Publication Search

Computational Thinking skills are students' ability to use mathematical concepts to solve problems systematically using algorithms and programming. This research aims to measure the development of Computational Thinking skills of students at SMA N 3 Medan on algebraic function limits material. The research method used is survey research with data collection using tests and observations. The subjects of this research were class XI students from SMA N 3 Medan who were studying algebraic function limits. The research results show that there is significant development in students' Computational Thinking skills after studying the material. The implication of this research is the need for a learning approach that focuses more on developing Computational Thinking skills in mathematics learning in high school.

MSIB (Certified Independent Study and Internship) is one of the activity programs at the Merdeka Campus which aims to help students improve their skills and develop themselves. MSIB appointed Orbit Future Academy as one of the partners in the Independent Study program. Founded in 2016 with the aim of improving the quality of life through innovation, education and skills training. In accordance with its mission, namely "We curate and localize international programs and courses for upskilling, re-skilling youth, and the workforce towards jobs of the future". Partners provide opportunities for students to take Artificial Intelligence programs and study online. Learning consists of eight material courses including Python Programming, AI Technology Logic and Concepts, AI Project Cycle, AI Research Methods, ChatGPT, Professional and Company Ethics, Financial Literacy and ending with a Final Project. The final project scope carried out is the Occupancy Detection in Parking Lot project. This project uses the Computer Vision domain with the selection of the YOLO model in detecting objects and pixel segmentation. The project begins with selecting a dataset using roboflow which then goes through data pre-processing for cloning, annotation and augmentation. Then the model is trained using machine learning and deep learning algorithms to understand patterns and characteristics related to parking spaces. Once trained, the AI model will be validated using test data. This aims to ensure that the model truly recognizes the presence of the vehicle. Next, form the application design in creating an informative interface using wireframes. Then enter the deployment stage so that the system can be accessed widely and easily via the web. Lastly, field testing is to find out the performance of the application that has been designed.      

Computer vision technology is used to improve work safety in the construction industry. The key in this project is the utilization of the YOLO method on the Roboflow platform. In addition to the Convolutional Neural Networks (CNN) algorithm, YOLO efficiently divides the image into a grid and classifies the objects in the grid by bounding box and confidence score. With the integration of YOLO, this project can achieve accurate and fast PPE detection. This project uses the YOLO method to detect head and body parts from input images. The detected body parts are then cropped and processed using the CNN method for classification. This project will also implement computer vision algorithms, including Deep Learning methods that currently have the most significant results in image recognition is CNN method, to automatically detect and monitor the use of PPE. This model achieves mAP 64.1%, Precision 73.2%, and Recall 60.2%. The Streamlit framework was used for deployment, creating a web application for PPE compliance tracking. This project, ''Health and Safety PPE Compliance Tracking'', aims to improve work safety in the construction industry. This project uses Computer Vision technology to detect, monitor, and ensure worker compliance with the use of appropriate PPE. The suggestion is to conduct further trials using other datasets in the form of photos or videos that can be done in real-time by ensuring that the colors of hats and vests do not vary too much to detect the conformity of labeling with PPE use.

Accurate image segmentation is a pivotal process in medical imaging, essential for supporting diagnosis, treatment planning, and monitoring disease progression. This study evaluates the effectiveness of machine learning algorithms, including U-Net, Fully Convolutional Networks (FCNs), and Mask R-CNN, in achieving high-precision segmentation of medical images. Experimental results demonstrate that these models significantly enhance segmentation accuracy, enabling more precise diagnostic outcomes in clinical settings and advancing the development of automated medical imaging technologies.

This study aims to review the use of generative artificial intelligence (AI) in strategic decision evaluation, with a focus on consistency and bias in business decision making. Through a qualitative literature review approach, this study analyzes various studies that examine how AI technology, such as the GPT model, can improve decision quality by providing more objective and consistent data analysis. Although it has great potential in reducing human bias, this study also shows the risk of algorithmic and data bias that can affect decision outcomes. Therefore, the use of AI in decision making must be accompanied by strict human supervision to ensure the quality and fairness of the resulting decisions. The results of this study provide an important contribution to the understanding of the challenges and opportunities of AI in strategic decision making in the business world.

Fraudsters increasingly exploit unauthorized credit card information for financial gain, targeting un-suspecting users, especially as financial institutions expand their services to semi-urban and rural areas. This, in turn, has continued to ripple across society, causing huge financial losses and lowering user trust implications for all cardholders. Thus, banks cum financial institutions are today poised to implement fraud detection schemes. Five algorithms were trained with and without the application of the Synthetic Minority Over-sampling Technique (SMOTE) to assess their performance. These algorithms included Random Forest (RF), K-Nearest Neighbors (KNN), Naïve Bayes (NB), Support Vector Machines (SVM), and Logistic Regression (LR). The methodology was implemented and tested through an API using Flask and Streamlit in Python. Before applying SMOTE, the RF classifier outperformed the others with an accuracy of 0.9802, while the accuracies for LR, KNN, NB, and SVM were 0.9219, 0.9435, 0.9508, and 0.9008, respectively. Conversely, after the application of SMOTE, RF achieved a prediction accuracy of 0.9919, whereas LR, KNN, NB, and SVM attained accuracies of 0.9805, 0.9210, 0.9125, and 0.8145, respectively. These results highlight the effectiveness of combining RF with SMOTE to enhance prediction accuracy in credit card fraud detection.

Analysis of the use of the Boruvka algorithm method in the context of electricity supply is an important topic in the development of electricity systems. In this study, we evaluate the effectiveness and application of the Boruvka algorithm in distribution optimization and power grid management. We study ways in which the Boruvka algorithm can be used to identify optimal electricity distribution paths, improve system efficiency, and minimize the potential for grid damage or failure. This research provides a deeper understanding of the potential and limitations of the Boruvka algorithm in the context of modern electricity infrastructure.

Collaborative filtering systems rely heavily on matrix factorization techniques, which often face scalability issues when handling large datasets. This paper presents an efficient parallel algorithm that leverages distributed computing to perform largescale matrix factorization. Experimental results show that our algorithm significantly reduces computation time while maintaining high accuracy. The approach has practical implications for recommendation systems, particularly in ecommerce and social media platforms.

This paper introduces a hybrid optimization approach that combines genetic algorithms with gradient descent for effective nonlinear function approximation in highdimensional data. Traditional methods struggle with computational efficiency and accuracy in such complex spaces. By integrating genetic algorithms to provide a global search strategy with gradient descent for finetuning, the proposed method achieves faster convergence and improved accuracy. Simulations and case studies demonstrate its effectiveness in applications like data mining, image recognition, and financial modeling.

Labor scheduling is allocating human resources to work stations according to needs. To increase productivity, companies must schedule labor optimally. In allocating labor, the Hotel Menara Lexus company has implemented labor scheduling with 3 work shifts, namely morning, afternoon and night. Because scheduling is still made manually, it is possible for schedule "collisions" to occur and there is no fairness in the distribution of shifts to each employee. To solve this problem, an appropriate algorithm is needed so that the scheduling process can run optimally. The application of genetic algorithm methods in scheduling problems is able to produce good solutions using chromosome representation, determining objective function values, fitness evaluation, determining probability function values, selection process using Roulette Wheel, crossover method using one cut-point crossover, and mutation at the above level. 0.5 can then be generated. From parameter testing, the results obtained  

This study explores the utilization of LightGBM, a gradient-boosting framework, to classify the inhibitory activity of beta-secretase 1 inhibitors, addressing the challenges of Alzheimer's disease drug discovery. The study aims to enhance classification performance by focusing on overcoming the limitations of traditional statistical models and conventional machine-learning techniques in handling complex molecular datasets. By sourcing a dataset of 7298 compounds from the ChEMBL database and calculating molecular descriptors for each compound as features, we employed LightGBM in conjunction with a set of carefully selected molecular descriptors to achieve a nuanced analysis of compound activities. The model's efficiency was benchmarked against traditional machine-learning algorithms, revealing LightGBM's superior accuracy (84.93%), precision (87.14%), sensitivity (89.93%), specificity (77.63%), and F1-score (88.17%) in classifying beta-secretase 1 inhibitor activity. The study underscores the critical role of molecular descriptors in understanding drug efficacy, highlighting LightGBM's potential in streamlining the virtual screening process. Conclusively, the findings advocate for LightGBM's adoption in computational drug discovery, offering a promising avenue for advancing Alzheimer's disease therapeutic development by facilitating the identification of potential drug candidates with enhanced precision and reliability.

Learning using media based on the Canva application has the advantage that the learning material is made more interesting. As support for direct learning, the use of the Canva application which has a variety of saved file formats, according to Rahmatullahet al (2020), the use of Canva-based audio visual learning media can improve students' informatics learning outcomes. There are also research results from Vivi et al (2021) which state that using the Canva application can help students fulfill online learning assignments at school. According to Yanti in Assidiqi (2020), learning using Canva-based learning media can help students understand learning even though they are at home because there are obstacles in the form of not having textbooks. This is known from the lack of student data recapitulation of the low learning outcomes of students who reach the KKM, namely 14 students out of a total of 38 students in class X.9

household industry or usually known quantity when UMKM  is an arising chance. Rephrase. This happens because there are very few job break. However, some UMKMs still have adversity determining the optimal  production abundance broad- based on accessible expedient, so  these troop cannot amplify. One of them is Bu Ida fruit juice, which consistently produces fruit juice according to crowd-pleasing orders of average consumer credit. admittedly utilize capital and basic materials enough, Ms. Ida's business is growing day by day. From this research, problems were distinguish and answer live needed, comprehend the employment of linear programming to determine the  sales abundance that absence to be optimized to attain maximum earnings from juice sales Ida's fruit. There are 6 (six) juices and 6 (six) ingredients. In this study, a simple method linear programming algorithm is applied to obtain an optimization model to maximize the  juice production quantity and obtain optimal profits. The POM-QM Windows application provides fast, accurate and precise  calculation results.

Indonesia's golden opportunity to take part in a world-class soccer competition at the U-20 World Cup competition was wiped out, as FIFA gave the decision to revoke Indonesia's status as host of the U-20 World Cup. Indonesian netizens who felt disappointed expressed their opinions and trended on social media Twitter. This research focuses on sentiment analysis of tweets using a combination of FastText embeddings method for word vectorization and using LSTM type RNN algorithm for sentiment classification. The dataset used totals 9,645 data consisting of 4,141 positive data and 5,504 negative data taken from March 29, 2023 to April 05, 2023. The test results on the LSTM model provide the best performance with an accuracy value of 74.92%, precision 74.74%, recall 74.92%, and f1-score 74.78%. The conclusion of this research is that the majority of datasets have negative sentiments, which means that people are more likely to give negative opinions than to provide support to Indonesian football which is experiencing problems. It is hoped that with this conclusion in the future people will better control their opinions and provide positive opinions when Indonesia is experiencing problems.

The occurrence of scorch during the production of flexible polyurethane is a significant issue that negatively impacts foam products' resilience and generally jeopardizes their integrity. The likelihood of foam product failure can be decreased by optimizing production variables based on machine learning algorithms used to predict the occurrence of scorch. Investigating technology is required because prevention is the best approach to dealing with this problem. Hence, machine learning algorithms were trained to predict the occurrence of scorch using the thermodynamic profile of polyurethane foam, which is made up of recorded production variables. A variety of heuristics algorithms were trained and assessed for how well they performed, namely XGBoost, Decision trees, Random Forest, K-nearest neighbors, Naive Bayes, Support Vector Machines, and Logistic Regression. The XGboost ensemble was found to perform best. It outperformed others with an accuracy of 98.3% (i.e., 0.983), followed by logistic regression, decision tree, random forest, K-nearest neighbors, and naïve Bayes, yielding a training accuracy of 88.1%, 66.7%, 84.2%, 87.5%, and 67.5% respectively. The XGBoost was finally used, yielding 2-distinct cases of non(occurrence) of scorch. Ensemble demonstrates that it is quite capable and is an effective way to predict the occurrence of scorch.

The increasing development of technology and increasing buying and selling activities require every business owner to adapt to technological developments, for business owners selling or processing sales data is very important as is done by PT.XYZ household furniture items such as tables, chairs, dressers, wardrobes, sofas and many more, where sales data is still done manually, such as a lack of reviewing what products consumers need and ineffective data storage. To overcome this problem the researcher tried to implement it using one of the methods available in data mining, namely the K-Means Algorithm. –certain data groups (clusters). So, by grouping this data, the company can find out which items are selling best and which are not selling well. So that the goods in the warehouse do not pile up. From this research, the resulting output is 5 of the best-selling items, and 5 of the least-selling items. With the data processing carried out, it is hoped that it can provide solutions to the company so that they can find out which items are the best-selling and best-selling items.

This research aims to investigate the critical role of an analysis stage in detecting spelling errors in news texts. Spelling errors in news text can hinder understanding and reduce the credibility of the information. The analysis stage in this research includes identifying spelling error patterns, selecting effective analysis methods, and developing a spelling error detection algorithm. Through this approach, this research aims to improve the quality of news texts by reducing the number of spelling errors that may occur. The research method involves analyzing news texts that have been identified as potential sources of spelling errors. It is hoped that the research results will provide valuable insights into the development of automatic systems to detect and correct spelling errors in news texts, thereby increasing the accuracy and credibility of information conveyed through this media

In the era of Big Data, securing sensitive information and ensuring data integrity have become paramount concerns due to the unprecedented volume and intricacy of data. Traditional security algorithms face significant challenges in adapting to the distinct characteristics of Big Data. This literature review explores the evolution of data security algorithms tailored explicitly for the Big Data landscape, aiming to address the increasing demand for robust security solutions capable of handling the unique challenges posed by the massive scale and complexity of data. By scrutinizing existing literature, the review unveils advancements, trends, and innovations developed by researchers and practitioners to mitigate vulnerabilities associated with handling vast datasets. The review also sheds light on emerging technologies and cryptographic techniques specifically designed for Big Data security, contributing to enhanced confidentiality, integrity, and availability in the face of evolving cyber threats. While these developments offer advantages such as improved data protection and threat detection, the review highlights challenges, including algorithmic bias, computational complexity, privacy trade-offs, and a shortage of skilled workforce. By considering these factors and emphasizing continuous improvement and ethical considerations, organizations can responsibly leverage data security algorithms to enhance information security in the era of Big Data.

The rapidly evolving landscape of cryptanalysis necessitates an urgent and detailed exploration of the high-degree non-linear functions that govern the relationships between plaintext, key, and encrypted text. Historically, the complexity of these functions has posed formidable challenges to cryptanalysis. However, the advent of deep learning, supported by advanced computational resources, has revolutionized the potential for analyzing encrypted data in its raw form. This is a crucial development, given that the core principle of cryptosystem design is to eliminate discernible patterns, thereby necessitating the analysis of unprocessed encrypted data. Despite its critical importance, the integration of machine learning, and specifically deep learning, into cryptanalysis has been relatively unexplored. Deep learning algorithms stand out from traditional machine learning approaches by directly processing raw data, thus eliminating the need for predefined feature selection or extraction. This research underscores the transformative role of neural networks in aiding cryptanalysts in pinpointing vulnerabilities in ciphers by training these networks with data that accentuates inherent weaknesses alongside corresponding encryption keys. Our study represents an investigation into the feasibility and effectiveness of employing machine learning, deep learning, and innovative random optimization techniques in cryptanalysis. Furthermore, it provides a comprehensive overview of the state-of-the-art advancements in this field over the past few years. The findings of this research are not only pivotal for the field of cryptanalysis but also hold significant implications for the broader realm of data security.

The work POLA-POLA was inspired by the online game Higgs Domino where the creator was interested in the algorithmic patterns of the game. This game is not only played by teenagers but is also played by children and even the elderly. This game has its own uniqueness, namely that there are boxes where the spin moves from top to bottom so that wins and losses appear in the box. The artist interpreted these boxes in the work POLA-POLA by using six cubes of different sizes and each side has a different color, consisting of red, yellow, green and white. In creating the dance work entitled POLA-OLA, the methods of observation, exploration, improvisation and formation were used. Supporting the work consists of six dancers, namely three male dancers and three female dancers. The POLA-POLA dance work was performed at the Hoerijah Adam Performance Building, Padangpanjang Indonesian Arts Institute. This dance work is presented in the form of a contemporary dance work.