Implementation of Oxygen Saturation Sensor Data Acquisition Based on Bluetooth Low Energy Protocol

Sensor data acquisition is used to obtain sensor data from IoT devices that already provide the required sensor data. To acquire sensor data, we can use Bluetooth Low Energy (BLE) protocol. This data acquisition aims to process further data which will later be sent to the server. Bluetooth Low Energy (BLE) has an architecture consisting of sensors, gateways, and data centers, but with this architecture, there are several weaknesses, namely the failure when sending data to the data center due to not being connected to internet network and data redundancy at the time of data delivery is done. The proposed solution to solve this problem is to create a system that can acquire sensor data using the Bluetooth Low Energy (BLE) protocol with use a store and forward mechanism and checking data redundancy. The proposed system will be implemented using sensors from IoT devices, the gateway used is Android devices, and using the Bluetooth Low Energy protocol to acquire data from sensors. Then the data will be sent to the cloud or server. The results of the test give the results of the system being successfully implemented and IoT devices can be connected to the gateway with a maximum distance of 10 meters. Then when the system stores, for every minute there is an increase in data of 4 kb. Then there is no data redundancy in the system.

Early Detection of COVID-19 Patient’s Survavibility Based On The Image Of Lung X-Ray Image Using Deep Neural Networks

SARS-CoV-2 causes an infection called COVID-19, which is caused by a new coronavirus. One of the symptomps that dangerous to the patients is developing pneumonia in their lungs. To detect pneumonia symptoms, one of the newest methods is using CNN (Convolution Neural Networks). The problem is when able to detect pneumonia, the patient's survivability, which knowing this will be helpful to decide the priority for each patient, is still in question. The CNN used in this research to classify the patient’s future condition, but met some major problems that the dataset is very few and unbalance. The image augmentation was used to multiply the dataset, and class weight was applied to prevent miscalculation on minority class. 6 CNN architectures used to find the best model. The result VGG19 architecture has the best overall accuracy, in training, it has 80% accuracy, 89% accuracy invalidation, and 82% f1 score accuracy on classifying the testing dataset means the best model if looking for accuracy on prediction, but this cost a prediction time that longest compared to other CNN architectures. MobileNet is the fastest, but it cost much worse on prediction accuracy, only 55%. The ResNet50 model has balanced prediction accuracy/time, it got 77% f1 accuracy, and also 8.49 seconds of prediction time, 9 seconds less than VGG19.

Effect of Error Level Analysis on The Image Forgery Detection Using Deep Learning

Digital image modification or image forgery is easy to do today. The authenticity verification of an image become important to protect the image integrity so that the image is not being misused. Error Level Analysis (ELA) can be used to detect the modification in image by lowering the quality of image and comparing the error level. The use of deep learning approach is a state-of-the-art in solving cases of image data classification. This study wants to know the effect of adding ELA extraction process in the image forgery detection using deep learning approach. The Convolutional Neural Network (CNN), which is a deep learning method, is used as a method to do the image forgery detection. The impacts of applying different ELA compression levels, such as 10, 50, and 90 percent, were also compared in this study. According to the results, adopting the ELA feature increases validation accuracy by about 2.7% and give the better test accuracy. However, the use of ELA will slow down the processing time by about 5.6%.

Deep Learning for Aspect-Based Sentiment Analysis on Indonesian Hotels Reviews

Tourism is one of the fastest-growing industries. Many travelers book hotels and share their experiences using travel e-commerce sites. To improve the quality of products and services, we can take advantage by analyzing their reviews. We can see the good and the bad thing reviews in every aspect of the hotel. However, research to analyze sentiment in every aspect using Indonesian hotel reviews is still relatively new. In this work, we propose to create an Aspect-based Sentiment Analysis (ABSA) using Indonesian hotel reviews to solve the problem. This research consists of four steps: collecting data, preprocessing, aspect classification, and sentiment classification. Our classification process compares with eight deep learning methods (RNN, LSTM, GRU, BiLSTM, Attention BiLSTM, CNN, CNN-LSTM, and CNN-BiLSTM). In aspect classification, we have six classes of aspects which are harga (price), hotel, kamar (room), lokasi (location), pelayanan (service), and restoran (restaurant). In sentiment analysis, we compared two scenarios to classify sentiments as positive or negative. The first one is to classify sentiment in all aspects, and the second one is to classify sentiment in every aspect. The results showed that LSTM achieved the best model for aspect classification with an accuracy value of 0.926. For sentiment classification, our experiments showed that classify sentiment in every aspect achieved a better result than classify sentiment in all aspects. The result showed that the CNN model gets an average accuracy score of 0.904.

Forensic Analysis of Braking Classification Based on Acceleration, Jerk, and Velocity Data

Nowadays, four-wheeled vehicles are equipped with an event data recorder (EDR) device to record sensors data. With advances in-memory technology, EDR provides evidence for forensic analysis after an accident happens, that uses information technology to facilitate forensic analysis to provide complete and valuable results using digital investigations. Several types of research have been conducted to reconstruct accidents from forensic data and Fuzzy Logic is an alternative method for classifying crash data taken from the accelerometer due to less complexity of implementation. Vehicle braking data is one of the most important evidence for digital investigation, since braking is a complex process determined by many factors, such as the condition of the vehicle, road construction, and the driver’s physiological condition. However, the existing digital investigation still process vehicle speed, deceleration, and varia- tion time of deceleration (known as a jerk) in separated manner to determine braking distance, driver response time, and braking category. The problem identified in this paper is how to use deceleration, velocity, and jerk to categorize the braking evidence forensic analysis. In this paper, forensic analysis is limited to produce forensic evident of braking events based on the collected data. The contribution of this paper is to propose a braking detection model by combining acceleration, speed, and jerk data into a Fuzzy Inference System. As a result, a forensic analysis of braking data can better understand the braking maneuvers, which can be further developed to identify the cause of the accident and provide recommendations on which actions to include in future analyses.

Application of Ant Colony Optimization for the Shortest Path Problem of Waste Collection Process

The search for the shortest path of the waste collection process is an interesting topic that can be applied to various cases, from a very practical basic problem to a complex automation system development. In a dense settlement, the waste collection system can be a challenging process, especially to determine the most optimized path. The obstacles can be circling streets, impassable roads, and dead-end roads. A wrong choice of method can result in wasteful consumption of energy. A possible method to solve the problem is the traveling salesman problem using ant colony search optimization, considering its relatively fast optimization process. Therefore, this paper proposes an application of ant colony and traveling salesperson problem in determining the shortest path of the waste collection process. The case study for the optimization algorithm application is the path UGM Sekip Lecturer Housing is considering. Firstly, the data was collected by measuring the distance between points. Then, the paths were modeled and then compared with the actual route used by waste transport vehicles. The last step is implementing the ant colony optimization and traveling salesman problem by determining the cost function and the parameters. The optimization process was conducted several times, considering the random generator within the algorithm. The simulation results show the probable shortest path with a value of about 752 meters so that the use of fossil fuels in waste transport vehicles can be more efficient. The results show that the algorithm can automatically recommend the minimized path length to collect waste.

Autonomous Visual Servoing for Alternately Working Arm Robots

Robots have infiltrated many aspects of human life up to this point, and with the term Industry 4.0, robots have even become the primary workforce in various factories. This condition necessitates that the robots collaborate without clashing. This paper discusses the application of two arm robot manipulators working alternately in sorting agricultural products. The proposed method employs simple image processing to detect the object and becomes the input to the system to control the robots. The effectiveness of the proposed method is enhanced by the application of a Fuzzy Logic Controller to smoothen robots’ joints motions. The average time required by the robot to finish their task from detecting to returning to standby position is 11.76 s for green tomatoes and 12.86 s for red tomatoes. The experimental results show that the proposed method is effective in controlling two robots to pick and place agricultural products using visual servoing.

Development of Digital Ventilator with Internet of Things for Preparation of the Next Outbreak in Indonesia

The emergency of the healthcare device unit, such as a ventilator, has been experienced during the COVID-19 pandemic in 2020. Therefore, ventilator usage is not hard suggested anymore for COVID-19 patients compared to the outbreak beginning. Despite that, it is still essential to have the ventilator ready, if possible, in each house, for the upcoming respiratory syndrome outbreak. Therefore, in this paper, a digital ventilator development is presented. The digital ventilator is comprised of three main parts, such as respiration mechanism (I), controller Internet of Things (IoT) module (II), and website application (III). The developed digital ventilator has been tested by comparing the measurement of respiratory data between the developed digital ventilator and gas flow analyzer. Results show that the respiratory data, such as Pressure Peak (PPeak), Positive End Respiratory Pressure (PEEP), Inspiratory Expiratory Ratio (IE Ratio), Breath per Minute (BPM), and Tidal Volume can be monitored and controlled both directly and online via website application consistently (standard deviation around 10%) with PPeak absolute error of 1.35 mbar, the PEEP absolute error of 0.16 mbar. Furthermore, the average time response of the digital ventilator to the input command from the website application is 0.23 s. Therefore, it is safe to assume that the doctor can use the website application to control the digital ventilator remotely.

Gold price prediction using Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM)

Gold has an important role in worldwide economics. Gold is not only used in jewelry but also can be a good deal for investment however several factors can affect the fluctuation in gold which can make the risk of investing in gold is bigger for many people. Therefore, is very important to predict the gold price for people who invest in gold in order to help reduce the investment risk. This study will implement a hybrid method from Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM). CNN can extract useful knowledge and learn the internal representation of time-series data, and LSTM networks will identify short-term and long-term dependencies effectively. This research will use daily time frame data and weekly time frame data. This research also tried some experiments to find the best hyperparameters of batch size and epochs in ratio data 60:40 and 80:20. The best result obtained in the daily time of ratio data 60:40 with RMSE 13.67953 and MAE 9,40998, while in ratio data 80:20 has RMSE 15,53199 and MAE 12,78120. In weekly time has obtained the RMSE 38,01949 and MAE 28,32035 for ratio data 60:40 while in ratio data 80:20 the result was RMSE 32,61283 and MAE 22,74638. Those results shows that CNN-LSTM model can predict the trend of daily time frame gold price.

Internet of Things for Water Quality Monitoring using Estimate Dissolved Oxygen and Free Ammonia

In fish farming or aquaponics, one of the problems that are often encountered is water quality. Several parameters that must be monitored are ammonia, temperature, pH, and dissolved oxygen. There are available measuring devices for oxygen and ammonia levels in the market, but the price of the tool is not suitable for small scales. This study uses the Emerson formula and the Benson-Krause formula to determine ammonia and dissolved oxygen value. In this study, the two values were measured using RMSE, MAE, and MAPE against NH3 and Dissolved Oxygen values from Seneye. The output of this research is the level of water quality using Fuzzy logic and implementing the Internet of things to minimize human intervention with objects