International Journal on Advanced Science, Engineering and Information Technology
Not a member yet
2006 research outputs found
Sort by
Customer Satisfaction Assessment System on Transactions E-commerce Product Purchases Using Sentiment Analysis
Get PDFCurrently, more products appear, and various services that offer similar products make it difficult for buyers to decide to buy before seeing reviews from other users. The growth of different e-commerce platforms exacerbates this. Users spent more time choosing products on each platform with many alternative considerations, such as looking at ratings, prices, and reviews from other buyers. This study conducted the optimization process of selecting e-commerce products so that users do not have to spend a long time reading every review when they want to buy a product. This research is expected to provide a comprehensive assessment of the purchase transaction of a product from the reviews provided. The data is sourced from product reviews on e-commerce in Indonesia, which are then classified into positive, negative, and neutral sentiments. The data is divided into 10 folds of data using stratified k-fold cross-validation, consisting of training and testing data with ratios of 90% and 10% of the total data. Our research proposed a system that implemented our modified Naive Bayes model to calculate a product's Customer Satisfaction (CSAT) score and compare it with the Google Cloud NLP model. In our model, the log prior and log-likelihood formulas are modified in the algorithm, adding the prefix "NOT_" after the negation words in the preprocessing. This doubled our model’s F1 score and increased the accuracy by 32%, from 59% to 91%, when compared to the Naive Bayes algorithm without modification
Sodium Counting System in Mass Catering for Therapeutic Diet Preparation
Get PDFSodium is a well-known substance to enhance food taste, but the intake must be restricted, especially for patients who require a low-sodium diet. The amount of added salt in cooking can be optimized and controlled by calculating the total sodium in the ingredients used. This process is cumbersome for hospital meal catering, in which food is prepared based on the number of daily orders and specific diets. The existing solution uses a spreadsheet for calculating the amount of sodium, which is vulnerable to errors and not user-friendly. This paper presents a systematic system that can monitor and control the amount of sodium during meal preparation for hospital catering. The system consists of two main parts: a desktop application and an automated salt dispenser. The application keeps track of sodium usage based on the final ingredient list and the meal plan, thus allowing the catering officer to check the feasibility of changing the sodium amount needed for cooking without doing any manual calculations. The application is then integrated with a salt dispenser to ensure the salt amount used in the cooking is as intended. The successful implementation of this system supports Malaysia’s 2021–2025 salt reduction strategy to prevent and control Non-Communicable Diseases (NCD). It is also consistent with SDG 3: Good Health and Well-Being, which calls for a global decrease in salt intake of 30%
College Course Recommender System based on Sentiment Analysis
Get PDFCollege plays a vital role in defining a student's future by providing relevant education, skills, and exposure. The choice of college courses heavily influences their career foundation and employment skill sets. However, the expanding number of college courses often leaves students struggling to make the best choice, leading to dropouts due to the lack of interest. Many systems rely on existing student reviews or the popularity of the course itself, which may not always result in relevant recommendations. Hence, some systems use sentiment analysis (SA) to evaluate students' opinions, considering qualitative and sentiment data to understand their interests better. However, current SA performance struggles to extract meaningful words due to dataset availability. Hence, a course recommendation system based on students' interests and competence would be valuable. This paper focuses on evaluating and understanding existing systems to provide students with an effective course recommendation system. It includes first gathering useful data that would improve the use of SA. Next, feature extraction techniques Term Frequency-Inverse Document Frequency (TF-IDF) and N-gram were implemented and compared. SA will be performed to increase the relevance of the student's interests to recommend a course by implementing Fuzzy Logic and K-nearest neighbors. These algorithms will be evaluated by performance metrics such as accuracy to determine the most efficient way to recommend a course. The findings highlight the importance of considering students' subjective preferences and interests for better outcomes regarding student success and graduation rates
In Silico Docking to Explore the Coronavirus-2 ACE2 Inhibitor Potential in Brown Seaweed Padina sp. from Morotai Island, North Maluku, Indonesia
Get PDFEfforts to explore new sources of antivirals for coronavirus-2 from abundant marine natural materials are highly encouraged. The study aimed to explore the potential compounds of brown seaweed Padina sp. from Morotai Island extracted using three solvents, i.e., n-hexane, ethyl acetate, and acetone, as an antiviral against coronavirus-2 through an entry inhibitor mechanism using bioinformatics tools. The target protein was Angiotensin-Converting Enzyme-related carboxypeptidase (ACE2) receptor. Protein structure was downloaded from PDB and prepared using Chimera. The interaction of compounds to ACE2 was predicted using AutoDock4 and AutoDockTools. MLN-4760 was used as a standard compound. Results showed that 15 selected compounds were potential as ACE2 inhibitors, resulting in negative binding energies, low inhibition constant, and varying binding modes. The conformation structure of all compounds was occupied on the ACE-2 active site. Four compounds were highly potential as ACE2 inhibitors with binding energy lower than a standard compound, comprised of Neophytadiene (diterpene); 6,9,12,15-Docosatetraenoic acid, methyl ester (fatty acid); N-Dimethylaminomethyl-tert-butyl-isopropylphosphine (alkaloid) and 8,11-Octadecadienoic acid, methyl ester (fatty acid). Ethyl acetate and acetone are suggested to be used as solvents for the extraction to produce compounds as ACE2 inhibitors, but ethyl acetate was found to be the most effective. Brown seaweed of Padina sp. is recommended to be developed as a pharmaceutical and nutraceutical preparation for COVID-19. Further in vivo and in vitro studies are suggested to confirm this study's results and provide stronger evidence
Measurement Analysis of Non-Invasive Blood Glucose On Sensor Coplanar Waveguide Loaded Square Ring Resonator with Interdigital Coupling Capacitor
Get PDFThis study presents the experimental results of a system with a sensor structure detecting human blood glucose levels. A microwave-based sensor is used for non-invasive blood glucose monitoring. The sensor design uses an asymmetrically loaded CPW structure as a square ring resonator with an interdigital coupling capacitor on the ground side. Simulated with a load of artificial finger tissue made from gelatin, modeled in four layers. The first layer is the skin is the outermost tissue, the next layer is fat, blood and bone. Each layer of tissue has a certain thickness size; skin (0.3mm), fat (0.2mm), blood (1.5mm), and bone (4mm). The measurement simulation is used, HFSS as modeling simulation and VNA as a measurement of the physical representation of the design results with parametric optimization methods. To verify the correlation and the expected sensitivity, media with different dielectrics were mounted on the surface of the sensor resonator with blood glucose levels of 1mg/dl, 72mg/dl, 126mg/dl, 162mg/dl and 216mg/dl. Reflection factor S11 was observed based on dielectric constant blood glucose levels (dB) fluctuations. Analysis of the data on the graph between the independent variables, namely blood glucose concentration and the dependent variable levels of S11 has an “R†correlation value of 0.97. The sensitivity level of the sensor on the S11 reflection factor with HFSS simulation averages 73.36mdB/mgdl-1 and VNA reaches 82.39mdB/mgdl-1. The results are interesting for developing a more optimal glucose sensor system
Business Category Classification via Indistinctive Satellite Image Analysis Using Deep Learning
Get PDFSatellite image analysis has numerous useful applications in various domains. Extracting their visual information has been made easier using remote sensing and deep learning technologies that intelligently interpret clear visual cues. However, satellite information has the potential for more complex tasks, such as recommending business locations and categories based on the implicit patterns and structures of the regions of interest. Nonetheless, this task is significantly more challenging due to the absence of obvious visual cues and the highly similar appearance of each location. This study aims to analyze satellite image similarity between business class categories and investigate the capabilities of state-of-the-art deep learning models for learning non-obvious visual cues. Specifically, a satellite image dataset is constructed using business locations and annotated with the business categories for image structural similarity analysis, followed by business category classification via fine-tuning of deep learning classifiers. The models are then analyzed by visualizing the features learned to determine if they could capture hidden information for such a task. Experiments show that business locations have significantly high SSIM regardless of categories, and deep learning models only recorded a top accuracy of 60%. However, feature visualization using Grad-CAM shows that the models learn biased features and disregard highly informative details such as roads. It is concluded that typical learning models and strategies are insufficient to effectively solve this complex visual problem; thus, further research should be done to formulate solutions for such non-obvious classifications with the potential to support business recommendation applications
Position and Temperature Detector for Autism Spectrum Disorder Children based on Sensor and Using IoT System
Get PDFChildren with Autism Spectrum Disorder (ASD) have characteristics where one cannot control emotions, which can cause tantrums that can impact behavior and body temperature. Based on this, they should be supervised by parents/relatives. To reduce the effects of these circumstances, this study seeks to design a technology system that can measure body temperature and detect the position of ASD children who can later monitor the activities they do. This system applies the ESP32 microcontroller and utilizes the GPS module to read the position of objects detected by the system and the MLX90614 temperature sensor, which can detect the body temperature of ASD children. Then, to facilitate checking, the control system is designed with an IoT system through the Blynk application to make it easier for users to supervise ASD children and can be accessed via smartphones in real time. In this study, detection testing was carried out on 3 ASD child subjects by grouping three conditions: namely, the child exits the location when the child is outside the predetermined location; then the child exits the body temperature when the child's body temperature is abnormal, and the child exits the location and body temperature outside normal. The results obtained show that the detector test results provide notifications to application users in the form of "child out of location," "child out of body temperature," and "child out of location and body temperature outside normal"
SADY: Student Activity Detection Using YOLO-based Deep Learning Approach
Get PDFAutomating human activity recognition is one of computer vision's most appealing and pragmatic research areas. In this article, we have addressed the problem of video-based student activity detection. The student’s activity detection using YOLO (SADY) aims to recognize the normal and abnormal student activities to ensure immediate intervention in case of any risk or necessity. We created our classroom data set of around 220 recordings depicting seven student classroom activities. The YOLOv4 Tiny model was retrained using 5000 labeled keyframes extracted from the train videos. The model was then tested for single or multiple activity detections. We presented the evaluated results for various values of hyperparameters like confidence threshold and Intersection Over Union (IoU) thresholds for the proposed model. The model assigns a unique confidence score and action label to each frame for the test videos by positioning recurrent activity labels. The proposed approach achieved a mean average precision (mAP) of 95% and a frame per second rate (FPS) of 45 for the student activity Class Room (CR) dataset and mAP of 95.18 % for the LIRIS dataset. The experimental findings using the Class Room recorded and LIRIS publicly accessible dataset show that our proposed approach outperforms existing approaches regarding recognition accuracy and speed. The comparable results obtained in this research work imply that the proposed framework could effectively monitor student’s activities in schools, colleges, and universities
The Effect of Annealing Modification on Increasing Glucomannan Content of Porang (Amorphophallus Muelleri Blume) Flour
Get PDFThis study aims to determine the effect of annealing on the glucomannan, protein, and water content of porang (Amorphophallus Muelleri Blume) flour. The material was a porang tuber from a farmer in Lubuk Pakam Regency in the second plant period. The method was completely randomized with temperature treatments of 30 °C, 40 °C, and 50 °C, and time; 3 hours, 4 hours, and 5 hours—parameters observation consisting of water content, protein, glucomannan, and yield. Annealing time significantly affects water content, glucomannan, and yield. It is related to the longer the time; the more water-soluble compounds were lost. It increases the amount of glucomannan and decreases the yield. The temperature significantly affects water content, protein level, glucomannan content, and yield. At the temperature reaches 40 degrees Celsius, the number of glucomannan increases, which is associated with the beginning of the gelatinization process. At the onset of gelatinization, the starch granule structure weakens, allowing it to be readily crushed and liberated from glucomannan. This study indicated that glucomannan content decreased significantly as the temperature increased to 50 °C. It is hypothesized that the gelatinization temperature of porang starch is low; therefore, gelatinization is complete at around 50 °C. However, it needs further research. The protein level decreases by increasing temperature due to protein denaturation. The annealing process at 40 °C for 5 hours gives the high glucomannan content. Glucomannan has a strong relationship with yield. The annealing process promised to be used in glucomannan production from porang tuber
Position Data Estimation System Based on Recognized Field Landmark Using Deep Neural Network for ERSOW Soccer Robot
Get PDFOne of the problems faced by soccer robots is how to find out the position of the robot itself and other robots on the field. A simple way to find out the robot's position is to use the odometry method. However, odometry is weak in accumulating position errors that reduce the accuracy of the moving robot's absolute position estimation and orientation. This paper presents a robot position data estimation system that is to be implemented on the ERSOW wheeled soccer robot. The robot can determine its position based on a unique landmark: an L-shaped line on the soccer robot field. We use a deep neural network method to recognize landmark L-shaped. Vision systems and deep learning inferences run on the Robot Operating System platform. After obtaining the distance of the robot to the L-shaped landmark, the robot's orientation and position relative to the field can be accurately determined based on the omnidirectional camera's perception. The results of the position estimation system in this study can be used to reduce position errors resulting from the odometry method. Based on the landmark L-shape recognition test results conducted on 641 datasets, the validation accuracy value is 0.806. The results of testing the robot position generated by vision obtained the largest error x about 2.32 cm and y about 1.99 cm