The project endeavored to design and develop a Web-based Performance Evaluation System with integration of rule-based algorithm that will analyze the performance of each reviewee based on the scores from their examination. The algorithm automatically provides suggested reading materials based on the result of the assessment. With this, the system will help the reviewees assess their own performance. The system was developed using Agile Methodology Model of Software Development. The software produced was tested using Alpha and Beta software tastings. The FURPS model was used to assess the software quality as perceived by the selected group of respondents. The system was found to be functional, meeting the designed functional requirements specifications.

The evaluation process to teachers is a method to ensure that instruction quality has been delivered to various stakeholders. Assessing the performance of faculty members is now an integral part of the educational system as it aims to assess faculty performance using sets of criteria and provide necessary academic intervention program anchored in the development of comprehensive faculty development program. With this, the researchers developed faculty evaluation system integrating decision support mechanism that can provide automatic report generation in terms of evaluation. In addition, a rule-based engine has been integrated as a mean to provide decision support mechanism in terms of automatic development plan. The evolutionary prototyping model was used in the development of the system. Based on expert evaluation result, using the ISO selected criteria, the system recorded a grand mean 4.31 which has an interpretation of “Very Satisfactory” This means that the system is ready for deployment.

The number of people being amputated is annually increasing, and it will continue to do so over the years. This, along with other reasons, makes the production of prosthesis fall short from its demands. One of the purposes of this project is to provide an alternative solution to this problem using 3D printing technology, which allows the prosthetic limbs to be produced faster. Another purpose of the project is to incorporate art and design into 3D printed prosthesis, which gives the amputees a chance to personalize their prosthetic limb/s. With this said, the study aims to develop a website designed for amputees, where they are able to personalize their 3D printed prosthetic limb/s.

This project aims to develop an augmented reality mobile application about endangered marine species. The program has made use of augmented reality technology to scan target photos on the card and show three-dimensional reconstructions of six endangered marine species. When a target image is detected, the program will show the 3D model of the sea species to the user. The application was built using Vuforia and Unity, the 3D models were modeled with Substance Painter and ZBrush, and the card layout was designed with Adobe Illustrator. The created system is one of the country's first augmented reality (AR) apps regarding endangered marine species, with the purpose of spreading information and increasing awareness about the situation of endangered marine species via AR. To demonstrate that the program is productive and user-friendly, the researchers surveyed ten IT experts. The survey findings demonstrate that the approach is effective and useable for sharing information and increasing awareness about the state of endangered marine species. Future researchers may enhance the system by incorporating some of the features and refining the mobile application so that people can still use it without the visual book.

It is always difficult to travel alone in a wheelchair without prior knowledge of the accessibility of the planned route. The majority of people prefer the shorter route. On the other hand, those with ambulatory limitations may prefer a longer route with proper ramps and drop curbs. This study aims to design obstacle management so that a registered user can report the accessibility of a ramp. The research includes an algorithm for generating barrier-free routes on the derived graph paths. When a wheelchair user encounters an obstacle while navigating the suggested route, the algorithm redirects them to their destination. A simulation test was performed, and the entire approach was evaluated using the survey method. The results showed that the proposed routing algorithm could find the shortest paths and reroute users to an unobstructed path. Respondents were highly pleased with the proposed navigation system’s performance and thought it was accessible, usable, and reliable. As a result, the study may provide a novel approach to designing a geographic information system for use in a wheelchair navigation system.

Early disease identification and detection have been an interest of experts to enhance productivity and performance in agriculture. This study aims to use deep learning algorithms to classify sugarcane diseases using leaf images. Deep learning algorithms are implemented to create models that can classify sugarcane diseases using 16,800 images of training data, 4,800 images for validation tasks, and 2400 images for testing. Results show that the InceptionV4 algorithm outperforms other models in classifying sugarcane leaf diseases at 99.61 accuracy. Different models such as VGG16, ResnetV2-152, and AlexNet achieve high accuracies of 98.88%, 99.23%, and 99.24%, respectively. Hence, this study provides evidence that deep learning models can perform better in classification problems. This study suggests some improvements to further its contribution.

Artificial intelligence-based medical image analysis promises an efficient and reliable diagnosis in today's healthcare. Traditional approaches for cataract screening by medical practitioners often results in subjectivity due to their varying levels of knowledge and expertise. Using transfer learning, ensembles of pre-trained convolutional neural networks, and stacked long short-term memory networks, we developed a non-invasive and streamlined pipeline for automatic cataract severity classification. Empirical results show that our proposed combined models of AlexNet, InceptionV3, Xception, and InceptionResNetV2 using a weighted average algorithm produces 99.20% (normal vs. cataract) and 97.76% (normal to severe) accuracies compared to standalone models. Furthermore, the ensemble model reduces classification error rates by an average of 2.17%. This study has the potential to help doctors to specify the magnitude of cataract stages with highly acceptable precision.

The incidence of falls is more commonly experienced by the ageing population globally, due to their increasing frailty. Fall detection systems are being developed combined with machine learning approaches that include those wearable devices, ambience-based systems, and vision-based systems. Wearable systems are omnipresent but intrusive while ambience-based systems are temperature-dependent that require a dedicated GPU. Challenges on vision-based fall detection systems include large computational cost, use of specialized camera, and system integration into smartphones. This research aims to propose a vision-based fall detection system ported on a smartphone application, that utilizes deep learning trained and tested from multiple RGB camera setups. In an event of a fall, the system is designed to provide an on-premises auditory alarm, an IoT notification, and a real-time video feed via the smartphone application. Using the pretrained MobileNetv2 CNN-based MediaPipe Pose and Random Forest Classifier for fall detection, experimental results show high-performance evaluation metrics.

Switches have been developed to connect a source and the load by opening or closing a circuit. This study developed an inexpensive but highly responsive capacitive switch prototype for lighting and motor control systems as a fine-tuned alternative. The control switch consists of a capacitive touch sensor mounted on a standard switch box with an internal control circuit board. Moreover, a capacitive touch switch is applied in one-way and three-way light switches and forward-reversed motor control. We devised a practical application of a capacitive touch switch to control a lamp from two locations and use it to function as a three-way button switch. The prototype was installed in a wooden case consisting of the following components: magnetic contactors, control relays, an Arduino board, a lamp, a capacitive touch control switch for dimmer light, and motor control. A series of test and analyses were performed on the prototype circuit to ensure the correct and targeted functions, meet the desired specifications, and operability and reliability. Based on the test performance, the level of acceptability for its safety (maximum voltage = 12VDC, temperature = 85 °C) and responsiveness (0.05) was highly acceptable with 25% cheaper costs than existing commercially available similar models.

The accumulation of waste tires leads to environmental degradation caused by uncontrolled dumping in landfills, which are prone to fire and emit harmful gases like carcinogens. Reusing this as reinforcement to self-compacting concrete (SCC) is an alternative way to address the issue. For over a decade, SCC emerged in the construction industry due to its enhanced mechanical properties and capacity to self-consolidate on its own. However, there is still limited literature describing the behavior of SCC with tire waste steel fiber (TWSF). This study provides an overview of the extraction, quantification, geometric characterization, surface characterization, and application of TWSF to self-compacting concrete to determine workability and the compressive strength of SCC with TWSF. A total of five mixes were prepared, including the control noted as SCC without fiber and SCC with TWSF, with fiber content ranging from 0.7 %, 1 %, 2 %, and 3 %. The fresh properties were evaluated using the European Federation for Specialist Construction Chemicals and Concrete (EFNARC) standards such as slump flow test, T500, L-Box, and wet sieving or GTM Screen Stability Test. In addition, the compressive strength was determined after 28 days. The investigation reveals that these fibers can be retrieved in three ways: manually cutting the tire's edge, using a specialized machine to pluck the fibers, or incinerating them. It was projected that 4.85 - 7.16 x 105 t of TWSF might be generated annually. The result of the inclusion of TWSF in SCC does not significantly affect the workability. However, there is a reduction in the passing ability of about 11.713 % and 186.75 % for GTM screen stability, but all mixes are still within the acceptable ranges specified on the EFNARC standard. In contrast, the results reveal that adding 3 % TWSF to SCC enhances compressive by 31 %, which might be due to the fiber's uneven surface, increasing the bond between the fiber and concrete. As a result, the TWSF can be utilized to strengthen the SCC and fully applied in the construction industry. Additionally, it is advantageous to combine TWSF with SCC to extend its life resulting in lower carbon emissions produced during the production processes.