Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control
Not a member yet
424 research outputs found
Sort by
Feature Selection Based on Multi-Filters for Classification of Mammogram Images to Look for Signs of Breast Cancer
Get PDFThe accuracy of classification results on mammogram images has a significant role in breast cancer diagnosis. Therefore, many stages consider finding the model has a high level of accuracy and minimizing the computing load, one of which is the accuracy in using the best feature. This needs to be prioritized considering that mammogram image has many features resulting from the mammogram extraction process. Our research has four stages: feature extraction, feature selection-multi filters, classification, and performance evaluation. Thus, in this research, we propose algorithms that can select the features by utilizing multiple filters simultaneously on the filter model for feature selection of mammogram images based on multi-filters/FSbMF. There are six feature selection algorithms with a filter approach (information gain, rule, relief, correlation, gini index, and chi-square) used in this research. Based on the testing result using 10-fold cross-validation, the features resulting from the FSbMF algorithm have the best performance based on the accuracy, recall, and precision from 72,63%, 70,38%, 75,01% to be 100%. Furthermore, the number of resulting features is the minimum because it results from intersection operation from the feature subsets resulting from the multi-filter
Javanese Character Recognition Based on K-Nearest Neighbor and Linear Binary Pattern Features
Get PDFJavanese script (Hanacaraka) is one of the cultures owned by Indonesia. Javanese script is found in temples, inscriptions, cultural and prehistoric sites, ancient Javanese manuscripts, Gulden series banknotes, street signage, and palace documents. Javanese script has a form with an article, and the use of reading above the script is a factor that affects the character detection process. Punctuation marks, clothing, Swara script, vowels, and consonants are parts of the script that are often found in Javanetest scripts. Preserving Javanese script in the digital era, of course, must use technology that can support the digitization of Javanese script through the script detection process. The concept of script image is the image of Javanese script in ancient manuscripts. The process of character detection using certain techniques can be carried out to extract characters so that they can be read. Detection of Javanese characters can be found by finding a testing image. Here, we had been used 10 words images consisting of 3 to 5 syllables with the vowel aiu. Dataset process by Linear Binary Pattern (LBP) feature extraction, which is used to characterize images and describe image textures locally. LBP has been used in r=4 and preprocessing is also done by thresholding with d=0.3. This process can be done using the K-Nearest Neighbor algorithm. In 10 datasets of Javanese script words, an average accuracy value of 90.5% was obtained. The accuracy value of 100% is the highest and 50% is the lowest
High Accuracy Electric Water Heater using Adaptive Neuro-Fuzzy Inference System (ANFIS)
Get PDFNowadays, water heater is a common household appliance. Water heater can be divided into three types, based on fuel sources: gas, diesel, and electric. Electric water heater is the most common due to its ease of use. The problems that often occur on electric water heater are over-temperature due to user error in setting up the thermostat and inaccurate readings caused by a conventional system control. These problems will cause a surge in power consumption. Over-temperature and conventional control inaccuracies can be overcome using the Artificial Intelligence (AI) control algorithm in the form of an adaptive neuro-fuzzy inference system (ANFIS). The proposed algorithm acts as a control by maintaining the stability of the temperature to obtain more accurate results. An accurate temperature reading can lower power consumption in electric water heater. This study tries to simulate Electric Water Heater temperature control using the ANFIS algorithm until stable readings can be achieved in all temperature settings. Results from disturbance tests in the form of external condition that causes sudden temperature change show that the system can maintain stability with an average error margin of 0.045% and the rate of accuracy of 99.955%
Optimization Improvement Using Pi Controller to Reach CCCV Method in Lead Acid Battery Load
Get PDFSolar energy can produce electrical energy with the help of photovoltaics so as to produce sufficient or even excessive supply to the electrical load. Therefore, it is necessary to store electrical energy (battery) from the excess unused energy. However, in the process of charging the battery, it takes a long time to fully charge the battery capacity and damage often occurs due to excessive voltage used. This can reduce battery life. The characteristics of the battery need to be considered so that the charging process can be carried out in accordance with the required provisions. The existence of the CCCV method can speed up the battery charging process with a constant current of 20% of the nominal current of the lead acid battery. To avoid overvoltage, the constant voltage method can anticipate the occurrence of damage. Utilization CUK Converter as charging can reduce output voltage ripple. The PI control on the CUK Converter produces a constant voltage of 13.8 Volts and a constant current of 1.44 Ampere. The average error generated by this system is 0.14%
Mass Classification of Breast Cancer Using CNN and Faster R-CNN Model Comparison
Get PDFThreat of breast cancer is a frightening type and threatens the female population worldwide. Early detection is preventive solution to determine cancer diagnosis or tumors in the female breast area. Today, machine learning technology in managing medical images has become an innovative trend in the health sector. This technology can accelerate diagnosing disease based on the acquisition of accuracy values. The primary purpose of this research is to innovate by comparing two deep learning models to build a prediction system for early-stage breast cancer. This research utilizes Convolutional Neural Network (CNN) sequential models and Faster Region-based Convolutional Neural Network (R-CNN) models that can determine the classification of normal or abnormal breast image data, which can determine the normal or abnormal classification of breast image. The dataset's source in this study came from the Mammographic Image Analysis Society (MIAS). This dataset consists of 322 mammogram data with 123 abnormal and 199 normal classes. The experimental results of this study show that the accuracy of the CNN and R-CNN models in image classification are 91.26% and 63.89%, respectively. Based on these results, the CNN sequential model has better accuracy than the Faster R-CNN model, because it does not require unique characteristics to detect breast cancer
CICM: A Collaborative Integrity Checking Blockchain Consensus Mechanism for Preserving the Originality of Data the Cloud for Forensic Investigation
Get PDFThe originality of data is very important for achieving correct results from forensic analysis of data for resolving the issue. Data may be analysed to resolve disputes or review issues by finding trends in the dataset that can give clues to the cause of the issue. Specially designed foolproof protection for data integrity is required for forensic purposes. Collaborative Integrity Checking Mechanism (CICM), for securing the chain-of-custody of data in a blockchain is proposed in this paper. Existing consensus mechanisms are fault-tolerant, allowing a threshold for faults. CICM avoids faults by using a transparent 100% agreement process for validating the originality of data in a blockchain. A group of agreement actors check and record the original status of data at its time of arrival. Acceptance is based on general agreement by all the participants in the consensus process. The solution was tested against practical byzantine fault tolerant (PBFT), Zyzzyva, and hybrid byzantine fault tolerant (hBFT) mechanisms for efficacy to yield correct results and operational performance costs. Binomial distribution was used to examine the CICM efficacy. CICM recorded zero probability of failure while the benchmarks recorded up to 8.44%. Throughput and latency were used to test its operational performance costs. The hBFT recorded the best performance among the benchmarks. CICM achieved 30.61% higher throughput and 21.47% lower latency than hBFT. In the robustness against faults tests, CICM performed better than hBFT with 16.5% higher throughput and 14.93% lower latency than the hBFT in the worst-case fault scenario
Accessibility Analysis of Websites of Provincial Governments in Indonesia
Get PDFWeb accessibility means that people with disabilities can use, navigate, and interact with the website. The World Wide Web Consortium (W3C) has provided important guidelines on web accessibility known as the Web Content Accessible Guidelines (WCAG). The Indonesian government encourages the use of new media, namely website, via Presidential Instruction number 3 of 2003 concerning the National Policy and Strategy for e-government development, which mandates every government agency to build a website. In the previous study, the tools used had limitations and were unable to complete the websites evaluation. Therefore, in this study, WCAG 2.0 standard was applied to analyze the websites of provincial governments in Indonesia. Two accessibility evaluation tools were employed, namely TAW and aXe. In addition, for data analysis and interpretation, descriptive statistics and normality tests were applied. The results showed that the most common violation was found in perceivable principle. It was expected that the findings of this study could provide insight and recommendation for web developers working on provincial government website in Indonesia
XGB-Hybrid Fingerprint Classification Model for Virtual Screening of Meningitis Drug Compounds Candidate
Get PDFMeningitis is an infection of the lining of the brain caused by diffuse inflammation, and this condition is caused by viruses or bacteria that cause Meningitis. Prevention for this disease is still in the form of strengthening antibodies with vaccines. There is no significant compound to relieve or treat Meningitis patients. In previous studies, they got seven proteins vital to Meningitis. We continued to investigate the compounds associated with the seven proteins. We chose the in-silico process by utilizing data in an open database. We use several databases for the data collection process. After that, the compound data were extracted for bonding features and chemical elements using molecular fingerprints. We use two fingerprint methods, where both we combine with three types of combinations. The combined results produce three types of datasets with different matrix sizes. We establish the Extreme Gradient Boosting (XGB) method to form the classification model for the three datasets, select the best classification model, and compare it with other classification algorithms. The XGB model has better quality than the classification model of other algorithms. We used this model to predict and quantify compounds that strongly bind to seven vital meningitis proteins. The compound with the highest predictive score (we found more than 0.99) became a drug candidate to inhibit or neutralize Meningitis
A Wearable Device for Enhancing Basketball Shooting Correctness with MPU6050 Sensors and Support Vector Machine Classification
Get PDFOne of the impacts of Covid-19 is the delay of basketball sports competitions, which influences the athlete’s fitness and the athlete’s ability to play, especially for shooting techniques. Existing research in wearable devices for basketball shooting correctness classification exists. However, there is still an opportunity to increase the classification performance. This research proposes designing and building a smartwatch prototype to classify the basketball shooting technique as correct or incorrect with enhanced sensors and classification methods. The system is based on an Internet of things architecture and uses an MPU6050 sensor to take gyroscope data in the form of X, Y, and Z movements and accelerometer data to accelerate hand movements. Then the data is sent to the Internet using NodeMCU microcontrollers. Feature extraction generates 18 new features from 3 axes on each sensor data before classification. Then, the correct or incorrect classification of the shooting technique uses the Support-Vector-Machine (SVM) method. The research compares two SVM kernels, linear and 3rd-degree polynomial kernels. The results of using the max, average, and variance features in the SVM classification with the polynomial kernel produce the highest accuracy of 94.4% compared to the linear kernel. The contribution of this paper is an IoT-based basketball shooting correctness classification system with superior accuracy compared to existing research
Robot Ankle Foot Orthosis with Auto Flexion Mode for Foot Drop Training on Post-Stroke Patient in Indonesia
Get PDFRobot Ankle Foot Orthosis (AFO) has been proven to assist the gait impairment, such as the foot drop. However, development challenge is still remains, such as the trade-off between complexity, functionality and cost. High functionality resulted in high cost, bulky, and complex device. But affordability and simplicity may decrease functionality. Therefore, this research proposed a robot AFO, which has the necessary function of auto dorsi-plantarflexion so it can keep the affordability and simplicity. The robot AFO consists of structure, electronics part and algorithm. The structure is custom made according to the user’s anatomy. A brushless DC (BLDC) motor, Force-Sensing Resistor (FSR) and microcontroller builds the electronic parts. The BLDC motor actuates the flexion, while the FSR detects the gait phase to determine the action. Both are integrated by the microcontroller with the P control algorithm that commands the BLDC motor to generate necessary torque so it rotates in a constant speed. A functionality test has been carried out on the robot AFO, where the robot AFO perform a dorsi-plantarflexion continuously in three conditions, such as no load, 1 Kg load, and foot load. The robot AFO successfully performed a constant velocity rotation in both directions, in all conditions. In the case of 1 Kg load, the maximum angular speed is 0.7 rad/s dorsiflexion and -1.8 rad/s plantarflexion. The torque keeps increasing and decreasing from -0.3 Nm to 4 Nm to keep the angular velocity. The result shows that the robot AFO can perform the necessary function to assist the foot drop training. Functionality test on the gait detection has also been done where it shows that the robot AFO can detect the four gait phases accurately. The robot AFO has been tested and future study should test the robot on a real post-stroke patient to see the effect of the gait control in reality