Jurnal Teknologi Informasi Universitas Lambung Mangkurat (JTIULM)
Not a member yet
142 research outputs found
Sort by
IMPLEMENTATION OF DOUBLE MOVING AVERAGE METHOD IN PREDICTION OF OMEGA 3 CHICKEN EGG HARVEST RESULT of PAK DARMO WEB-BASED: Peternakan Pak Darmo, Desa Rejosari, Kecamatan Sugio, Kabupaten Lamongan
No full textEggs are one of the livestock products that help people to get adequate nutrition. In Indonesia, chicken eggs are in great demand, besides being easy to obtain and the price is affordable, Omega 3 chicken eggs are one of the chicken eggs that have very many benefits. However, chicken eggs themselves become a rare staple because the yield of chicken eggs is too little. The main purpose of this application was to forecast or predict the yield of chicken eggs accurately so that it became one of the options for egg supply in the future. Making the application used the Double Moving Average method as a forecasting method. The Double Moving Average was used as an analysis of data on the yield of chicken eggs Omega 3 with the right calculation produced an accurate forecast. Implementation of the application is intended for users starting from filling in data, filling in forecasting movements, and additional features such as recaps. The results of the application are explained from the forecasting results which have a high level of accuracy and a small error rate using the Double Moving Average method
COMPARISON OF DETECTION WITH TRANSFER LEARNING ARCHITECTURE RESTNET18, RESTNET50, RESTNET101 ON CORN LEAF DISEASE
Get PDFThe occurrence of diseases that impact the leaves of corn plants presents a substantial obstacle in agriculture, leading to a reduction in the overall yield of crops. This study aims to perform a comparative analysis of transfer learning methodologies by employing three distinct ResNet architectures: ResNet18, ResNet50, and ResNet101. The dataset utilized by the author consists of a compilation of images portraying corn leaves that demonstrate varying levels of disease severity. Transfer learning refers to leveraging a pre-existing ResNet model and retraining the network by employing the corn leaf dataset. The experimental results demonstrate that the ResNet18, ResNet50, and ResNet101 models achieved accuracy rates of 96.68%, 95.73%, and 95.26%, respectively. The ResNet101 model shows superior performance in terms of precision and recall metrics. This research indicates that utilizing a more complex and sophisticated network structure can improve the effectiveness of disease identification in corn plant leaves. The result above is essential in promoting sustainable agricultural methodologies and efficiently managing corn plant diseases
IMPLEMENTATION OF FUZZY MEMBERSHIP FUNCTION TO DETERMINNE AUTOMATIC WATERINNG TIME
Get PDFGlobal warming impacted to agriculture, thereby affecting the world's food supply. UNICEF calls on every country to be able to meet its food needs independently. Indonesia, as an agricultural country, has the potential to meet food needs independently. One effort that can be made is to combine existing agricultural systems with technology through the application of IoT. In this research, it was applied to citrus plants using DHT11 temperature and air humidity sensors and soil moisture sensor YL69. Data processing uses a fuzzy algorithm approach, using two input variables, namely temperature and humidity variables and soil moisture variables. Meanwhile, the resulting output variable is the time for watering the plants. At the fuzzification stage, each variable is grouped into three memberships. For temperature and humidity variables are divided into hot, medium and cold membership. Soil moisture variables are divided into wet, medium and dry membership. And the time variables consists of short, medium and long time membership. In the defuzzification stage, it is calculated by calculating the center of area value for each graphic area resulting from the fuzzification stage. Furthermore, to test the system, the testing carried out resulted in an error value of 5.19 percent. Based on test and trial that be done, the result shows that average volume of water that flows every second when watering plants is 36.5 ml per second
THE THE EFFECTS OF PERCEIVED EASE OF USE, PERCEIVED SECURITY, AND PERCEIVED SATISFACTION ON INITIAL TRUST OF BRI MOBILE BANKING USERS (CUSTOMER STUDY IN SAMARINDA CITY)
No full textThe purpose of this study was to examine the effect of perceived ease of use, perceived security, and trust on customer satisfaction using BRI mobile banking in Samarinda. This study uses a quantitative approach to the type of associative research. The number of samples used was 96 respondents using the purposive sampling method. The criteria for respondents in this study were BRI customers who are domiciled in the city of Samarinda and are over 18 years old and have been using BRI mobile banking for the last 6 months. The analysis technique used is data quality test, classic assumption test, multiple linear regression test, and hypothesis testing which is processed using the SPSS version 22 application. From the test results it is known that perceived security and perceived satisfaction have a positive and significant effect on the initial trust of BRI mobile banking users both partially and simultaneously, while perceived ease of us does not have a partial significant effect on the initial trust of BRI mobile banking users. So it can be concluded that perceived security and perceived satisfaction affect the initial trust level of BRI mobile banking customers
A COMPARATIVE STUDY OF SENTIMENT CLASSIFICATION: TRADITIONAL NLP VS. NEURAL NETWORK APPROACHES
Get PDFThe current research compares traditional natural language processing methods, such as Naive Bayes and Support Vector Machine, to neural network approaches, particularly Multi-Layer Perceptron, to classify positive and negative sentiments regarding company customer service. This research is motivated by the need to understand the effectiveness of these two approaches in analyzing and classifying sentiment in customer reviews, a crucial aspect of enhancing the quality of customer service. The author evaluated accuracy, speed, and adaptability to complex and diverse review content using a dataset containing various business customer reviews. The findings of this study indicate that neural network approaches, particularly Multi-Layer Perceptron, tend to provide superior performance in classifying customer sentiment with greater precision, albeit at a higher computational cost. Traditional methods such as Naive Bayes and Support Vector Machine still apply in situations with limited resources. The results of this research provide valuable guidance for companies in selecting an appropriate approach to analyzing customer sentiment, with the potential to increase understanding of customer views and improve overall customer service. Nave Bayes achieves 68.75% accuracy, Support Vector Machine achieves 87.5% accuracy, and Multi-Layer Perceptron achieves 100% accuracy
IMPLEMENTATION AND EVALUATION OF INTERACTIVE EDUCATIONAL GAME OF WADAI BANJAR AS AN EFFORT TO PRESERVE TRADITIONAL CAKES OF SOUTHERN KALIMANTAN
Get PDFWhen tourists visit an area, they will definitely look for food and cakes typical of the area to taste or as souvenirs. Including when tourists visit South Kalimantan, tourists will also look for special cakes often called Wadai Banjar, there are various types of Wadai Banjar and must be preserved. Among the younger generation now, Wadai Banjar must also be introduced and strived to be preserved. Based on the information, the purpose of this research is to design, developing, and evaluating the Wadai Banjar educational game. The method used is the Game Development Life Cycle (GDLC) which consists of initiation, pre-production, production, testing, beta, and release, followed by an evaluation of the use of the game application. A total of 100 users have been involved in evaluating the use of the application. The result of this study is that the educational game that has been designed and built can function according to the test features provided and can be released to users. The evaluation of use provides results that the Wadai Banjar educational game application can provide fun, comfort, a fitting and attractive user interface, can find out and learn in recognizing the types of Wadai Banjar, useful and as an effort to introduce and preserve traditional cakes typical of South Kalimantan. Then as many as 70 users are more interested if this Wadai Banjar educational game is desktop-based, because interaction is more flexible and a mouse is needed as the main input media in performing actions
INVESTIGATION AND PENETRATION OF DIGITAL ATTACKS ON ZIGBEE-BASED IOT SYSTEMS
Get PDFThe market for Internet of Things (IoT) products and services has grown rapidly. It has been predicted that the deployment of these IoT applications will grow exponentially in the near future. However, the rapid growth of IoT brings new security risks and potentially opens up new types of attacks for systems and networks. This article outlines various techniques to carry out attacks on ZigBee-based IoT systems. We conducted penetration experiments on various possible attacks on Zigbee-based IoT. The purpose of this experiment’s results is for reference in developing an Intrusion Detection System (IDS) specifically for ZigBee-based IoT
MITIGATION VOLTAGE SAG USING DVR WITH HYSTERESIS-ANN CONTROL ON DISTRIBUTION SYSTEM PUJON FEEDER LINE
Get PDFThis paper discusses the mitigation of voltage sag in distribution systems. Voltage sag often occurs due to a single line-to-ground short circuit fault. The fault has a major impact on load damage in the Pujon feeder line. Voltage sag can be mitigated by using Dynamic Voltage Restorer (DVR). DVR is one type of custom device that is effective in overcoming voltage sag. DVR requires control in regulating voltage recovery on the Pujon line. This research proposes DVR control using Hysteresis combined with ANN to mitigate voltage sag on Pujon line. Simulation results show that the Hysteresis-ANN control is able to recover the load voltage in phase R on average by 97% due to voltage sag that occurs on the Pujon feeder line
ANALYSIS OF QRIS USER EXPERIENCE USING THE USER EXPERIENCE QUESTIONNAIRE (UEQ) METHOD
Get PDFDigital wallets are the most preferred payment method by Indonesian digital society compared to other payment methods, namely cash payments and bank transfers. Over time, several digital wallet platforms began to create QR codes according to their respective platforms. With each platform having its own QR Code, this can make it difficult for users to make payments. Therefore, Bank Indonesia, which is tasked with regulating and maintaining the smooth running of the payment system, regulates the QR Code to be standardized in accordance with the International Europe Mastercard Visa standard. A product is said to be successful if the product is able to meet user needs to increase user satisfaction. In order for the product to meet the standards, it is necessary to evaluate the quality of the product. One of the evaluations that must be carried out is an evaluation of the User Experience. The sampling technique used is purposive sampling with the criteria that domiciled in Samarinda and has used QRIS as a payment system. Data analysis was carried out using the UEQ Data Analysis Tool. The results of this study are that user experience (attractiveness, perspicuity, efficiency, dependability, stimulation, novelty) has a positive impression on QRIS user satisfaction in Samarinda City
ANOMALY DETECTION IN ZIGBEE-BASED IOT USING SECURE AND EFFICIENT DATA COLLECTION
Get PDFThis article outlines various techniques for detecting types of attacks that may arise in ZigBee-based IoT system. The researchers introduced a hybrid Intrusion Detection System (IDS), combining rule-based intrusion detection and machine learning-based anomaly detection. Rule-based attack detection techniques are used to provide an accurate detection method for known attacks. However, determining accurate detection rules requires significant human effort that is susceptible to error. If it is done incorrectly, it can result in false alarms. Therefore, to alleviate this potential problem, the system is being upgraded by combining it (hybrid) with machine learning-based anomaly detection. This article expounds the researchers’ IDS implementation covering a wide variety of detection techniques to detect both known attacks and potential new types of attacks in ZigBee-based IoT system. Furthermore, a safe and efficient meth-od for large-scale IDS data collection is introduced to provide a trusted reporting mechanism that can operate on the stringent IoT resource requirements appropriate to today's IoT systems