Air pollution has emerged as a critical issue in several major urban areas, including Jakarta. To address this problem, the study explores the utilization of the Artificial Neural Network (ANN) method. Three distinct ...
Air pollution has emerged as a critical issue in several major urban areas, including Jakarta. To address this problem, the study explores the utilization of the Artificial Neural Network (ANN) method. Three distinct approaches are investigated in this research: the Support Vector Classifier (SVC), Deep Artificial Neural Network (Deep ANN), and Long Short-Term Memory (LSTM). The available Air Quality data were utilized to train and assess the performance of the proposed ANN models. The research findings reveal that the Deep ANN model surpasses the other methods, achieving an impressive accuracy of approximately 96.57% and a cross-entropy value of 0.1103. In predicting air quality in Jakarta, Deep ANN has demonstrated superior performance when compared to SVC and LSTM. These results highlight the significant potential of deploying Deep Artificial Neural Network (ANN) methodologies for air quality prediction. Such an approach could play a pivotal role in the development of monitoring and early warning systems aimed at effectively addressing air quality issues in the Jakarta region.
Microservices and monolithic systems are two prevalent architectural approaches in software development. This study provides a complete review and analysis of the key components involved in design, development, and op...
Microservices and monolithic systems are two prevalent architectural approaches in software development. This study provides a complete review and analysis of the key components involved in design, development, and operation in software development. A systematic review of the literature was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. This review examined various sources debating monolithic systems vs microservices, highlighting their benefits, drawbacks, and implementation in enterprises. The paper addresses important research questions, aiming to further analyze this architectural approach. Findings show that microservices offer benefits such as scalability, flexibility, and independent deployment, while monolithic systems provide simplicity and ease of development. However, challenges related to network communication, data consistency, and operational complexity were also found with microservices. This research focuses on discussing the trade-offs and factors to consider when deciding between monolithic systems and microservices, which provides in-depth information for practitioners in decision-making for software development. This research aims to help readers understand the effects of using monolithic or microservice-based systems in software development.
Artificial Intelligence (AI) and chatbot technology have emerged as promising solutions to improve healthcare services. AI chatbots can mimic human-like interactions and assist with tasks such as triaging patients and...
Artificial Intelligence (AI) and chatbot technology have emerged as promising solutions to improve healthcare services. AI chatbots can mimic human-like interactions and assist with tasks such as triaging patients and providing medical advice. This review focuses on the use of AI chatbots in predicting diseases, aiming to explore their effectiveness and potential for early intervention and treatment. The purpose of this systematic literature review is to analyze studies related to AI chatbot technology in disease prediction. A systematic literature review was conducted, analyzing a total of 24 selected journals based on predefined inclusion and exclusion criteria. The review protocol involved examining studies published from various years, with a particular emphasis on articles from 2020. The findings indicate that AI chatbots have the potential to play a significant role in predicting diseases and assisting healthcare professionals in making informed decisions. AI chatbot technology shows tremendous potential in disease prediction. By utilizing machine learning algorithms and techniques, chatbots can enhance the accuracy and quickness of disease diagnosis. Continued research and development efforts are necessary to refine AI chatbots' capabilities and revolutionize healthcare delivery for improved patient outcomes and disease management.
This paper explores using artificial intelligence (AI) to predict stock market movements and build optimal portfolios. The research methodology involves using LSTM networks to predict stock performance. The study aims...
This paper explores using artificial intelligence (AI) to predict stock market movements and build optimal portfolios. The research methodology involves using LSTM networks to predict stock performance. The study aims to combine AI with human expertise to develop an intelligent trading system. The findings emphasize the importance of selecting appropriate AI approaches for accurate predictions and optimal portfolio management. The results of this study state that with LSTM, we can predict stock prices that are very close to their real prices. The average LSTM method in predicting stock prices is about 97.2938%. Average error obtained when using LSTM was 2.3223%.
Typhoid fever is an endemic disease that burdens Indonesia and has a potentially fatal infection multisystem. Salmonella typhi bacterium is responsible for typhoid fever disease. Poor sanitation, crowding, and slums a...
详细信息
Speech content is closely related to the stability of speaker embeddings in speaker verification tasks. In this paper, we propose a novel architecture based on self-constraint learning (SCL) and reconstruction task (R...
详细信息
Traffic accidents are a serious problem in developing countries. These accidents can be caused by poor infrastructure and the arrogance of drivers. In addition, semi-autonomous systems are automotive technologies that...
Traffic accidents are a serious problem in developing countries. These accidents can be caused by poor infrastructure and the arrogance of drivers. In addition, semi-autonomous systems are automotive technologies that have been developed in recent years and are starting to be implemented in existing vehicles. This literature review paper will mainly discuss how object detection works in semi-autonomous systems, how semi-autonomous systems operate, and whether semi-autonomous cars can reduce the number of traffic accidents. Its main objective is to build driver confidence in semi-autonomous systems by assisting in driving and supporting the development of them. In the process of writing this paper, the method we use follows the guidelines stated in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA-P).
The ability to identify cancer at an early stage is critical to increase the likelihood of effective treatment or stopping the progression of the disease in the body. Lung cancer is one of the most common deadly disea...
The ability to identify cancer at an early stage is critical to increase the likelihood of effective treatment or stopping the progression of the disease in the body. Lung cancer is one of the most common deadly diseases and quickly kills the patient. The number of deaths caused by lung cancer surpasses those of colon, rectal, breast, and prostate cancers combined. Unfortunately, only two percent of patients with advanced lung cancer survive for five years or more. However, the survival rates are better, with 49 % of the patients surviving for five years or more, if the disease is detected early. On the other hand, in this modern era, machine learning has become one of the most reliable tools in the world for healthcare, as machine learning can learn from the data obtained and process the data that will later be used to help complete certain tasks. Therefore, in this research, machine learning or especially classification algorithms such as K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Random Forest, Logistic Regression, and XGBoost are used to identify three main symptoms that serve as markers for early detection of lung cancer. Determining the three main symptoms is done by combining the results of the feature importance score on each model using the rank-averaging method. The result is an average ranking of each feature based on its combined importance with an accuracy of 93.5 %.
Hate speech is any act of provoking or insulting another person or group based on their ethnicity, religion, race, gender, sexual orientation, physical ability, or other characteristics. This can be done in a variety ...
Hate speech is any act of provoking or insulting another person or group based on their ethnicity, religion, race, gender, sexual orientation, physical ability, or other characteristics. This can be done in a variety of ways. Because of this, a study was conducted that focused on ascertaining the most effective technique for identifying hate speech. It was discovered that LSTM, a form of recurrent neural network (RNN) that is particularly good at modeling sequential data, is the most successful technique to apply. Many natural language processing (NLP) activities may be performed with it, including hate speech detection, which makes LSTM have advantages over other techniques, both in terms of long-term memory mechanisms. LSTM, also able to handle long and complex text sequences, capable of recognizing more complex patterns, able to understand and consider the context in the text, to recognize the meaning of a word or phrase in a broader context. Then, based on the dataset obtained from the Kaggle website with a total of 24783 data, then the dataset is divided into training data and validation data, with a ratio of training data to validation data. of 80%: 20%. Then, the testing step results in an accuracy value of 87.10%.
Emotionally-aware chatbots are chatbots that are equipped with emotional intelligence. Based on the literature, using emotional elements in chatbots can improve user engagement and believability factors. This study at...
Emotionally-aware chatbots are chatbots that are equipped with emotional intelligence. Based on the literature, using emotional elements in chatbots can improve user engagement and believability factors. This study attempts to make a novel contribution by empirically evaluating the impact of emotion recognition models on the believability factor of chatbots. This study examines the impact of the emotions model and avatar on chatbot interactions through three implementations. Thirty-one participants volunteered to evaluate emotionally aware chatbots. The participants evaluated the interaction with the chatbot using the Godspeed Questionnaire Series (GQS). The questionnaire results are utilized to measure the effect of the emotions model on the chatbot's believability factor. The evaluation results of the experiment show that implementing the emotion recognition model on the chatbot increases its believability factors. On average, the believability measures in interaction type B (with an emotion model) are enhanced 1.71 times compared to interaction type A (a basic model). Furthermore, the believability measures in interaction type C (with an emotion model and an avatar) are enhanced 2 times more than in interaction type A (a basic model). The believability factor is also heightened by integrating a chatbot avatar into the interaction system. Using avatars in chatbots increases the believability variables of the system by 1.17 times if compared with not using avatars.
暂无评论