embedded in the education system anymore. Still, the emotional engagement between teachers and students is in doubt. Teachers tend to teach without knowing whether the students understand it. It is due to the limitation of having an online class instead of a physical one, but can be overcome by embedding emotion detection using facial expressions during the course. However, the next challenge is to overcome the system's limitations. Even though people have a well-equipped tool for themselves in this situation, it is hard to determine the exact emotion of the students during the learning process. The facial image captured in the real world usually has multiple issues, including (1) an obstacle in front of the face, (2) a different distance between the head and camera, and (3) low-resolution images due to the low bandwidth, causing the low accuracy of emotion classification. Thus, this paper took the initiative to compare two different datasets tested on four classifiers: CNN, DCNN, Transfer Learning, and Multiple Pipeline. Four different algorithm sets were used to test two datasets: FER2013 and our dataset. The result shows that the data captured in a real-world situation using an independent device setting contains some issues with the slightly low accuracy of emotion classification and leads to false classification compared to FER2013 data. Thus, this can be improved by having a powerful emotion detection system that can capture all the issues in a real-world situation [ABSTRACT FROM AUTHOR]

Software testing phase is one of the most significant phases within the System Development Life Cycle (SDLC) since software bugs can be costly and traumatic. Originally, the implementation of software testing was done manually where the test cases are executed manually without any support from scripts or tools and then multiples of tools were introduced to help the implementation where automated testing taking place throughout the process where test cases are executed using the scripts, tools and software to provide the level of product's quality. Now with the widely known Artificial Intelligence (AI) technologies gaining place in industry, it is transforming the software testing in ways that could not have been dreamt of a decade ago including reducing the need for test maintenance, simplifying test creation, and driving new ways to assess the results. Machine learning is one of the branches in AI that happen to be a new dominated technology in industry. The use of machine learning in software testing's process is rarely implemented in the market, machine learning focus on the use of data and algorithms to imitate the way that humans learn, gradually improve its accuracy and it is an important component of the growing field of data science. In machine learning there are multiple algorithms to be use as an approach that can help to improve the way to manage and understand the results produced from the testing. The use of machine learning in this area theoretically able to provide a faster and easier test creation, make the test analysis much simpler and reducing the needs of test maintenance. In this study will be focusing on using clustering which is one of algorithm in machine learning as an approach to increase the mutation score of mutation testing which is one of the software testing methods on fault-based testing. [ABSTRACT FROM AUTHOR]

Although researchers have shown huge interest in searching for better Android testing approaches, usable and effective techniques that can show a reasonable number of effective tests are still insufficient, especially for Android applications. Due to the challenges faced by Android applications, this research has proposed a set of mutation operators that can discover more faults and simulate more categories during mutation testing processes. Based on a literature review of existing research, analysis on mutation testing operators and the evaluation standard metric that has been implemented by MDroid+ were parts of the main activities during the initial study. The proposed method of this study has discovered an additional category in mutant generation with a higher effectiveness. The implemented Android mutation operator for code annotation in mutation testing was able to increase the coverage of mutation testing in Android applications. Furthermore, it can reduce number of non-compilable mutants during the testing phase. [ABSTRACT FROM AUTHOR]

Constructive alignment is a student-centered teaching principle to support students in developing meaningful and effective learning in achieving the intended learning outcomes. The 21st Century learning skills have been introduced to prepare learners the abilities required to adapt better in modern working environment. This idea is rapidly implemented as part of the intended learning outcomes in educational curriculum globally. Artificial Intelligence (AI) is currently one of the highly demand course to realize the Fourth Industrial Revolution (4IR), that requires re-aligning the syllabus with this 21st century learning skills specifically in critical thinking and creativity skills. Nevertheless, it was observed that learners' performance in previous AI course are not up to par with the requirement, reflecting just average problem solving and creativity skills. Thus, this study proposed to implement Design Thinking as a systematic problem-solving approach to strengthen learner's problem-solving skills in AI course. The course learning outcomes were re-aligned and mapped with the 21st century learning skills and include the Design Thinking systematic phases for improvement. The changes resulted a more positive experiences and reflections from learners towards Design Thinking approach in their AI project execution. [ABSTRACT FROM AUTHOR]

Computational thinking (CT) is useful in daily problem-solving process while educational robotic (ER) is well known as a pedagogical tool in attracting students to learn more in problem solving activities. Both subjects are widely used for formal learning and informal learning regardless of the age and gender of the learners. Both also have big scope. However, there is lack of studies in integrating CT into ER. Thus, nurturing CT through ER remain a challenge. This study focuses on designing a conceptual model of the integration between CT and ER. Validity and reliability are crucial for research trustworthiness. Thus, grounded theory analysis (GTA) is used to analyze the CT concept and ER development from various sources such as the literature, book, discussion and web of contents to make the model more effective, fit, relevance, workability and modifiability. The finding represents a conceptual model that detail out the relationship between CT and ER besides giving benefits to community who are planning, designing, or revising new framework or platform in nurturing CT through ER. [ABSTRACT FROM AUTHOR]