A stock is a high-risk, high-return investment product. Prediction is one way to minimize risk by estimating future prices based on past data. There are limitations to solving the stock prediction problem from previous research: limited stock data, practical aspects of application, and less than optimal stock price prediction results. The main objective of this study is to improve the prediction performance by formulating and developing the stock price prediction framework. Furthermore, the research provides a stock price prediction framework that can produce better prediction results than the previous study with fast computation time. The proposed framework deals with data generation, pre-processing and model prediction. In further, the proposed framework includes two prediction methods for predicting stock closing prices: stored model prediction and current model prediction. This study uses an artificial neural network with Rectified Linear Units as an activation function and Adam Optimizer to predict stock prices. The model we have built for each forecasting method shows a better MAPE value than the model in previous studies. Previous research showed that the lowest MAPE was 1.38% for TLKM shares and 0.81% for BBRI. Our proposed framework based on the stored model prediction method shows a MAPE value of 0.67% for TLKM shares and 0.42% for BBRI. While the current model prediction method shows a MAPE value of 0.69% for TLKM shares and 0.89% for BBRI. Furthermore, the stored model prediction method takes 1.0 seconds to process a single prediction request, while the current model prediction takes 220 seconds.
股票是一种高风险、高回报的投资产品。预测是一种通过根据过去的数据估计未来价格来降低风险的方法。以往的研究对股票预测问题的解决存在着局限性:股票数据有限，应用的实际方面，股票价格预测结果不是最优的。本研究的主要目的是通过制定和发展股票价格预测框架来提高预测绩效。此外，该研究还提供了一种计算速度快、预测结果优于以往研究的股票价格预测框架。该框架涉及数据生成、预处理和模型预测。此外，本文提出的框架还包括两种预测股票收盘价的方法:存储模型预测和当前模型预测。本研究以整流线性单元(Rectified Linear Units)为激活函数的人工神经网络和Adam Optimizer来预测股票价格。我们为每种预测方法所建立的模型都比以往研究的模型显示出更好的MAPE值。先前的研究表明，TLKM股票的MAPE最低为1.38%，bri股票的MAPE最低为0.81%。我们提出的基于存储模型预测方法的框架显示，TLKM份额的MAPE值为0.67%，bri的MAPE值为0.42%。而目前的模型预测方法显示，TLKM份额的MAPE值为0.69%，bri的MAPE值为0.89%。此外，存储的模型预测方法处理单个预测请求需要1.0秒，而当前模型预测需要220秒。
{"title":"Rectified Linear Units and Adaptive Moment Estimation Optimizer on ANN with Saved Model Prediction to Improve The Stock Price Prediction Framework Performance","authors":"Sekhudin Sekhudin, Yuli Purwati, Fandy Setyo Utomo, Mohd Sanusi Azmi, Pungkas Subarkah","doi":"10.33096/ilkom.v15i2.1586.271-282","DOIUrl":"https://doi.org/10.33096/ilkom.v15i2.1586.271-282","url":null,"abstract":"A stock is a high-risk, high-return investment product. Prediction is one way to minimize risk by estimating future prices based on past data. There are limitations to solving the stock prediction problem from previous research: limited stock data, practical aspects of application, and less than optimal stock price prediction results. The main objective of this study is to improve the prediction performance by formulating and developing the stock price prediction framework. Furthermore, the research provides a stock price prediction framework that can produce better prediction results than the previous study with fast computation time. The proposed framework deals with data generation, pre-processing and model prediction. In further, the proposed framework includes two prediction methods for predicting stock closing prices: stored model prediction and current model prediction. This study uses an artificial neural network with Rectified Linear Units as an activation function and Adam Optimizer to predict stock prices. The model we have built for each forecasting method shows a better MAPE value than the model in previous studies. Previous research showed that the lowest MAPE was 1.38% for TLKM shares and 0.81% for BBRI. Our proposed framework based on the stored model prediction method shows a MAPE value of 0.67% for TLKM shares and 0.42% for BBRI. While the current model prediction method shows a MAPE value of 0.69% for TLKM shares and 0.89% for BBRI. Furthermore, the stored model prediction method takes 1.0 seconds to process a single prediction request, while the current model prediction takes 220 seconds.","PeriodicalId":33690,"journal":{"name":"Ilkom Jurnal Ilmiah","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135021662","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
FF Jaya Glass is a shop that supplies and installs 3 mm to 12 mm glass. The store obtained glass from suppliers to be processed in shape and size according to customers’ order. After completing the customer's order, the shop worker will install the glass at the requested location. Unfortunately, currently stores do not utilize sales data to predict sales either manually or by utilizing technology. As a result, the store cannot predict when the number of glass orders will increase or decrease. In addition, errors often occur when ordering glass for the next period. As a result, stores often run out of glass supplies due to the large number of glass orders so that the achievement of profits is not optimal. This study aims to identify sales variables in glass sales data and build a general regression neural network model as a data mining method. In addition, this study aims to iterate to find the best value in the sales data training process, design and create applications according to user needs, and conduct system validation tests. The general regression neural network method is used to predict sales. The results of this study indicate that the application of general regression neural networks can be used to predict sales. This will make it easier for the store to provide glass supplies in the coming months with an accuracy of 98.1%.
{"title":"Application of General Regression Neural Network Algorithm in Data Mining for Predicting Glass Sales and Inventory Quantity","authors":"Suryani Suryani, Indo Intan, Farhan Mochtar Yunus, Adammas Haris, Faizal Faizal, Nurdiansah Nurdiansah","doi":"10.33096/ilkom.v15i2.1562.229-239","DOIUrl":"https://doi.org/10.33096/ilkom.v15i2.1562.229-239","url":null,"abstract":"FF Jaya Glass is a shop that supplies and installs 3 mm to 12 mm glass. The store obtained glass from suppliers to be processed in shape and size according to customers’ order. After completing the customer's order, the shop worker will install the glass at the requested location. Unfortunately, currently stores do not utilize sales data to predict sales either manually or by utilizing technology. As a result, the store cannot predict when the number of glass orders will increase or decrease. In addition, errors often occur when ordering glass for the next period. As a result, stores often run out of glass supplies due to the large number of glass orders so that the achievement of profits is not optimal. This study aims to identify sales variables in glass sales data and build a general regression neural network model as a data mining method. In addition, this study aims to iterate to find the best value in the sales data training process, design and create applications according to user needs, and conduct system validation tests. The general regression neural network method is used to predict sales. The results of this study indicate that the application of general regression neural networks can be used to predict sales. This will make it easier for the store to provide glass supplies in the coming months with an accuracy of 98.1%.","PeriodicalId":33690,"journal":{"name":"Ilkom Jurnal Ilmiah","volume":"95 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135023335","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-16DOI: 10.33096/ilkom.v15i2.1539.250-261
Anton Yudhana, Herman Herman, Suwanti Suwanti, Muhammad Kunta Biddinika
Over five years, the implementation of the library information system at IKIP Muhammadiyah Maumere faced a challenge, frequent errors during data input that hindered users from fully utilizing the system. These issues not only affected users’ interest but also highlighted the significance of the human factor in shaping the quality of an information system. To make full use of the system, it was crucial to identify and address the problems associated with it. This research delved into the experiences of 242 library information system users, including lecturers, students, and librarians, by using the PIECES method. The goal was to analyze users’ satisfaction and uncover any underlying issues within the system. The results of the PIECES analysis revealed average satisfaction scores, showcasing users' contentment with the system's performance (3.77), information (3.79), economy (3.80), control (3.77), efficiency (3.77), and service (3.89). These findings suggest that the library information system has been meeting users' expectations. However, a significant problem emerged in the performance variable, particularly in the system stability. Additionally, issues related to data compatibility, duplication in storage, and users’ authority management, access control, and system errors were observed in the information and control variables. Based on these identified challenges, recommendations for system improvement were made by targeting low satisfaction levels. Proposed solutions involve enhancing data management, storage practices, user access control, and reducing the risk of system errors, ensuring more efficient and reliable library information system
{"title":"Evaluating The Application of Library Information System Technology using the PIECES Method in Remote Areas","authors":"Anton Yudhana, Herman Herman, Suwanti Suwanti, Muhammad Kunta Biddinika","doi":"10.33096/ilkom.v15i2.1539.250-261","DOIUrl":"https://doi.org/10.33096/ilkom.v15i2.1539.250-261","url":null,"abstract":"Over five years, the implementation of the library information system at IKIP Muhammadiyah Maumere faced a challenge, frequent errors during data input that hindered users from fully utilizing the system. These issues not only affected users’ interest but also highlighted the significance of the human factor in shaping the quality of an information system. To make full use of the system, it was crucial to identify and address the problems associated with it. This research delved into the experiences of 242 library information system users, including lecturers, students, and librarians, by using the PIECES method. The goal was to analyze users’ satisfaction and uncover any underlying issues within the system. The results of the PIECES analysis revealed average satisfaction scores, showcasing users' contentment with the system's performance (3.77), information (3.79), economy (3.80), control (3.77), efficiency (3.77), and service (3.89). These findings suggest that the library information system has been meeting users' expectations. However, a significant problem emerged in the performance variable, particularly in the system stability. Additionally, issues related to data compatibility, duplication in storage, and users’ authority management, access control, and system errors were observed in the information and control variables. Based on these identified challenges, recommendations for system improvement were made by targeting low satisfaction levels. Proposed solutions involve enhancing data management, storage practices, user access control, and reducing the risk of system errors, ensuring more efficient and reliable library information system","PeriodicalId":33690,"journal":{"name":"Ilkom Jurnal Ilmiah","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135023338","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-16DOI: 10.33096/ilkom.v15i2.1758.335-343
Aditia Yudhistira, Imas Sukaesih Sitanggang, Hari Agung Adrianto
The SOLAP system for Indonesian Agricultural Commodities is a successful development based on previous studies. Agricultural commodity data are managed in a data warehouse with a galactic schema, which has 7 fact tables, namely cut flower horticulture, ornamental plant horticulture, horticulture, food crops, plantation, livestock population, and livestock production, as well as 3 dimensional tables, namely location, time, and commodity. The results of SOLAP operations on the system can be visualized in the form of crosstabs, graphs and maps. The system uses a web platform so that it can be accessed by the public. However, the SOLAP system cannot update data in real time. This study aims to develop a data warehouse for Indonesian Agricultural Commodities SOLAP in real time by creating a scraping system. This study has succeeded in developing a data warehouse in real time on the indonesian agricultural commodity SOLAP system by creating a real time scraping system that is applied to the SOLAP server and has succeeded in making the ETL process run in real time on the SOLAP server and optimizing polygon-based spatial data visualization using the Douglas-Peucker. This study has also carried out functional testing of OLAP features and functions on the Indonesian Agricultural Commodity SOLAP system using the black box testing method. The results of this study provide accurate and real-time data on the SOLAP of Indonesian Agricultural Commodities, with the results of SOLAP feature testing achieving 100 percent pass and the data conformity test results of OLAP function as expected. In addition, the results of this study make it possible to automatically update the data according to a predetermined schedule to provide real-time information.
{"title":"Development ETL (Extract, Transform and Load) Module in Indonesian Agricultural Commodities OLAP System","authors":"Aditia Yudhistira, Imas Sukaesih Sitanggang, Hari Agung Adrianto","doi":"10.33096/ilkom.v15i2.1758.335-343","DOIUrl":"https://doi.org/10.33096/ilkom.v15i2.1758.335-343","url":null,"abstract":"The SOLAP system for Indonesian Agricultural Commodities is a successful development based on previous studies. Agricultural commodity data are managed in a data warehouse with a galactic schema, which has 7 fact tables, namely cut flower horticulture, ornamental plant horticulture, horticulture, food crops, plantation, livestock population, and livestock production, as well as 3 dimensional tables, namely location, time, and commodity. The results of SOLAP operations on the system can be visualized in the form of crosstabs, graphs and maps. The system uses a web platform so that it can be accessed by the public. However, the SOLAP system cannot update data in real time. This study aims to develop a data warehouse for Indonesian Agricultural Commodities SOLAP in real time by creating a scraping system. This study has succeeded in developing a data warehouse in real time on the indonesian agricultural commodity SOLAP system by creating a real time scraping system that is applied to the SOLAP server and has succeeded in making the ETL process run in real time on the SOLAP server and optimizing polygon-based spatial data visualization using the Douglas-Peucker. This study has also carried out functional testing of OLAP features and functions on the Indonesian Agricultural Commodity SOLAP system using the black box testing method. The results of this study provide accurate and real-time data on the SOLAP of Indonesian Agricultural Commodities, with the results of SOLAP feature testing achieving 100 percent pass and the data conformity test results of OLAP function as expected. In addition, the results of this study make it possible to automatically update the data according to a predetermined schedule to provide real-time information.","PeriodicalId":33690,"journal":{"name":"Ilkom Jurnal Ilmiah","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135021660","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-16DOI: 10.33096/ilkom.v15i2.1634.317-325
Diana Tri Susetianingtias, Eka Patriya, Rini Arianty
Baby Fish counting must be counted accurately so it will not cause any loss, especially for fish seeds or fingerlings that have a small size. Generally, people still use conventional counting methods that produce low accuracy values. This research will make a Nila Baby Fish fingerlings counter program using the YOLOv3 algorithm and Blobb detection technique. The annotation data process will use LabelImg, and the dataset training will use Google COLABoratory with the Darknet framework in an online environment. Images that will predict in this program will be called and detected with an object detector. The object with a confidence score of more than 0.3 will be converted into a blob. The blob value will be forwarded to the output layer for scaling the bounding box objects. The output of this program is the predicted image, blob value, prediction time, and the number of Nila seeds. The model performance is evaluated using a confusion matrix and got a 98.87% for accuracy score.
{"title":"Combination of YOLOv3 Algorithm and Blob Detection Technique in Calculating Nile Tilapia Seeds","authors":"Diana Tri Susetianingtias, Eka Patriya, Rini Arianty","doi":"10.33096/ilkom.v15i2.1634.317-325","DOIUrl":"https://doi.org/10.33096/ilkom.v15i2.1634.317-325","url":null,"abstract":"Baby Fish counting must be counted accurately so it will not cause any loss, especially for fish seeds or fingerlings that have a small size. Generally, people still use conventional counting methods that produce low accuracy values. This research will make a Nila Baby Fish fingerlings counter program using the YOLOv3 algorithm and Blobb detection technique. The annotation data process will use LabelImg, and the dataset training will use Google COLABoratory with the Darknet framework in an online environment. Images that will predict in this program will be called and detected with an object detector. The object with a confidence score of more than 0.3 will be converted into a blob. The blob value will be forwarded to the output layer for scaling the bounding box objects. The output of this program is the predicted image, blob value, prediction time, and the number of Nila seeds. The model performance is evaluated using a confusion matrix and got a 98.87% for accuracy score.","PeriodicalId":33690,"journal":{"name":"Ilkom Jurnal Ilmiah","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135021663","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-16DOI: 10.33096/ilkom.v15i2.1349.326-334
Alders Paliling, Muh Nurtanzis Sutoyo
Tuition payments at State Universities (PTN) use a Single Tuition Fee (UKT) payment system. It has been implemented to make it easier for students to pay their tuition. The UKT system is divided into several groups starting from the UKT group I to VIII. Universitas Sembilanbelas November (USN) Kolaka is a state university and the university should determine the amount of tuition fees for each student according to the UKT system. In determining the UKT group for each student, several variables were used to make it easier to group student into their UKT groups. However, the large number of students, a number of variables and the limited time to determine the amount of UKT for each student become an issue, so a method was needed to help USN Kolaka in grouping UKT for each student. One thing that can be done was to use the MADM model Yager and k-NN in order to make it easier to group UKT students. The results of the study showed that the use of the MADM Model Yager and k-NN could determine the UKT group of the students, and the results obtained for the UKT group I were 63 people (21.95%), the UKT group II were 72 people (25.09%), the UKT group III were 120 people (41.81%), UKT group IV were 7 people (2.44%), and UKT group V were 25 people (8.71%).
{"title":"Combination of the MADM Model Yager and k-NN to Group Single Tuition Payments","authors":"Alders Paliling, Muh Nurtanzis Sutoyo","doi":"10.33096/ilkom.v15i2.1349.326-334","DOIUrl":"https://doi.org/10.33096/ilkom.v15i2.1349.326-334","url":null,"abstract":"Tuition payments at State Universities (PTN) use a Single Tuition Fee (UKT) payment system. It has been implemented to make it easier for students to pay their tuition. The UKT system is divided into several groups starting from the UKT group I to VIII. Universitas Sembilanbelas November (USN) Kolaka is a state university and the university should determine the amount of tuition fees for each student according to the UKT system. In determining the UKT group for each student, several variables were used to make it easier to group student into their UKT groups. However, the large number of students, a number of variables and the limited time to determine the amount of UKT for each student become an issue, so a method was needed to help USN Kolaka in grouping UKT for each student. One thing that can be done was to use the MADM model Yager and k-NN in order to make it easier to group UKT students. The results of the study showed that the use of the MADM Model Yager and k-NN could determine the UKT group of the students, and the results obtained for the UKT group I were 63 people (21.95%), the UKT group II were 72 people (25.09%), the UKT group III were 120 people (41.81%), UKT group IV were 7 people (2.44%), and UKT group V were 25 people (8.71%).","PeriodicalId":33690,"journal":{"name":"Ilkom Jurnal Ilmiah","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135023332","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-16DOI: 10.33096/ilkom.v15i2.1544.262-270
Dadang Priyanto, Ahmad Robbiul Iman, Deny Jollyta
Indonesia, with its very large population, is a potential market for drugs trafficking. Hence, seriousness is needed in cracking down or preventing drug trafficking. Narcotics are substances or drugs that can cause dependence or addicted and other negative impacts on users. The problem is that drug users do not realize and even ignore diseases caused by drug addiction. The diseases can be life-threatening for users, such as inflammation of the liver, heart disease, hypertension, stroke, and others. The prevalence rate of drug abuse in West Nusa Tenggara (NTB) is included in the high category, reaching 292 cases or around 37.24% cases. This study aimed to create an application that can classify various diseases of drug users using the naïve bayes and KNN methods. The results of this study indicated that there was a very close relationship between drug users and various deadly diseases. The prediction results showed that the naive bayes method provided a prediction accuracy of 94.5% while the KNN showed a prediction accuracy of 92.5%. This shows that the naive bayes method provides better predictive performance than the KNN in the data set of drug addicts in NTB.
{"title":"Naïve Bayes and K-Nearest Neighbor Algorithm Approach in Data Mining Classification of Drugs Addictive Diseases","authors":"Dadang Priyanto, Ahmad Robbiul Iman, Deny Jollyta","doi":"10.33096/ilkom.v15i2.1544.262-270","DOIUrl":"https://doi.org/10.33096/ilkom.v15i2.1544.262-270","url":null,"abstract":"Indonesia, with its very large population, is a potential market for drugs trafficking. Hence, seriousness is needed in cracking down or preventing drug trafficking. Narcotics are substances or drugs that can cause dependence or addicted and other negative impacts on users. The problem is that drug users do not realize and even ignore diseases caused by drug addiction. The diseases can be life-threatening for users, such as inflammation of the liver, heart disease, hypertension, stroke, and others. The prevalence rate of drug abuse in West Nusa Tenggara (NTB) is included in the high category, reaching 292 cases or around 37.24% cases. This study aimed to create an application that can classify various diseases of drug users using the naïve bayes and KNN methods. The results of this study indicated that there was a very close relationship between drug users and various deadly diseases. The prediction results showed that the naive bayes method provided a prediction accuracy of 94.5% while the KNN showed a prediction accuracy of 92.5%. This shows that the naive bayes method provides better predictive performance than the KNN in the data set of drug addicts in NTB.","PeriodicalId":33690,"journal":{"name":"Ilkom Jurnal Ilmiah","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135023334","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-16DOI: 10.33096/ilkom.v15i2.1632.344-352
Arif Munandar, Wiga Maulana Baihaqi, Ade Nurhopipah
Cardiovascular disease is one of the deadliest diseases, claiming around 17 million lives worldwide each year. According to data from the World Health Organization (WHO), more than four out of five deaths from cardiovascular disease are caused by heart attacks and strokes, and one-third of these deaths occur prematurely in people under the age of 70. Machine learning approaches can be used to detect the disease. This research aims to improve the prediction model of cardiovascular heart failure patient survival using C4.5, KNN, Logistic Regression algorithms, and the ensemble learning method of Voting Classifier. Based on the testing results, each model showed a significant increase in accuracy in the 70:30 ratio. Logistic Regression and C4.5 achieved the same accuracy, 89.47%, KNN obtained 91.23%, and Voting Classifier experienced a considerable improvement, reaching 94.74%. In testing with ratios of 90:10, 80:20, and 70:30, KNN demonstrated high accuracy but had significant overfitting, with a difference of 7-9% between training and testing accuracy scores in the 90:10 and 80:20 ratios. On the other hand, Voting Classifier showed stable performance in the 70:30 ratio, with an accuracy difference between training and testing scores below 1%. The conclusion of this research is that the Voting Classifier can assist the performance improvement of algorithms for classifying the survival expectancy of cardiovascular heart failure patients into 'Survived' or 'Deceased', compared to Logistic Regression, KNN, and C4.5.
{"title":"A Soft Voting Ensemble Classifier to Improve Survival Rate Predictions of Cardiovascular Heart Failure Patients","authors":"Arif Munandar, Wiga Maulana Baihaqi, Ade Nurhopipah","doi":"10.33096/ilkom.v15i2.1632.344-352","DOIUrl":"https://doi.org/10.33096/ilkom.v15i2.1632.344-352","url":null,"abstract":"Cardiovascular disease is one of the deadliest diseases, claiming around 17 million lives worldwide each year. According to data from the World Health Organization (WHO), more than four out of five deaths from cardiovascular disease are caused by heart attacks and strokes, and one-third of these deaths occur prematurely in people under the age of 70. Machine learning approaches can be used to detect the disease. This research aims to improve the prediction model of cardiovascular heart failure patient survival using C4.5, KNN, Logistic Regression algorithms, and the ensemble learning method of Voting Classifier. Based on the testing results, each model showed a significant increase in accuracy in the 70:30 ratio. Logistic Regression and C4.5 achieved the same accuracy, 89.47%, KNN obtained 91.23%, and Voting Classifier experienced a considerable improvement, reaching 94.74%. In testing with ratios of 90:10, 80:20, and 70:30, KNN demonstrated high accuracy but had significant overfitting, with a difference of 7-9% between training and testing accuracy scores in the 90:10 and 80:20 ratios. On the other hand, Voting Classifier showed stable performance in the 70:30 ratio, with an accuracy difference between training and testing scores below 1%. The conclusion of this research is that the Voting Classifier can assist the performance improvement of algorithms for classifying the survival expectancy of cardiovascular heart failure patients into 'Survived' or 'Deceased', compared to Logistic Regression, KNN, and C4.5.","PeriodicalId":33690,"journal":{"name":"Ilkom Jurnal Ilmiah","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135023337","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-16DOI: 10.33096/ilkom.v15i2.1686.390-397
Huzain Azis, Purnawansyah Purnawansyah, Nirwana Nirwana, Felix Andika Dwiyanto
Support Vector Regression (SVR) is a supervised learning algorithm to predict continuous variable values. The basic goal of the SVR algorithm is to find the most suitable decision line. SVR has been successfully applied to several issues in time series prediction. In this research, SVR is used to predict the price of staple commodity, which are constantly changing in price at any time due to several factors making it difficult for the public to get groceries that are easy to reach. National staple commodity data consisting of 17 commodities, including shallots, honan garlic, kating garlic, medium rice, premium rice, red cayenne peppers, curly red chilies, red chili peppers, meat of broiler chicken, beef hamstrings, granulated sugar, imported soybeans, bulk cooking oil, premium packaged cooking oil, simple packaged cooking oil, broiler chicken eggs, and wheat flour. With a data set for the last 3 years, including from January 1, 2020, to December 31, 2022. There are 3 variables in the data set, namely commodity, date, and price. This research divides the entire dataset into 80% training and 20% testing data. The results of this research show that SVR using the RBF kernel produces good forecasting accuracy for all datasets with an average Mean Square Error (MSE) training data of 6,005 while data testing is 6,062, Mean Absolute Deviation (MAD) of training data is 6,730 while data testing is 6.6831, Mean Absolute Percentage Error (MAPE) training data is 0.0148 while data testing is 0.0147, and Root Mean Squared Error (RMSE) training data is 7.772 while data testing is 7.746.
{"title":"The Support Vector Regression Method Performance Analysis in Predicting National Staple Commodity Prices","authors":"Huzain Azis, Purnawansyah Purnawansyah, Nirwana Nirwana, Felix Andika Dwiyanto","doi":"10.33096/ilkom.v15i2.1686.390-397","DOIUrl":"https://doi.org/10.33096/ilkom.v15i2.1686.390-397","url":null,"abstract":"Support Vector Regression (SVR) is a supervised learning algorithm to predict continuous variable values. The basic goal of the SVR algorithm is to find the most suitable decision line. SVR has been successfully applied to several issues in time series prediction. In this research, SVR is used to predict the price of staple commodity, which are constantly changing in price at any time due to several factors making it difficult for the public to get groceries that are easy to reach. National staple commodity data consisting of 17 commodities, including shallots, honan garlic, kating garlic, medium rice, premium rice, red cayenne peppers, curly red chilies, red chili peppers, meat of broiler chicken, beef hamstrings, granulated sugar, imported soybeans, bulk cooking oil, premium packaged cooking oil, simple packaged cooking oil, broiler chicken eggs, and wheat flour. With a data set for the last 3 years, including from January 1, 2020, to December 31, 2022. There are 3 variables in the data set, namely commodity, date, and price. This research divides the entire dataset into 80% training and 20% testing data. The results of this research show that SVR using the RBF kernel produces good forecasting accuracy for all datasets with an average Mean Square Error (MSE) training data of 6,005 while data testing is 6,062, Mean Absolute Deviation (MAD) of training data is 6,730 while data testing is 6.6831, Mean Absolute Percentage Error (MAPE) training data is 0.0148 while data testing is 0.0147, and Root Mean Squared Error (RMSE) training data is 7.772 while data testing is 7.746.","PeriodicalId":33690,"journal":{"name":"Ilkom Jurnal Ilmiah","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135021655","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Floods are a common disaster in watersheds, and flood control is difficult. However, losses can be reduced by quickly disseminating alert status information. This paper proposes a prototype of a monitoring system that can determine the status of flood alerts in real time and quickly disseminating to the community, allowing people to be better prepared for flood disasters. The system was developed using the RD method and consists of hardware and software development. The hardware comprises several sensor modules to read the discharge, temperature, humidity, and water level and to transmit the readings to the software. The software is divided into two applications: a website application and a Telegram application. The public can find the flood alert status history data from the website and obtain flood alert status warning messages and the latest alert status from Telegram. The results of the tests indicated that the sensors were very accurate, with a MAPE value of less than 10%. The software test also showed that the input and output were according to design. The proposed system can potentially reduce flood losses by providing early warning information to the community. The system is also scalable and adaptable to other watersheds.
{"title":"Design and Build of IoT Based Flood Prone Monitoring System at Semani’s Pump House Drainage System","authors":"'Aisyah 'Aisyah, Aji Ery Burhandenny, Happy Nugroho, Didit Suprihanto","doi":"10.33096/ilkom.v15i2.1581.303-316","DOIUrl":"https://doi.org/10.33096/ilkom.v15i2.1581.303-316","url":null,"abstract":"Floods are a common disaster in watersheds, and flood control is difficult. However, losses can be reduced by quickly disseminating alert status information. This paper proposes a prototype of a monitoring system that can determine the status of flood alerts in real time and quickly disseminating to the community, allowing people to be better prepared for flood disasters. The system was developed using the RD method and consists of hardware and software development. The hardware comprises several sensor modules to read the discharge, temperature, humidity, and water level and to transmit the readings to the software. The software is divided into two applications: a website application and a Telegram application. The public can find the flood alert status history data from the website and obtain flood alert status warning messages and the latest alert status from Telegram. The results of the tests indicated that the sensors were very accurate, with a MAPE value of less than 10%. The software test also showed that the input and output were according to design. The proposed system can potentially reduce flood losses by providing early warning information to the community. The system is also scalable and adaptable to other watersheds.","PeriodicalId":33690,"journal":{"name":"Ilkom Jurnal Ilmiah","volume":"373 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135021661","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：