Garuda - Garba Rujukan Digital

A Workshop on Basic Statistics With The R Programming Language at SMA Erenos Tangerang Valensius Jimy; Dhela Asafiani Agatha; Ferdinand Nathaniel Widjaya; Bakti Siregar
Jurnal Syntax Transformation Vol 4 No 11 (2023): Jurnal Syntax Transformation
Publisher : CV. Syntax Corporation Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.46799/jst.v4i11.869

The goal of the article is to find out and interpret the effect of giving a workshop on basic statistics with the R Programming Language at SMA ERENOS Tangerang. Researchers used descriptive statistics and basic inferential statistics to analyse student performances. The sample was taken as seventy-two students from SMA Erenos, where the actual number of students was ninety-six for grades X and XI majoring in science and social studies. The score of students was carried out directly from the Quizziz website and it is divided into two sessions, namely the Pre-Test and Post-Test. The results of this study obtained a real influence on learning basic statistics before and after the workshop. Therefore it can be concluded that when students have been given the right method to learn statistics or any other subject, it would be increased the performance of the students.

Analisis Data dalam Supply Chain Management: Klasifikasi dan Pemantauan Data Gudang PT Rajawali Nusantara Indonesia Natalie, Karen; Siregar, Bakti
FARABI: Jurnal Matematika dan Pendidikan Matematika Vol 7 No 1 (2024): FARABI
Publisher : Program Studi Pendidikan Matematika FKIP UNIVA Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47662/farabi.v7i1.705

This internship report discusses the application of data analysis in supply chain management, with a focus on classification and monitoring of warehouse data at PT Rajawali Nusantara Indonesia. In this report, Excel and R Studio are adopted as data analysis tools to improve efficiency in warehouse management. This research covers two main aspects: first, implementation of a warehouse data classification system using R Studio and Excel to optimize the allocation and placement of goods. Second, using R Studio and Excel for real-time monitoring of warehouse activities, including item selection, delivery, and inventory conditions. Through an experimental approach and historical data analysis, this research evaluates the impact of using Excel and R Studio on operational efficiency and warehouse response speed. The results are expected to provide insight for companies in improving warehouse management.

The Comparative Analysis of Integrated Moving Average and Autoregressive Integrated Moving Average Methods for Predicting Bitcoin Returns Brigita Tiara Elgityana Melantika; Kalfin; Siregar, Bakti; Wiwik Wiyanti
Journal of Mathematics, Computations and Statistics Vol. 7 No. 2 (2024): Volume 07 Nomor 02 (Oktober 2024)
Publisher : Jurusan Matematika FMIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35580/jmathcos.v7i2.3788

The rise in popularity of cryptocurrencies such as Bitcoin across various platforms has attracted the attention of young investors, making it easier for them to invest. However, due to the volatile nature of Bitcoin, this type of investment carries a high risk. Therefore, this research conducts an analysis of stock return prices to minimize losses and help investors make effective investment decisions through stock price prediction. The focus of this study is on predicting Bitcoin stock returns by analyzing closing price data over the past five years (2019-2024). The methods used are a comparison between Integrated Moving Average (IMA) and Autoregressive Integrated Moving Average (ARIMA) with a quantitative approach using R Studio software. One of the main focuses of this research is the comparison of error estimation values between the two methods, namely Mean Absolute Error (MAE) and Root Mean Square Error (RMSE). The data analyzed comprises the daily closing prices of Bitcoin over the last five years, which is publicly accessible data. The best model for predicting the daily return of Bitcoin stock is the ARIMA (1,0,1) model. The predicted values for the next five days, from May 27, 2024, to May 31, 2024, are 0.0016632438, 0.0007991618, 0.0013415932, 0.0010010794, and 0.0012148386. The ARIMA (1,0,1) model has error measurement values with an MAE of 2.3% and an RMSE of 3.5%. It is hoped that this research will provide a better understanding of the effectiveness and relative advantages of the IMA and ARIMA methods in forecasting cryptocurrency returns, thereby offering more accurate guidance for investors in making investment decisions.

Comparative Analysis of K-Means and K-Medoids Algorithms for Product Sales Clustering and Customer Yosia; Siregar, Bakti
Journal of Mathematics, Computations and Statistics Vol. 7 No. 2 (2024): Volume 07 Nomor 02 (Oktober 2024)
Publisher : Jurusan Matematika FMIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35580/jmathcos.v7i2.4053

In today's rapidly evolving business landscape, effective product management is crucial for maintaining a company's competitive advantage. Comprehensive analysis is essential for providing insights that inform strategic business development decisions. This study examines the sales data of PT XYZ from July 2020 to May 2024 using the K-Medoids algorithm, with dimensionality reduction applied through Principal Component Analysis (PCA). The clustering results identified three customer segments: Cluster 1 with 46 customers, Cluster 2 with 76 customers, and Cluster 3 with 62 customers. For product segmentation, four clusters were identified: Cluster 1 with 52 products, Cluster 2 with 12 products, Cluster 3 with 20 products, and Cluster 4 with 53 products. The K-Medoids algorithm demonstrated superior performance compared to K-Means in terms of cluster separation and interpretability, with visualizations that enhance the understanding of customer and product distributions. This research aids the company in enhancing customer satisfaction, optimizing inventory, and increasing profitability.

Comparison of Holt Winter's and SARIMA Methods on the data of the Number of Foreign Tourist Visits in Bali Province Evania, Clara Della; Siregar, Bakti
Journal of Mathematics, Computations and Statistics Vol. 7 No. 2 (2024): Volume 07 Nomor 02 (Oktober 2024)
Publisher : Jurusan Matematika FMIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35580/jmathcos.v7i2.4211

The tourism sector is a critical component of a country’s economy, including in Indonesia, where its impact is felt both nationally and regionally, such as in provinces and cities. This research focuses on Bali Province and aims to conduct a comparative analysis of the Holt-Winter’s and Seasonal Auto-regressive Integrated Moving Average (SARIMA) methods for forecasting foreign tourist arrivals. The analysis centers on two primary entry points: Ngurah-Rai Airport and the seaport. The primary objective is to forecast the number of foreign tourist arrivals from February 2024 to January 2025. The results indicate that the Holt-Winter’s model has a Mean Absolute Percentage Error (MAPE) of 5.2631%, which is lower than the MAPE of 5.8920% for the SARIMA model. Additionally, the Mean Absolute Error (MAE) for the Holt-Winter’s model is 19,149.18, compared to 20,883.20 for the SARIMA model. Consequently, this study concludes that the Holt-Winter’s model provides more accurate predictions and is closer to the actual values than the SARIMA model. Bali, Holt-Winter’s, forecasting, SARIMA, tourism.

ANALYZING THE INFLUENCE OF VARIOUS FACTORS ON GENERATION Z'S BEAUTY PRODUCT PURCHASE BEHAVIOR Natalie, Karen; Siregar, Bakti
Jurnal Lebesgue : Jurnal Ilmiah Pendidikan Matematika, Matematika dan Statistika Vol. 5 No. 3 (2024): Jurnal Lebesgue : Jurnal Ilmiah Pendidikan Matematika, Matematika dan Statistik
Publisher : LPPM Universitas Bina Bangsa

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.46306/lb.v5i3.743

In the digital era, Generation Z predominantly relies on digital platforms to select beauty products, both local and international, significantly impacting their purchasing decisions. This study investigates the influence of social media marketing, brand awareness, quality, price, online customer reviews, and discounts on Generation Z’s purchasing behaviours using multiple linear regression and correlation coefficients, analysed with R Studio and SPSS Statistics 25. A quantitative approach was employed, utilizing a questionnaire to collect data from 213 respondents. The results indicate that for international beauty products, social media marketing, brand awareness, product quality, and discounts are significant predictors of purchase intention (R² = 0.832). Conversely, for local beauty products, the key factors influencing purchase decisions are quality, price, online customer reviews, and discounts (R² = 0.790). These findings highlight that the determinants of Generation Z’s purchasing decisions vary based on the product type. While international products are more influenced by marketing and branding, local products are more affected by price and customer feedback. These insights are crucial for beauty brands aiming to tailor their marketing strategies effectively to engage Generation Z.

Analisis Segmentasi Pelanggan Toko Online Marketplace Berdasarkan Recency, Frequency, Monetary, Time, Satisfaction (Rfmts) Menggunakan Algoritma K-Medoids Clustering Windjaya, Putri Angelina; Siregar, Bakti
Mutiara: Multidiciplinary Scientifict Journal Vol. 2 No. 2 (2024): Multidiciplinary Scientifict Journal
Publisher : Al Makki Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57185/mutiara.v2i2.144

The marketplace is an e-commerce website that applies traditional market concepts and implements them online. Due to the rapid development of the times, a business will seek various ways to maintain its business so that it continues to grow and generate a high income. One way is to build and maintain long-term relationships with customers. Therefore, it is necessary to implement a marketing strategy based on customer relationship management, all of which aims to increase turnover and profit while retaining customers. From these problems, this study applies customer segmentation to detect diversity among customers, so those segments represent potential customers to improve marketing strategies. This segmentation needs to consider the value of the customer's recency, frequency, monetary, time, and satisfaction (RFMTS) variables. Recency is the customer's length of time since the last payment. Frequency is how often customers make transactions. Monetary is the number of transactions made by the customer. Time is the time interval between two consecutive purchases by a customer. Satisfaction is the level of customer satisfaction based on the total rating and number of reviews. This study uses the K-Medoids Clustering algorithm to perform segmentation according to customer characteristics. This study uses primary data from a database of one of the online marketplace stores in Indonesia.

Analisis Klasifikasi Diagnosa Penyakit Diabetes Melitus Berdasarkan Komparasi Algoritma Supervised Learning Siridion, Sherly Taurin; Siregar, Bakti
Mutiara: Multidiciplinary Scientifict Journal Vol. 2 No. 3 (2024): Multidiciplinary Scientifict Journal
Publisher : Al Makki Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57185/mutiara.v2i2.159

Diabetes mellitus is a serious health problem that threatens various age groups, including children, teenagers, and adults. In 2021, the mortality rate due to diabetes mellitus reached alarming levels, making it a global threat, especially in Indonesia, where the number of patients reached 19.5 million. Efforts to address diabetes mellitus include early prediction of the disease's risk in patients, and machine learning approaches have shown potential in this regard. This study employs a quantitative method by utilizing secondary data from the UC Irvine Machine Learning Repository titled "Early Stages Diabetes Risk Prediction". The data was obtained from questionnaires filled out by diabetes patients at Sylhet Diabetes Hospital and validated by healthcare professionals. A total of 520 data samples with 17 attributes were used for analysis. The tested Supervised Learning algorithms include Logistic Regression, K-Nearest Neighbour, Support Vector Machine, Random Forest, Naïve Bayes Classifier, Artificial Neural Network, Decision Tree C4.5, and Gradient Boosting Classifier. The research findings reveal that the Random Forest algorithm achieved the highest accuracy of 98.71% in diagnosing diabetes mellitus. This study significantly contributes to enhancing the understanding of diabetes mellitus and has the potential for further development in finding the best algorithm for early disease prediction. It is hoped that this research will make a significant contribution to the efforts in preventing and managing diabetes mellitus, ultimately improving the quality of life for patients and reducing its impact on the population.

Analisis Segmentasi Pelanggan Toko Online Marketplace Berdasarkan Recency, Frequency, Monetary, Time, Satisfaction (Rfmts) Menggunakan Algoritma K-Medoids Clustering Windjaya, Putri Angelina; Siregar, Bakti
Mutiara: Multidiciplinary Scientifict Journal Vol. 2 No. 2 (2024): Mutiara: Multidiciplinary Scientifict Journal
Publisher : Al Makki Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57185/mutiara.v2i2.144

The marketplace is an e-commerce website that applies traditional market concepts and implements them online. Due to the rapid development of the times, a business will seek various ways to maintain its business so that it continues to grow and generate a high income. One way is to build and maintain long-term relationships with customers. Therefore, it is necessary to implement a marketing strategy based on customer relationship management, all of which aims to increase turnover and profit while retaining customers. From these problems, this study applies customer segmentation to detect diversity among customers, so those segments represent potential customers to improve marketing strategies. This segmentation needs to consider the value of the customer's recency, frequency, monetary, time, and satisfaction (RFMTS) variables. Recency is the customer's length of time since the last payment. Frequency is how often customers make transactions. Monetary is the number of transactions made by the customer. Time is the time interval between two consecutive purchases by a customer. Satisfaction is the level of customer satisfaction based on the total rating and number of reviews. This study uses the K-Medoids Clustering algorithm to perform segmentation according to customer characteristics. This study uses primary data from a database of one of the online marketplace stores in Indonesia.

Analisis Klasifikasi Diagnosa Penyakit Diabetes Melitus Berdasarkan Komparasi Algoritma Supervised Learning Siridion, Sherly Taurin; Siregar, Bakti
Mutiara: Multidiciplinary Scientifict Journal Vol. 2 No. 3 (2024): Mutiara: Multidiciplinary Scientifict Journal
Publisher : Al Makki Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57185/mutiara.v2i2.159

Diabetes mellitus is a serious health problem that threatens various age groups, including children, teenagers, and adults. In 2021, the mortality rate due to diabetes mellitus reached alarming levels, making it a global threat, especially in Indonesia, where the number of patients reached 19.5 million. Efforts to address diabetes mellitus include early prediction of the disease's risk in patients, and machine learning approaches have shown potential in this regard. This study employs a quantitative method by utilizing secondary data from the UC Irvine Machine Learning Repository titled "Early Stages Diabetes Risk Prediction". The data was obtained from questionnaires filled out by diabetes patients at Sylhet Diabetes Hospital and validated by healthcare professionals. A total of 520 data samples with 17 attributes were used for analysis. The tested Supervised Learning algorithms include Logistic Regression, K-Nearest Neighbour, Support Vector Machine, Random Forest, Naïve Bayes Classifier, Artificial Neural Network, Decision Tree C4.5, and Gradient Boosting Classifier. The research findings reveal that the Random Forest algorithm achieved the highest accuracy of 98.71% in diagnosing diabetes mellitus. This study significantly contributes to enhancing the understanding of diabetes mellitus and has the potential for further development in finding the best algorithm for early disease prediction. It is hoped that this research will make a significant contribution to the efforts in preventing and managing diabetes mellitus, ultimately improving the quality of life for patients and reducing its impact on the population.

Title

Found 26 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 26 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 26 Documents
Search