SHM Publisher Journals
Not a member yet
315 research outputs found
Sort by
Comparison of gridsearchcv and bayesian hyperparameter optimization in random forest algorithm for diabetes prediction
Get PDFDiabetes Mellitus (DM) is a chronic disease whose complications have a significant impact on patients and the wider community. In its early stages, diabetes mellitus usually does not cause significant symptoms, but if it is detected too late and not handled properly, it can cause serious health problems. To overcome these problems, diabetes detection is one of the solutions used. In this research, diabetes detection was carried out using Random Forest with gridsearchcv and bayesian hyperparameter optimization. The research was carried out through the stages of study literature, model development using Kaggle Notebook, model testing, and results analysis. This study aims to compare GridSearchCV and Bayesian hyperparameter optimizations, then analyze the advantages and disadvantages of each optimization when applied to diabetes prediction using the Random Forest algorithm. From the research conducted, it was found that GridSearchCV and Bayesian hyperparameter optimization have their own advantages and disadvantages. The GridSearchCV hyperparameter excels in terms of accuracy of 0.74, although it takes longer for 338,416 seconds. On the other hand, Bayesian hyperparameter optimization has a lower accuracy rate than GridSearchCV optimization with a difference of 0.01, which is 0.73 and takes less time than GridSearchCV for 177,085 seconds
Optimizing the implementation of the BFS and DFS algorithms using the web crawler method on the kumparan site
Get PDFEfficient access to timely information is critical in today's digital era. Web crawlers, automated programs that navigate the Internet, play an important role in collecting data from websites such as Kumparan, a leading news site in Indonesia. This research shows the effectiveness of the Breadth-First Search (BFS) and Depth-First Search (DFS) algorithms in indexing Kumparan content. The results of the research show that BFS consistently indexes more files comprehensively but with longer execution times compared to DFS, which provides faster initial results but with fewer files. For example, at depth 4 BFS indexed 949 files in 886.94 seconds, while DFS indexed 470 files in 233.02 seconds. These findings highlight the balance between precision and speed when selecting a crawling algorithm tailored to the needs of a particular website. This research provides insights into optimizing web crawler technology for complex websites such as Coil and suggests avenues for further research to improve permission efficiency and adaptability across a variety of crawling scenarios
Developing a classification system for brain tumors using the ResNet152V2 CNN model architecture
Get PDFAccording to The American Cancer Society, in 2021 there were 24,530 cases of brain and nervous system tumors. The National Cancer Institute reports that there are approximately 4.4 new cases of brain tumors per 100,000 men and women per year. Brain tumors can be detected using magnetic resonance imaging (MRI), a scanning tool that uses a magnetic field and a computer to record brain images and is able to provide clear visualization of differences in soft tissue such as white matter and gray matter. However, this cannot be done optimally because it still relies on manual analysis, so it cannot classify brain tumor types on larger datasets with the potential for error and a low level of accuracy. To accurately determine the type of brain tumor, a better classification method is needed. The aim of this study is to determine the accuracy of brain tumor calcification using the deep learning model. In this study, the classification of brain tumor types was carried out using the ResNet152V2 convolutional neural network (CNN) model which has a depth of 152 layers. The dataset used in this study was 7,023 MRI images of brain tumors consisting of 1,645 meningiomas, 1,621 gliomas, 1,757 pituitary and 2,000 normal. Research results show an accuracy value of 94.44%, so it can be concluded that the ResNet152V2 model performs well in classifying brain tumor images and can be used as a medium for physicians to more accurately diagnose brain tumor patients more accurately
Quadrotor height control system using LQR and recurrent artificial neural networks
Get PDFThe quadorotor is a type of unmanned flying vehicle known as Unmanned Aerial Vehicle (UAV). In recent years, quadrotors have attracted much attention from researchers around the world due to their excellent maneuverability. A good control system in this quadrotor system is needed for ease of use of this quadrotor. One control system that is often used is the Linear Quadratic Regulator (LQR) control system. This control system has challenges for dynamic system disturbances in quadrotor control. Researchers proposed a recurrent artificial neural network (RNN) system to address these challenges.RRN is used to change the value of the feedback component in the LQR control system. The nature of the feedback component in LQR, which is static, is changed based on the system error value based on changes in the error value entered into the RNN. The result of this RNN is a change in the value of the LQR feedback component based on the input of the system. The results of this research show that LQR control with RNN produces a faster system response of 0.075 seconds and a faster settling time of 0.221 seconds. Compensation for the system response speed produces a higher overshot value
Sentiment based-emotion classification using bidirectional long short term-memory (Bi-LSTM)
Get PDFSocial media is now an important platform for sharing information, expressing opinions, and daily feelings or emotions. The expression of emotions such as anger, sadness, fear, happiness, disappointment, and so on social networks can be further analyzed either for business purposes or just analyzing the habits of a community or someone's posts. However, analyzing manually will be a time-consuming process, and the use of conventional methods can affect the results of less accurate accuracy. This research aims to improve the accuracy of recognizing emotions in text by using the Bidirectional Long Short Term Memory (Bi-LSTM) method, which is a subset of RNNs that tend to be more stable during training and show better performance on various NLP and other processing tasks. The method used includes several stages, namely preprocessing, tokenization, sequence padding, and modeling. The results of this study show that the Bi-LSTM model is capable of predicting emotions in text with an accuracy of 94.45% because it excels in handling the temporal context and can avoid vanishing gradients
Analysis of ODP point placement using algorithm K-means in RW. 01 Gendongan village (case study: PT. Indomedia)
Get PDFThis study discusses the placement of ODP points for designing Fiber To The Home (FTTH) networks in RW. 01 Gendongan Village using the K-Means algorithm. The purpose of this study is to facilitate the determination of the optimal location of the Optical Distribution Point (ODP) without the need for manual determination by the network designer. The initial stage of the study began with five stages, namely Determining the location of the study, Conducting surveys and data collection, Determining the location of the ODP placement using K-Means, Network design, Finished. The K-Means algorithm is used to determine the best ODP placement point after the study was conducted. The results of this study are divided into two stages, namely determining the location of the ODP placement and creating an FTTH access network scheme using Google Earth Pro software. The results obtained from using the K-Means algorithm with a value of K = 8 need to be adjusted. ODP adjustments are made to ODPs located in houses or in the middle of the road which will later be shifted to the shoulder of the road. Distribution cable design is carried out at the location of the ODP point that has been adjusted. The design of this distribution cable has 4 paths, each path has 2 ODPs. Previous research has focused on residential areas with relatively small coverage, while the current research covers a wider area, namely RW. This significant difference shows a shift in focus from a small scale to a larger scale so that it can optimize the deployment of FTTH networks in wider areas and improve more services
The potential of the mohamad toha area as a street food tourism destination in cirebon city
No full textThe Mohamad Toha area is located in Kebonbaru Village, Kejaksan District, Cirebon City, in the center of Cirebon city and only ± 200m from the Cirebon mayor's office. This street is filled with various food vendors presenting typical Indonesian flavors with a local touch, from heavy food to sweet snacks, this street has everything to pamper the taste buds of its visitors. This research aims to discover the potential of the Mohamad Toha area as a tourist destination in Cirebon City. By using a qualitative approach and 3A’s analysis, this research identifies what culinary tourism in the Mohamad Toha area could be a potential tourist visit, as well as how to analyze the 3A’s, namely the attractions presented, the amenity available in the area and the accessibility to the area. The findings show that the Mohamad Toha area of Cirebon City has culinary tourism potential which can be an attraction for local tourists and tourists who are on holiday in Cirebon City, and based on the 3A’s analysis the Mohamad Toha Tourism Area is quite supportive of being a culinary tourism attraction in Cirebon City, having various attractions, street culinary, easy access and close to the city center, as well as sufficient supporting facilities
Comparative study of marker-based and markerless tracking in augmented reality under variable environmental conditions
Get PDFAugmented reality (AR) technology integrates virtual content into real environments using two main methods: marker-based and markerless tracking. Marker-based tracking relies on printed markers for object placement, while markerless uses environmental features for flexibility and accuracy. This research aims to evaluate the combined impact of environmental factors-distance, angle, and lighting-on these two methods. The Multimedia Development Life Cycle (MDLC) methodology was applied by testing 72 combinations of indicators: distance (5-120 cm), angle (30°, 45°, 90°), and light color (red, blue, green, yellow) using Xiaomi Note 8 and Google Pixel 4. Results show markerless tracking is superior in all conditions, achieving a 94.4% success rate on both devices. In contrast, marker-based tracking only achieved 72.2% (Xiaomi Note 8) and 77.8% (Google Pixel 4). Markerless tracking was optimally performed from 50 cm away and up close, while marker-based tracking degraded in performance at long distances and red lighting. Markerless tracking proved to be more reliable and consistent, suitable for dynamic and diverse environments, while marker-based methods remained relevant for short distances and controlled lighting. These findings provide guidance for AR developers in choosing a tracking methodology according to application needs
Detection and prediction of rice plant diseases using convolutional neural network (CNN) method
No full textRice is a basic staple food in many Asian countries and is generally irreplaceable. Rice accounts for almost half of Asia food expenditure. Rice is too a crop that is prone to plant disease. It can appear and cause a decline in the quality of rice. However, constant monitoring of the rice fields can prevent the infection of the disease. Therefore, detection and prediction of rice plant diseases is one of the topics that will be discussed in this research. The purpose of this research is to help farmers to quickly pinpoint the disease of rice plants and take care of it properly. The methods used in this paper is researching and redesigning the previous attempt to hopefully make it better and more accurate. We will be using Convolutional Neural Network (CNN) models VGG16 as our algorithm. The results are that our proposed method has more accuracy than previous research using a similar dataset. The novelty of this paper is the increased accuracy of rice plant disease detection
IoT-Integrated Smart Attendance and Attention Monitoring System For Primary and Secondary School Classroom Management
No full textThe monitoring of student attendance is a crucial aspect of the assessment of academic performance. The conventional methods for monitoring student attendance have inherent limitations in terms of both time efficiency and accuracy. Consequently, there is a clear need for a more expedient and precise attendance system. The objective of this research is to present the design of a real-time attendance recording and monitoring system for students from elementary school to senior high school, which will be implemented using the concept of the Internet of Things (IoT). The proposed system employs biometric technology in the form of face recognition. The methodology commences with the capture of images of students who leave the classroom during the instructional period via an active camera positioned on the classroom door. The system employs a Convolutional Neural Network (CNN) algorithm and a powerful computer vision tool, OpenCV, to perform real-time face recognition. Teachers will be informed of student absences and returns, as well as at the 15th and 30th minutes. An absence exceeding 30 minutes is classified as truancy. The integration of sophisticated technologies, such as machine learning and image processing, not only enhances the precision of attendance records but also equips educators with an efficient and automated system for streamlining classroom attendance management. This not only optimizes the learning environment but also facilitates more advanced and efficient pedagogical practices