Nowadays, online buying and selling transactions have become a habit and lifestyle for Indonesian people. Online-based buying and selling transaction activities make it easier and faster for people in terms of service and convenience. One of the services in question is the practicality of sending goods and not requiring buyers to go to the seller's location. Generation Z is one of the generations born with the advancement of information technology. Therefore, it is necessary to develop a learning module that explains the design or development of a Website-based E-Commerce platform using the Waterfall method. The Waterfall method is used as a guide for researchers in developing an E-Commerce website design consisting of several stages. While the final result of the website design will be presented in the form of a module accompanied by an installation process to facilitate understanding of Generation Z in implementing an E-Commerce website. This study involved 2 experts, including academics and practitioners who know about the digital economy and informatics. The involvement of experts in the validation session by applying the Likert scale as a measurement scale. The results of the E-Commerce design validation can be said to be good so that it needs to be developed in a further stage. Testing on the E-commerce design made using the Black Box Testing technique to measure the suitability of the features contained in the E-Commerce design.

This study explores the application of the Transformer model in sentiment analysis of tweets generated by ChatGPT. We used a Kaggle dataset consisting of 217,623 instances labeled as "Good", "Bad", and "Neutral". The Transformer model demonstrated high accuracy (90%) in classifying sentiments, particularly predicting "Bad" tweets. However, it showed slightly lower performance for the "Good" and "Neutral" categories, indicating areas for future research and model refinement. Our findings contribute to the growing body of evidence supporting deep learning methods in sentiment analysis and underscore the potential of AI models like Transformers in handling complex natural language processing tasks. This study broadens the scope for AI applications in social media sentiment analysis.

Cyber-crime is becoming more massive as online activities increase. Cybercrime is a criminal act that exploits digital technology to damage, harm, and destroy property. Therefore, it is crucial for internet users to have knowledge of cybersecurity and the world of technology and the internet in order to avoid falling victim to cybercrime. The aim of this study is to develop a chatbot system as a centralized information medium on cybersecurity, technology, and the internet for internet users. The development of this chatbot aims to reduce the risks of cybercrimes and help enhance internet users' awareness of cybercrime. This research employs the AI Project Cycle method in chatbot development and utilizes the Long Short-Term Memory (LSTM) deep learning model algorithm to develop a model that achieves high accuracy. The training results of the LSTM model achieved an accuracy score of 100% and a loss of 3.09% with 400 epochs. Consequently, it can be concluded that the LSTM algorithm is highly effective for training and developing a chatbot model.

XYZ University has a scholarship program intended for outstanding students. Determination of recipients of outstanding scholarships is still constrained by the unclear system for determining scholarship recipients. This can affect the fairness of receiving scholarships. Students who really deserve to get a scholarship do not receive the scholarship. The solution to overcome this problem, the researcher came up with an idea in the form of a Multi-Objective Optimization by Ratio Analysis (MOORA) Decision Support System method, with the aim of being able to solve the problem of selecting an objective scholarship acceptance process. The MOORA method was chosen because it can solve the problem with a fairly effective method. After conducting research, it was found that 5 (five) candidates were eligible to receive scholarships for outstanding students, namely Alt7, Alt10, Alt37, Alt49, and Alt35. Furthermore, from the results of this study, it was concluded that students who get achievement scholarships are students with the highest scores from the results of the MOORA method, there is no element of subjectivity.

Based on the results of previous studies, research on machine learning for predicting ICU patients is crucial as it can aid doctors in identifying high-risk individuals. A high accuracy in machine learning models is necessary for assisting doctors in making informed decisions. In this study, machine learning models were developed using two models, namely Random Forest and Artificial Neural Network (ANN), to predict patient mortality in the ICU. Patient data was obtained from The Global Open Source Severity of Illness Score (GOSSIS) and underwent preprocessing to address issues of missing values and imbalanced data. The data was then divided into training, validation, and testing sets for model training and evaluation. The results of the study indicate that the Random Forest model performs better with an accuracy of 93% on the testing data compared to the ANN which only achieved an accuracy of 86% on the testing data. Consequently, the Random Forest model can be utilized as a solution for predicting patient mortality in the ICU.

Information and communication technology continues to develop and progress which demonstrated by the presence of gadget technology. Gadgets are smart electronic devices that assist in making it simple for users to accomplish various task. The use of gadget technology in children are unable to be separated. According to the 2020 KPAI survey, approximately 71,3% of school-age children own and have played with gadgets for a longer time. As a result, it is expected that early detection of gadget addiction can be carried out to ensure that mental and emotional disorders in children who use gadgets can be properly addressed. The aim of this research is to create a prototype expert system for early detection of gadget addiction levels in children using the fuzzy tsukamoto. The fuzzy tsukamoto method was used in this study. This study included 74 respondents aged 9 to 12 years old. The DAS (Digital Addiction Scale : For Children) was used as the data collection method in this study. The system’s as performance will be evaluated using 74 respondents data by comparing the result of expert calculations and fuzzy tsukamoto method calculations. Fuzzy Tsukamoto reasoning with 64 rule bases in used to build this expert system. According to evaluation with 74 respondent data, this expert system has a system acurracy rate of 87,83%, which indicates that it proceeds succesfully.

Measurement of temperature and humidity generally uses a tool that can determine the value of two physical quantities of a material or system (thermometer or hygrometer). When it comes to hatchery incubators, success and protection are priorities. The microcontroller is a small computer in the form of a chip, DHT22 temperature and humidity sensor with a range of (-40)-80°C. Formulation of the problem, (1) How to get digital data of hatching incubator room temperature using a microcontroller? (2) How is the application of the system to hatching incubators?. Methods of research, analysis, implementation, and simulation (a) Analysis, the lowest ideal temperature for hatching eggs shows a figure of approximately 38°C and the highest is 38.5–39°C. (b) Implementation, starting with the use of DHT22 giving a temperature signal according to whether or not when the condition of the incandescent lamp as a heater will turn on or off, the display will be displayed on the LCD. (c) Simulation, the first stage is the DHT22 schematic to Arduino, the second stage is the relay schematic to Arduino, the third is I2C LCD to Arduino, the fourth is relay to lights and indicators. Based on the analysis, implementation and simulation, conclusions are drawn (1) Digital data related to temperature from the DHT22 sensor displayed on the LCD can be applied to help monitor the hatching process of chicken eggs using an incubator. (2) Arduino board-based microcontrollers can be applied as controllers related to system flow in chicken egg hatching incubators.

This study aims to evaluate usability on the user's M-Pise LPD Digital application page in Jembrana Regency by using the usability testing method with Performance Measurement and RTA (Retrospective Think Aloud) techniques. The usability aspects reviewed are effectiveness, efficiency, and user satisfaction. In this study there were 20 respondents who were involved consisting of a group of advanced respondents and a group of novice respondents. The results showed that (1) the M-Pise LPD Digital application was still not effective when viewed from errors (errors) when the respondent was doing the task, (2) the M-Pise LPD Digital application in terms of efficiency has proven to be efficient seen from statistical testing Mann Whiteney showed that there was no difference in time between the beginner and advanced groups of respondents so that it could be said to be efficient, (3) User satisfaction was still unsatisfied as seen from the SUS questionnaire score of 64. Thus the M-Pise application page did not have good usability. Thus the recommendations for improvement given are based on the results of performance measurements, namely errors. Recommendations for improvement will focus on improving menu components and features. Meanwhile, based on the results of the problems and suggestions for the results of the RTA, namely simplification of features and improvements as well as clarity of the layout of letters, numbers and icons.

Unemployment is a major social problem in many regions, including West Java province in Indonesia. West Java province is one of the most populous regions with a high level of urbanization. With population growth and urbanization, the challenge of creating enough jobs becomes more difficult. Therefore, the purpose of this study is to cluster open unemployment data in West Java communities classified by the number of unemployed people by district or city. This research uses CRISP-DM method with K-Means clustering algorithm. The result of this research is 10 regencies/cities that have low level of unemployment, then there are 15 regencies/cities that have medium level of unemployment and there are 2 regencies/cities that have high level of unemployment. The result of the test using Davies Bouldin Index cluster = 3 has the best cluster quality, because the value of the Davies Bouldin Index test result with c = 3 is the smallest value of 0.28, which is the lower, the better the cluster.