Universal Face Recognition Using Multiple Deep Learning Agent and Lazy Learning Algorithm

Mainstream face recognition systems have a problem regarding the disparity of recognizing faces from different races and ethnic backgrounds. This problem is caused by the imbalances in the proportion of racial representations found in mainstream datasets. Hence, the research proposes using a multi-agent system to overcome this problem. The system employs several face recognition agents according to the number of races that are necessary to make data encodings for the classification process. The first step in implementing this system is to develop a race classifier. The number of races is arbitrary or determined differently in a caseby-case scenario. The race classifier determines which face recognition agent will try to recognize the face in the query. Each face recognition agent is trained using a different dataset according to their assigned race, so they have different parts in the system. The research utilizes lazy learning algorithms as the final classifier to accommodate a system with the constant data flow of the database. The experiment divides the data into three racial groups, which are black, Asian, and white. The experiment concludes that dividing face recognition tasks based on racial groups into several face recognition models has better performance than a single model with the same dataset with the same imbalances in racial representation. The multiple agent system achieves 85% on the Face Recognition Rate (FRR), while the single pipeline model achieves only 80.83% using the same dataset.

General Cybersecurity Maturity Assessment Model: Best Practice to Achieve Payment Card Industry-Data Security Standard (PCI-DSS) Compliance

The use of technology in the era of the Industrial Revolution 4.0 is essential, marked by the use of technology in the economy and business. This situation makes many companies in the payment sector have to improve their information technology security systems. In Indonesia, Bank Indonesia and the Financial Services Authority (Otoritas Jasa Keuangan - OJK) are agencies that provide operational permits for companies by making Payment Card Industry-Data Security Standard (PCI-DSS) certification as one of the requirements for companies to obtain operating permits. However, not all companies can easily get PCI-DSS certification because many companies still do not meet the PCI-DSS requirements. The research offers a methodology for measuring the level of technology and information maturity using general cybersecurity requirements adopted from the cybersecurity frameworks of CIS, NIST, and Cobit. Then, the research also performs qualitative calculations based on interviews, observations, and data surveys conducted on switching companies that have been able to implement and obtain certification. PCI-DSS to produce practical cybersecurity measures, in general, can be used as a measure of the maturity of technology and information security. The results and discussion provide a model assessment tool on the procedures and requirements needed to obtain PCI-DSS certification. The maturity level value of PT XYZ is 4.0667 at maturity level 4, namely quantitatively managed, approaching level 5 as the highest level at maturity level.

Decision Support System Design Model for Choosing Effective Learning Method in Higher Education Institution

In an educational institution, especially higher education, producing qualified graduates is the highest priority achievement. The quality of the graduates is highly dependent on the applied learning method. However, sometimes universities have difficulty in determining appropriate learning for their students. Determination of this decision requires a tool that can facilitate decisionmakers in analyzing all the considerations in choosing decisions. Another challenge is when the decision is shared among multiple stakeholders with equal contribution. It consequently creates a need for a mechanism that can provide an equal contribution for each stakeholder. The research tries to create a model to design a tool that can determine the decisions that need to be taken. This model is built with the Multi-Criteria Decision Analysis (MCDA) approach. MCDA is selected because it can be implemented in a collaborative ecosystem where multiple stakeholders are involved. Then, a literature study is conducted to determine the attributes and decision-making parameters. In addition, the primary data derived from interviews to a total of 40 respondents consisting of students, lecturers, and staff are taken into consideration to complete the literature study. The literature review and interview output are then translated to some parameters that influence an effective learning system. The result shows that this model can be used as a reference in developing a web-based application as an implementation of MCDA.

Enhanced Financial Business Competitiveness by Leveraging Technology and Innovation

The financial industry has developed rapidly with the emergence of financial technology companies. Moreover, the use of technology in the financial industry has developed rapidly. However, conventional financial companies face tough challenges in doing business. They have to adopt the technology to survive. The adoption of technology will increase the competitiveness of established financial companies against financial technology companies. The research aims to analyze the impact of technology on improving the work process of established financial companies. It examines the business model of established financial companies and financial technology companies. The work process is almost identical for all companies. The research applies a qualitative method. The chosen companies are multi-purpose financing for the purchase of two-wheeled and four-wheeled motorized vehicles. The total number of companies involved in the research is four companies. The result concludes that conventional consumer finance companies need to adopt technology into the work process to compete with financial technology companies. Technologies that can be used are Artificial Intelligence (AI), automation job allocation, chatbot, and face recognition. The work process is reviewed by digitizing, eliminating, and moving. Then, technology adoption will require a large investment. Hence, consumer finance companies can collaborate with technology companies, including financial technology companies.

Evaluation on User Perception Effect based on Interaction Techniques in the Stereoscopic Environment

The interaction of user performance with three-dimensional (3D) objects has become an important issue in the recent development of virtual reality applications. Additionally, the basic conviction of current Virtual Reality (VR) supports the development of the viable interface between humans and machines. The research focuses on the user’s interaction technique by considering two approaches (direct and indirect interaction techniques) for the users while interacting with threedimensional objects. Numerous possible uses can benefit from virtual reality by considering a few fundamental visual and cognitive activities in the Virtual Environment (VE), such as the interpretation of space that users of clear and indirect perception are not well established. The experiment is performed in a stereoscopic environment using a reciprocal tapping task. Participants are expected to use direct pointing as well as indirect cursor techniques to select a stereoscopic spherical target. The results show that, in the sense of a direct interaction technique, user recognition of an object appears to converge in the center of a simulated area. Unfortunately, this convergence is not demonstrated in the indirect cursor situation. The pointing estimation from the users is more accurate when using the indirect interaction approach. The findings provide an understanding of the interaction characteristics done by the users in the stereoscopic environment. Importantly, developers of a virtual environment may use the result when developing effective user interface perception in specific interaction techniques.

Service Level Management of Information Technology and System in ITB STIKOM Bali

ITB STIKOM Bali is a technology-based university so that all business processes, especially related to academics, student affairs, and finance, utilize information technology. Management of the information system and technology is carried out by the Computer and Network Center unit (Pusat Komputer dan Jaringan - PUSKOMJAR). In the management of Information Technology (IT) services, an agreement is needed in the form of Service Level Management (SLM) which contains the level of service, quality of service, and complaints handling, so users’ needs are aligned with the system developer. Meanwhile, in managing the internal system of ITB STIKOM Bali, the agreement between PUSKOMJAR and related units is only in the form of submitting a form for making or changing the system. Therefore, it raises a problem if there is a mismatch in the manufacture or development of the system because there is no contract document that regulates in detail about the system. Hence, designing an SLM of IT and system services at ITB STIKOM Bali to overcome these problems is necessary. The research presents a case study by analyzing and designing the SLM of service system and IT in ITB STIKOM Bali. The applied research method is literature study and observation, analysis of existing SLM, designing new SLM, and conclusions. The result of the research is a document of SLM for internal IT and system services at ITB STIKOM Bali. Based on the analysis and design of SLM, the SLM document can help the PUSKOMJAR in documenting IT service information and making it easier for users to get information related to the provided IT services.

Audio Forensics on Smartphone with Digital Forensics Research Workshop (DFRWS) Method

Audio is one of the digital items that can reveal a happened case. However, audio evidence can also be manipulated and changed to hide information. Forensics audio is a technique to identify the sound’s owner from the audio using pitch, formant, and spectrogram parameters. The conducted research examines the similarity of the original sound with the manipulated voice to determine the owner of the sound. It analyzes the level of similarity or identical sound using spectrogram analysis with the Digital Forensics Research Workshop (DFRWS) Method. The research objects are original and manipulated files. Both files are in mp3 format, which is encoded to WAV format. Then, the live forensics method is used by picking up the data on a smartphone. Several applications are also used. The results show that the research successfully gets digital evidence on a smartphone with the Oxygen Forensic application. It extracts digital evidence in the form of two audio files and two video files. Then, by the hashing process, the four obtained files are proven to be authentic. Around 90% of the data are identical to the original voice recording. Only 10% of the data are not identical.

Leady: A Multisensory Approach in Mobile Application for Dyslexic Children

Dyslexia is considered the most common appearance of a learning disability in the world. Dyslexic students often suffer from emotional issues because of their educational failure experiences. Dyslexia cannot be cured. However, proper and continuous treatment can help students to achieve better results in their studies. The sooner the intervention happens, the better the outcome will be. The research presents the design and development of a mobile application to help dyslexic students to learn how to read. The application is called Learning Dyslexia (Leady). The Analysis, Design, Development, Implementation, and Evaluation (ADDIE) model is used for developing the application. It uses a multisensory approach as the teaching method. Leady aims at young learners so that they can have the intervention from an earlier age. The application design also complies with the appropriate design for dyslexic students, such as in the selection of used fonts and background color. Then, Leady is demonstrated to five dyslexia caretakers (the teachers or parents) from a dyslexia school in Bandung. From the conducted survey, it can be inferred that the method is integrated well since all respondents agree that Leady can help dyslexic students to learn how to read. Most of them agree that the used method in Leady is suitable for dyslexic students.

The Readiness of Learning Object Materials during the COVID-19 Pandemic

The current COVID-19 pandemic requires everyone to limit their activities outside home. This situation also affects the learning process, which requires students and teachers to do their work from home without having to meet face to face following the government policies in schools. Therefore, they must use electronic learning (e-learning) to solve these problems. However, this e-learning model has not been well responded by teachers for learning activities. The research examines the readiness of Learning Object Materials (LOM) during the pandemic at two vocational schools in Magelang. To assess the teachers’ readiness, a mixed-method is used, namely qualitative and quantitative methods. Interview and technical assistance are also done in the research. The results show how the material can be completed using the available time so that the teachers can focus on deepening the material and increasing correspondence. The use of free software using a computer can answer the e-learning model. Moreover, after mentoring, there is a 27% increase in understanding about LOM materials. However, it is also found that teachers are not ready and do not understand how to start online learning and use elearning. There are problems in knowledge, the Internet, the LOM process, facilities and infrastructures, as well as the combination of theory and practice.

Impact of E-Learning on Learning Mandarin in Elementary Schools During Covid-19

In the teaching and learning process, the roles of teachers and students will be balanced if there is good communication between them. However, learning methods from face-to-face to e-learning suddenly prompt the teachers to look for various alternatives to convey the material due to COVID-19 pandemic. The research analyzes the impact of COVID-19 on the learning process, especially Mandarin, in an elementary school. The research uses a descriptive-analytical method with questionnaires. The questionnaires are filled out by 34 students from the fifth grade in Bina Kusuma Elementary School, Jakarta. The research finds the positive impacts of elearning in the school regarding Mandarin learning. First, the subject is more interesting and easier to understand. Second, students can learn the right notes and pronunciation independently. Third, using WhatsApp and Zoom can help students recall previously learned vocabulary. Fourth, students can learn new vocabulary independently. Fifth, e-learning can motivate students to learn and gain experience from new learning methods. Sixth, students can manage their time in learning, so they become more independent. Last, students can study anywhere and anytime without time and space limits. However, the weakness of e-learning is the speed of the internet. It affects the students in obtaining the provided subject by the teachers.