ejournal.nusamandiri.ac.id (STMIK Nusa Mandiri)
Not a member yet
1504 research outputs found
Sort by
PENERAPAN METODE APPLICATION DEVELOPMENT DALAM PERANCANGAN PROGRAM PENJUALAN TANAH KAVLING BERBASIS WEBSITE
Get PDFCurrently, property developer agencies are progressing rapidly, the supporting elements of home ownership are also facilitated by the presence of home ownership loans. In the process of selling land property, it still uses conventional methods so that customers must be present in person to find and order land plots. The sales process is still manual, causing errors in reporting and making it difficult for employees to find documents. The purpose of this research is to design a website-based land plot sales program that can facilitate the process of providing online land availability information, the land plot sales process which is equipped with a payment scheme calculator feature and a track booking feature. This research uses the Rapid application Development (RAD) model with the aim that the application can be efficient in processing time, and uses the black-box testing method in its testing. The result of this research is a land plot sales website with a payment scheme calculator feature and a track booking feature, so that the sales process becomes efficient, reporting data becomes accurate and data search becomes easier
ANALYSIS OF WHISPER AUTOMATIC SPEECH RECOGNITION PERFORMANCE ON LOW RESOURCE LANGUAGE
Get PDFImplementing Automatic Speech Recognition Technology in daily life could give convenience to its users. However, speeches that can be recognized accurately by the ASR model right now are in languages considered high resources, like English. In previous research, a few regional languages like Javanese, Sundanese, Balinese and Btaknese are used in automatic speech recognition. This research aim is to improve speech recognition using the ASR model on low-resource language. The dataset used in this research is the Javanese dataset specifically because there is a high-quality Javanese speech dataset provided by previous research. The method used is fine-tuning the Whisper model which has been trained on 680,000 hours of multilingual voice data using a Javanese speech dataset. To reduce computation requirements, parameter efficient fine-tuning (PEFT) implemented in the fine-tuning process. The trainable parameter is reduced to <1% because the implementation of PEFT reduces the computation required by the model for fine-tuning. The best WER evaluation result is 13.77%, achieved by the fine-tuned Whisper large-v2 model compared to the base model of Whisper large-v2, which achieves 89.40% in WER evaluation. Performance improvement in WER evaluation showed that fine-tuning effectively improves the performance of the Whisper automatic speech recognition model on recognizing speeches in low-resource languages like the Javanese language compared to the Original Whisper model performance with minimal computational cost needed for fine-tuning large model
IMPLEMENTATION OF ARMA MODEL FOR BENGAWAN SOLO RIVER WATER LEVEL AT JURUG MONITORING POST
Get PDFThe amount of annual rainfall in the Bengawan Solo watershed causes high water flow (water discharge) in several rivers. In addition, high flow rates significantly increased the water surface level at some observation sites. The Bengawan Solo River burst its banks in November 2016, causing flooding in several areas in eastern Solo. At that time, the river stage at the Jurug monitoring post passed ten. Therefore, a flood early warning system would be useful for predicting water levels in this context. Every day, one post on the Bengawan Solo River measures the water level. The time series data used in this study is the water level. Autoregressive Moving Average (ARMA) is a predictive method for measuring time set data. The assumption of homoscedasticity or constant error variance is used in this model. However, the study will use the ARMA model if the variance changes randomly. The study used 60 pieces of data from January to February 2018. This study can directly use ARMA because the data results are stationary based on ADF value 0.0036. After the first pause, the ACF and PACF are disconnected based on the correlogram pattern. This shows that the water level of the Bengawan Solo River in that period can appear on the Autoregressive Moving Average with orders p = 1 and q = 1 ARMA(1,1). Thus, the total average residue is 0.0668384, so the short error is 6.68384%
DEVELOPMENT OF CINEVERSE FILM WEBSITE UTILIZING THEMOVIEDB'S API FOR DYNAMIC CONTENT MANAGEMENT
Get PDFThe development of websites in this digital era is crucial to creating captivating and relevant online experiences. The combination of server-side programming and client-side technologies along with MySQL database management forms the foundation for a dynamic user interface emphasizes the significance of integrating various technologies to achieve this goal. This project involves the use of PHP, HTML, CSS, JavaScript, and MySQL, with the integration of The Movie Database (TMDB) API, showcasing the intricate fusion of creativity, technical prowess, and data integration. The resulting website offers a comprehensive list of films with detailed information and posters, enhancing the user experience and making it an essential read for those interested in crafting immersive online experiences. The abstract of this research aims to explore the process of website development using diverse technologies and data integration and to analyze its impact on user experience. By examining aspects such as security, performance, and routine maintenance, this study aims to provide in-depth insights into producing captivating and relevant online experiences in the context of modern web development
DEVELOPMENT OF FIELD WORK PRACTICE MANAGEMENT IN INFORMATICS INTEGRATED SERVICE SYSTEM
Get PDFField work practice is crucial for aligning college education with workplace skills, but manual management poses challenges such as time-consuming processes and limited information access, particularly within the Informatics Program at Mulawarman University. This research explores integrating field work practice with mobile technology through Rapid Application Development, resulting in a comprehensive field work practice management mobile app. Data collected from direct observation and student questionnaires inform the app's development. The app covers various field work practice aspects, facilitating efficient student engagement, and Black Box testing confirms its functionality without errors. This research significantly contributes to field work practice management, offering stakeholders an efficient, transparent, and well-integrated solution for field work practice implementation in academic settings, ultimately aiming to enhance convenience and accelerate the field work practice process for all involved parties
THE DEVELOPMENT OF AUTOMATIC CIGARETTE SMOKE DETECTION SYSTEM USING TA12-100 AND MQ-135 SENSORS
Get PDFIndonesia is ranked 3rd with the highest number of active smokers in the world after India and China. Cigarettes contain more than 4000 chemical compounds that are harmful to active smokers and passive smokers. The large number of active smokers in Indonesia also has an impact on passive smokers where passive smokers also inhale cigarette smoke. The purpose of this research is to build an automatic cigarette smoke detector to support air hygiene control, protect passive smokers, and save electricity used. The stages in designing this research include analysis of product specifications, assembly, programming, and trials. This study was designed using Arduino UNO as the microcontroller, the MQ-135 sensor as a tool to detect cigarette smoke, the TA12-100 sensor to calculate electrical power consumption, the relay as an automatic switch, and the exhaust fan as a smoke neutralizer. The results of this study are that the MQ-135 sensor can detect cigarette smoke properly with the Relay as an automatic switch that functions to turn on the exhaust fan automatically. Based on the test results on the TA12-100 sensor, the consumption of electrical energy with an automatic mechanism is more efficient than the consumption of electrical energy in a manual way, with a difference of 0.0000385 kWH and Rp. 0.05205. Future research could focus on developing a more complex system for overall air cleanliness control, rather than just focusing on cigarette smoke detection
MACHINE LEARNING FOR PREDICTING SPREAD OF COVID-19 IN INDONESIA
Get PDFIn previous research, we carried out an analysis using the FBProphet model to predict the COVID-19 outbreak in Indonesia. The application of the FBProphet model to time series data is considered quite good because it produces a MAPE of 22.60% with a linear distribution. Additionally, based on the pattern in the previous dataset and the total number of active cases currently stands at 2,606, in this research we tried to use the Linear Regression (LR) model as a comparison with the FBProphet model by using additional data from the same data source, KAWALCOVID19 website. Data collection started from March 2, 2020 to December 19, 2021. The aim of this research is the same as previous research, namely predicting the spread of COVID-19. The analysis process is carried out by preprocessing the data by validating missing data and validating the format of the data variables. Then carry out descriptive analysis and data visualization so that it can be seen that in this 657 data there is a fluctuates data that non-periodically from July to August 2021. Next, model analysis is carried out using FBProphet and LR and validating the results of each model. The research results are in the form of evaluation metrics where the LR model gets better RMSE, MAE and MAPE values compared to FBProphet, namely 292.91; 178, 81 and 12.79%
PERFORMANCE OF THE YOLOV5 ALGORITHM TO DETECT HUMANS IN THE REAR EXCAVATOR AREA
Get PDFWork involving excavators carries a high risk of accidents that can result in fatalities, making Occupational Safety and Health (OSH) critically important. Most excavator accidents are caused by blind spots at the rear, where the operator's limited field of view increases the risk of hitting nearby objects or workers. Despite safety features such as reverse alarms and rear cameras, these technologies only display real-time video without automatically detecting workers, thus still posing a significant risk. This study aims to develop a human detection system for the rear area of excavators using the YOLOv5 algorithm based on image processing. The system's main features include real-time human detection, distance estimation, and audible warnings if a human is detected within a high-risk distance. The system was tested using three video recordings depicting human objects behind the excavator in different scenarios. Despite the limited number of video samples, the human objects provided sufficient complexity to evaluate the system's effectiveness. The test results showed an average accuracy of 80.5% and an F1-score of 87.78%. These findings indicate that the YOLOv5-based detection system performs well in various video conditions and shows potential effectiveness in real operational situations. Consequently, this system is expected to reduce the risk of work accidents with excavators caused by rear blind spots and improve on-site worker safety. This research contributes to the field of occupational safety by integrating image processing algorithms into the development of heavy equipment safety technology, thereby enhancing worker protection
IMPLEMENTASI TEKNIK SMOTE UNTUK MENGATASI IMBALANCE CLASS DALAM KLASIFIKASI SENTIMEN MENGENAI CHATGPT
Get PDFChatGPT is a chatbot or computer program in the form of a virtual robot that can simulate human-like conversations. ChatGPT is widely used in various fields in academia. The impact of the use of ChatGPT on academia and public perception of this technology is significant. Sentiment analysis can be used to determine the polarity of a text or opinion that is positive or negative. In this research, social media is used as a data source to collect public opinion regarding ChatGPT instantly. The methods used in this reserach are the KNN algorithm and Naive Bayes algorithm. The aim of this research is to find the best algorithm model for sentiment classification in terms of public opinion for ChatGPT which contains English text. Before testing the algorithm model, a text processing stage was carried out which included the processes of case folding, tokenizing, stopword removal, and stemming. Word weighting using TF-IDF was carried out before the data was ready to be processed. Splitting data used in this research includes 80% of the dataset as training data and 20% of the dataset as testing data. The application of the SMOTE technique to the KNN and Naive Bayes algorithms to overcome the imbalance class of the public opinion dataset regarding ChatGPT. The research results show that combining SMOTE and Naive Bayes algorithm gives the best results with an accuracy value of 85.00%, a precision value of 87.64%, a recall value of 84.78% and an f1-score of 86.18%
PENGEMBANGAN DAN IMPLEMENTASI LITERASI DIGITAL UNTUK MENANGKAL HOAKS KEBENCANAAN DI CIANJUR
Get PDFThe flood of information following the earthquake disaster in Cianjur has had an impact on the spread of hoaxes which is difficult to stop. The impact not only causes anxiety and fear, it even triggers division and conflict. To educate the public in preventing hoaxes, the Volunteer Team from Putra Indonesia University and Jayabaya University for three months after the earthquake carried out activities to Strengthen Digital Literacy and Psychosocial Assistance at earthquake victim posts spread across Cianjur. Implementation of community service uses a service-learning method that adapts the learning system to the situation and needs of the community. Community service activities are carried out using teaching aids that can be received by earthquake victims in the form of fun-learning with a duration of two hours. The first session is filled with introductions and storytelling, then learning while playing in the second session. The group was divided into two separate locations, namely the children's group and the parents' group. The aim is for each group to receive different material reinforcement according to their individual needs. Implementation of activities is carried out in collaboration with various parties (between universities, company CSR and local stakeholders). The results of the activities carried out showed that participants were able to understand digital culture, ethics in using social media, and the Personal Data Protection Law