Pub Date : 2023-04-07DOI: 10.33096/ilkom.v15i1.1481.21-31
Mutammimul Ula, M. Ula, Desvina Yulisda, S. Susanti
In this research, the clustering of food prone areas in Aceh Utama is based on the Index Ketahanan Pangan (IKP) indicators compiled by Badan Ketahanan Pangan (BKP) using Fuzzy C-Means (FCM) and Borda algorithms. The fuzzy C-Means algorithm was used to classify food-prone areas with three clusters: very prone, moderately prone, and prone. The Borda algorithm was used to choose the most prone area from very prone clusters, which are considered urgently to be followed up by decision-makers. Based on the research results, it was found that in the aspect of food availability, four sub-districts are moderately prone, 10 are prone, and 13 are very prone. Regarding food affordability, it found that 12 sub-districts are moderately prone, seven are prone, and eight are very prone. Regarding food utilization, one sub-district is moderately prone, three are prone, and 23 are very prone. The results of voting using the Borda algorithm in very prone clusters are obtained Sawang District from the aspect of food availability, Syamtalira Aron District from the aspect of food affordability, and Lapang District from the aspect of food utilization. The clustering system is built based on the web using the PHP programming language.
{"title":"Fuzzy C-Means with Borda Algorithm in Cluster Determination System for Food Prone Areas in Aceh Utara","authors":"Mutammimul Ula, M. Ula, Desvina Yulisda, S. Susanti","doi":"10.33096/ilkom.v15i1.1481.21-31","DOIUrl":"https://doi.org/10.33096/ilkom.v15i1.1481.21-31","url":null,"abstract":"In this research, the clustering of food prone areas in Aceh Utama is based on the Index Ketahanan Pangan (IKP) indicators compiled by Badan Ketahanan Pangan (BKP) using Fuzzy C-Means (FCM) and Borda algorithms. The fuzzy C-Means algorithm was used to classify food-prone areas with three clusters: very prone, moderately prone, and prone. The Borda algorithm was used to choose the most prone area from very prone clusters, which are considered urgently to be followed up by decision-makers. Based on the research results, it was found that in the aspect of food availability, four sub-districts are moderately prone, 10 are prone, and 13 are very prone. Regarding food affordability, it found that 12 sub-districts are moderately prone, seven are prone, and eight are very prone. Regarding food utilization, one sub-district is moderately prone, three are prone, and 23 are very prone. The results of voting using the Borda algorithm in very prone clusters are obtained Sawang District from the aspect of food availability, Syamtalira Aron District from the aspect of food affordability, and Lapang District from the aspect of food utilization. The clustering system is built based on the web using the PHP programming language.","PeriodicalId":33690,"journal":{"name":"Ilkom Jurnal Ilmiah","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43555589","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 : 2023-04-07DOI: 10.33096/ilkom.v15i1.1337.175-185
I. Irawanto, Cynthia Widodo, Atin Hasanah, Prema Adhitya Dharma Kusumah, Kusirini Kusrini, Kusnawi Kusnawi
so the preprocessing process must be carried out as was done in research [5], which retrieved Twitter data on the theme of COVID 2019. Moreover, weighting must be applied to tasks that must be completed prior to classification. This study uses VADER or commonly known as the lexicon. [6] It uses a lexicon that combines lexical dictionary features as a polarity assessment. Sentiment scores of 5 additional criteria, namely exclamation marks, large alphabet, level of word order, polarity shift due to the term "but," and using the tri-gram feature to study negation [7]. Once the text has been labeled, we will classify it using the sentiment analysis. The Nave Bayes technique, Random Forest, and SVM are some reliable classifications that have been demonstrated in numerous research (Support Vector Machine). A popular algorithm that is frequently employed by researchers is Naïve Bayes. The following researchers have used the Naive Bayes method for sentiment analysis research: : [8] analyzing the online store JD.ID, [9] regarding awareness of procedures to prevent COVID 2019. Random Forest is rarely implemented in research on sentiment analysis, although it has recently been investigated to gauge its accuracy. It is used by a number of researchers, including [10], who achieves an accuracy of about 0.829. Moreover, the SVM (Support Vector Machine) approach, whose accuracy is 85%, is also being investigated in sentiment analysis study by [11]. Hence, researchers want to compare the values of the 3 methods namely Naïve Bayes, Random Forest and SVM (Support Vector Machine) to find out the difference in accuracy of the three when using the same data. As for the accuracy will be calculated using the calculation on the confusion matrix. In addition, the researcher also wants to compare the results of classifying sentiment statements which are divided into positive, negative and neutral sentiments.
{"title":"Sentiment Analysis and Classification of Forest Fires in Indonesia","authors":"I. Irawanto, Cynthia Widodo, Atin Hasanah, Prema Adhitya Dharma Kusumah, Kusirini Kusrini, Kusnawi Kusnawi","doi":"10.33096/ilkom.v15i1.1337.175-185","DOIUrl":"https://doi.org/10.33096/ilkom.v15i1.1337.175-185","url":null,"abstract":"so the preprocessing process must be carried out as was done in research [5], which retrieved Twitter data on the theme of COVID 2019. Moreover, weighting must be applied to tasks that must be completed prior to classification. This study uses VADER or commonly known as the lexicon. [6] It uses a lexicon that combines lexical dictionary features as a polarity assessment. Sentiment scores of 5 additional criteria, namely exclamation marks, large alphabet, level of word order, polarity shift due to the term \"but,\" and using the tri-gram feature to study negation [7]. Once the text has been labeled, we will classify it using the sentiment analysis. The Nave Bayes technique, Random Forest, and SVM are some reliable classifications that have been demonstrated in numerous research (Support Vector Machine). A popular algorithm that is frequently employed by researchers is Naïve Bayes. The following researchers have used the Naive Bayes method for sentiment analysis research: : [8] analyzing the online store JD.ID, [9] regarding awareness of procedures to prevent COVID 2019. Random Forest is rarely implemented in research on sentiment analysis, although it has recently been investigated to gauge its accuracy. It is used by a number of researchers, including [10], who achieves an accuracy of about 0.829. Moreover, the SVM (Support Vector Machine) approach, whose accuracy is 85%, is also being investigated in sentiment analysis study by [11]. Hence, researchers want to compare the values of the 3 methods namely Naïve Bayes, Random Forest and SVM (Support Vector Machine) to find out the difference in accuracy of the three when using the same data. As for the accuracy will be calculated using the calculation on the confusion matrix. In addition, the researcher also wants to compare the results of classifying sentiment statements which are divided into positive, negative and neutral sentiments.","PeriodicalId":33690,"journal":{"name":"Ilkom Jurnal Ilmiah","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42018401","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 : 2023-04-07DOI: 10.33096/ilkom.v15i1.1541.45-63
Yudha Islami Sulistya, Elsi Titasari Br Bangun, Dyah Aruming Tyas
This study provides a review of CNN's ensemble learning method for transfer learning by highlighting sections such as review studies, datasets, pre-trained models, transfer learning, ensemble learning, and performance. The results indicate that the trend of ensemble learning, transfer learning ensemble, and transfer learning is growing every year. In 2022, there will be 35 papers reviewed related to this topic in this study. Some datasets contain apparent information starting from the dataset name, total data points, dataset splitting, target dataset availability, and type classification. ResNet-50, VGG-16, InceptionV3, and VGG-19 are used in most papers as pre-trained models and transfer learning processes. 50 (90.1%) papers use ensemble learning, and 5 (9.1%) do without ensemble learning. The reviewed paper summarizes several performance measurements, including accuracy, precision, recall, f1-score, sensitivity, specificity, training accuracy, validation accuracy, test accuracy, training losses, validation losses, test losses, training time, and AUC, DSC. In the last section, 49 papers produce the best model performance using the proposed model, and 6 other papers use DenseNet, DeQueezeNet, Extended Yager Model, InceptionV3, and ResNet-152.
{"title":"CNN Ensemble Learning Method for Transfer learning: A Review","authors":"Yudha Islami Sulistya, Elsi Titasari Br Bangun, Dyah Aruming Tyas","doi":"10.33096/ilkom.v15i1.1541.45-63","DOIUrl":"https://doi.org/10.33096/ilkom.v15i1.1541.45-63","url":null,"abstract":"This study provides a review of CNN's ensemble learning method for transfer learning by highlighting sections such as review studies, datasets, pre-trained models, transfer learning, ensemble learning, and performance. The results indicate that the trend of ensemble learning, transfer learning ensemble, and transfer learning is growing every year. In 2022, there will be 35 papers reviewed related to this topic in this study. Some datasets contain apparent information starting from the dataset name, total data points, dataset splitting, target dataset availability, and type classification. ResNet-50, VGG-16, InceptionV3, and VGG-19 are used in most papers as pre-trained models and transfer learning processes. 50 (90.1%) papers use ensemble learning, and 5 (9.1%) do without ensemble learning. The reviewed paper summarizes several performance measurements, including accuracy, precision, recall, f1-score, sensitivity, specificity, training accuracy, validation accuracy, test accuracy, training losses, validation losses, test losses, training time, and AUC, DSC. In the last section, 49 papers produce the best model performance using the proposed model, and 6 other papers use DenseNet, DeQueezeNet, Extended Yager Model, InceptionV3, and ResNet-152.","PeriodicalId":33690,"journal":{"name":"Ilkom Jurnal Ilmiah","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49541906","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 : 2023-04-07DOI: 10.33096/ilkom.v15i1.1381.118-123
Andi Bode, Z. Y. Lamasigi, Ivo Colanus Rally Drajana
Academic services are actions taken by state and private universities to provide convenience for student’s academic activities. During the current covid-19 pandemic, every university remains active in academic activities. This study aimed to apply the K-Nearest Neighbor algorithm in predicting the level of student satisfaction with online lectures at University Ichsan Gorontalo. Our main aim was to obtain quantitative information to measure student satisfaction with online lectures during the pandemic, which should be taken into account when making decisions. K-Nearest Neighbor is a non-parametric Algorithm that can be used for classification and regression, but K-Nearest Neighbor are better if feature selection is applied in selecting features that are not relevant to the model. Feature Selection used in this research is Forward Selection and Backward Elimination. Seeing the results of experiments that have been carried out with the application of the K-nearest Neighbor algorithm and the selection feature, the results of the forecasting can be used for consideration or policy in decision making. The highest level of accuracy in the K-Nearest Neighbor algorithm model used Forward Selection with an accuracy rate of 98.00%. Thus, the experimental results showed that feature selection, namely forward selection, was a better model in the relevant selection variables compared to backward elimination.
{"title":"The K-Nearest Neighbor Algorithm using Forward Selection and Backward Elimination in Predicting the Student’s Satisfaction Level of University Ichsan Gorontalo toward Online Lectures during the COVID-19 Pandemic","authors":"Andi Bode, Z. Y. Lamasigi, Ivo Colanus Rally Drajana","doi":"10.33096/ilkom.v15i1.1381.118-123","DOIUrl":"https://doi.org/10.33096/ilkom.v15i1.1381.118-123","url":null,"abstract":"Academic services are actions taken by state and private universities to provide convenience for student’s academic activities. During the current covid-19 pandemic, every university remains active in academic activities. This study aimed to apply the K-Nearest Neighbor algorithm in predicting the level of student satisfaction with online lectures at University Ichsan Gorontalo. Our main aim was to obtain quantitative information to measure student satisfaction with online lectures during the pandemic, which should be taken into account when making decisions. K-Nearest Neighbor is a non-parametric Algorithm that can be used for classification and regression, but K-Nearest Neighbor are better if feature selection is applied in selecting features that are not relevant to the model. Feature Selection used in this research is Forward Selection and Backward Elimination. Seeing the results of experiments that have been carried out with the application of the K-nearest Neighbor algorithm and the selection feature, the results of the forecasting can be used for consideration or policy in decision making. The highest level of accuracy in the K-Nearest Neighbor algorithm model used Forward Selection with an accuracy rate of 98.00%. Thus, the experimental results showed that feature selection, namely forward selection, was a better model in the relevant selection variables compared to backward elimination.","PeriodicalId":33690,"journal":{"name":"Ilkom Jurnal Ilmiah","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48581955","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 : 2023-04-07DOI: 10.33096/ilkom.v15i1.1526.165-174
T. A. Cahyanto, Rizky Dwi Antoko, Taufiq Timur Warisaji, S. Santosa, Rodianto Rodianto
Current technological advancements make it easier for users to do their work effectively and efficiently, including the use of wireless networks to exchange data via File Transfer Protocol (FTP) and video conferencing services (VCS). A Mobile AdHoc Network (MANET) is a wireless network technology that applies a dynamic set of nodes. Data transmission on the MANET does not require the use of devices such as base stations. Because each node on the MANET can act as a router in determining the direction of the data sent, the number of nodes in the MANET will influence the quality of the data sent. Using the OPNET Modeler simulator, this paper shows how to assess the quality of FTP and VCS based on delay, jitter, and packet loss parameters. The simulation scenario employs five, fifteen, and thirty nodes with low, medium, and high traffic loads, using the Dynamic Source Routing (DSR) protocol. According to the measurement results, the FTP service with the bad category is the packet loss parameter in high traffic loads, which has the highest packet loss value of 56.6 percent with 15 nodes. In contrast, good results for VCS are only produced on the delay parameter. The jitter increases with the number of nodes, and it is 5 in this case. In all scenarios, the packet loss parameter yields poor results, with the highest packet loss value approaching 100%.
{"title":"Analysis of the Dynamic Source Routing Protocol on the Performance of File Transfer Protocol and Video Conference Services in the Mobile AdHoc Network Simulation","authors":"T. A. Cahyanto, Rizky Dwi Antoko, Taufiq Timur Warisaji, S. Santosa, Rodianto Rodianto","doi":"10.33096/ilkom.v15i1.1526.165-174","DOIUrl":"https://doi.org/10.33096/ilkom.v15i1.1526.165-174","url":null,"abstract":"Current technological advancements make it easier for users to do their work effectively and efficiently, including the use of wireless networks to exchange data via File Transfer Protocol (FTP) and video conferencing services (VCS). A Mobile AdHoc Network (MANET) is a wireless network technology that applies a dynamic set of nodes. Data transmission on the MANET does not require the use of devices such as base stations. Because each node on the MANET can act as a router in determining the direction of the data sent, the number of nodes in the MANET will influence the quality of the data sent. Using the OPNET Modeler simulator, this paper shows how to assess the quality of FTP and VCS based on delay, jitter, and packet loss parameters. The simulation scenario employs five, fifteen, and thirty nodes with low, medium, and high traffic loads, using the Dynamic Source Routing (DSR) protocol. According to the measurement results, the FTP service with the bad category is the packet loss parameter in high traffic loads, which has the highest packet loss value of 56.6 percent with 15 nodes. In contrast, good results for VCS are only produced on the delay parameter. The jitter increases with the number of nodes, and it is 5 in this case. In all scenarios, the packet loss parameter yields poor results, with the highest packet loss value approaching 100%.","PeriodicalId":33690,"journal":{"name":"Ilkom Jurnal Ilmiah","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45941129","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 : 2023-04-07DOI: 10.33096/ilkom.v15i1.1357.186-192
Abdul Rachman Manga’, A. N. Handayani, H. Herwanto, R. A. Asmara, Yudha Islami Sulistya, Kasmira Kasmira
Arabic character handwriting is one of the patterns and characteristics of each person's writing. This characteristic makes Arabic writing more challenging if the letter recognition process is based on a dataset of Arabic scripts. This Arabic script has been presented in a dataset totaling 16800, each representing a class of hijaiyah letters starting from alif to yes, consisting of 600 data for each class. The accuracy of the data used can be increased using the ensemble method. By using multiple algorithms at simultaneously, the ensemble technique can raise the level or result of a score in machine learning. This study's primary goal is to evaluate the ensemble method classifier's performance on datasets of handwritten Arabic characters. The classifier uses the ensemble method by applying the proposed soft voting to provide a multiclass classification of three machine learning algorithms, namely, SVM, Random Forest, and Decision Tree for classification. This research process produces an accuracy value for the voting classifier of 0.988 and several
{"title":"Analysis of the Ensemble Method Classifier's Performance on Handwritten Arabic Characters Dataset","authors":"Abdul Rachman Manga’, A. N. Handayani, H. Herwanto, R. A. Asmara, Yudha Islami Sulistya, Kasmira Kasmira","doi":"10.33096/ilkom.v15i1.1357.186-192","DOIUrl":"https://doi.org/10.33096/ilkom.v15i1.1357.186-192","url":null,"abstract":"Arabic character handwriting is one of the patterns and characteristics of each person's writing. This characteristic makes Arabic writing more challenging if the letter recognition process is based on a dataset of Arabic scripts. This Arabic script has been presented in a dataset totaling 16800, each representing a class of hijaiyah letters starting from alif to yes, consisting of 600 data for each class. The accuracy of the data used can be increased using the ensemble method. By using multiple algorithms at simultaneously, the ensemble technique can raise the level or result of a score in machine learning. This study's primary goal is to evaluate the ensemble method classifier's performance on datasets of handwritten Arabic characters. The classifier uses the ensemble method by applying the proposed soft voting to provide a multiclass classification of three machine learning algorithms, namely, SVM, Random Forest, and Decision Tree for classification. This research process produces an accuracy value for the voting classifier of 0.988 and several","PeriodicalId":33690,"journal":{"name":"Ilkom Jurnal Ilmiah","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"69492591","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 : 2023-04-07DOI: 10.33096/ilkom.v15i1.1504.64-71
Muhammad Faisal, Maryam Hasan, Kartika Candra Pelangi
The identification of the maturity level of dragon fruit in this study was divided into two groups of ripeness: the unripe and the ripe. This study aims to classify the maturity level based on dragon fruit images using the feature extraction method, the gray level co-occurrence matrix (GLCM). This research method consists of converting RGB data to grayscale, image normalization, detection of dragon fruit maturity, feature extraction, and identification. Data collection from real data totaled 60 images used in this study consisting of 40 training data and 20 testing data which are RGB image data in JPG format. Each data consists of 2 maturity categories. Training data consists of 20 images of 99% ripe dragon fruit and 20 images of 85%. Meanwhile, the testing data consisted of 10 of 99% ripe dragon fruit images and 10 of 85% ripe dragon fruit images. The image data is processed into a grayscale image which then detects the ripeness of the dragon fruit. After the maturity of the dragon fruit is obtained, segmentation is carried out on the location of the dragon fruit found. Then the feature calculation is performed using the Gray Level Co-Occurrence Matrix (GLCM). The Artificial Neural Network (ANN) algorithm is used for the identification process. The final test results show that the proposed method has been able to detect dragon fruit maturity level with an accuracy of = 9/10* 100% = 90%, calculated using the confusion matrix. Thus, implementing the Gray Level Co-Occurrence Matrix
{"title":"The Implementation of GLCM and ANN Methods to Identify Dragon Fruit Maturity Level","authors":"Muhammad Faisal, Maryam Hasan, Kartika Candra Pelangi","doi":"10.33096/ilkom.v15i1.1504.64-71","DOIUrl":"https://doi.org/10.33096/ilkom.v15i1.1504.64-71","url":null,"abstract":"The identification of the maturity level of dragon fruit in this study was divided into two groups of ripeness: the unripe and the ripe. This study aims to classify the maturity level based on dragon fruit images using the feature extraction method, the gray level co-occurrence matrix (GLCM). This research method consists of converting RGB data to grayscale, image normalization, detection of dragon fruit maturity, feature extraction, and identification. Data collection from real data totaled 60 images used in this study consisting of 40 training data and 20 testing data which are RGB image data in JPG format. Each data consists of 2 maturity categories. Training data consists of 20 images of 99% ripe dragon fruit and 20 images of 85%. Meanwhile, the testing data consisted of 10 of 99% ripe dragon fruit images and 10 of 85% ripe dragon fruit images. The image data is processed into a grayscale image which then detects the ripeness of the dragon fruit. After the maturity of the dragon fruit is obtained, segmentation is carried out on the location of the dragon fruit found. Then the feature calculation is performed using the Gray Level Co-Occurrence Matrix (GLCM). The Artificial Neural Network (ANN) algorithm is used for the identification process. The final test results show that the proposed method has been able to detect dragon fruit maturity level with an accuracy of = 9/10* 100% = 90%, calculated using the confusion matrix. Thus, implementing the Gray Level Co-Occurrence Matrix","PeriodicalId":33690,"journal":{"name":"Ilkom Jurnal Ilmiah","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45910403","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 : 2023-04-07DOI: 10.33096/ilkom.v15i1.1524.32-44
F. Alzami, Fikri Diva Sambasri, Mira Nabila, Rama Aria Megantara, Ahmad Akrom, R. A. Pramunendar, D. P. Prabowo, Puri Sulistiyawati
E-commerce is selling and buying goods through an online or online system. One of the business models in which consumers sell products to other consumers is the Customer to Customer (C2C) business model. One thing that needs to be considered in the business model is knowing the level of customer loyalty. By knowing the level of customer loyalty, the company can provide several different treatments to its customers to maintain good relationships with customers and increase product purchase revenue. In this study, the author wants to segment customers on data in E-commerce companies in Brazil using the K-Means clustering algorithm using the RFM (Recency, Frequency, Monetary) feature and display it in the form of a dashboard using the Streamlit framework. Several stages of research must be carried out. Firstly, taking data from the open public data site (Kaggle), then merging the data to select some data that needs to be used, understanding data by displaying it in graphic form, and conducting data selection to select features/attributes. The step follows the proposed method, performs data preprocessing, creates a model to get the cluster, and finally displays it as a dashboard using Streamlit. Based on the results of the research that has been done, the number of clusters is 4 clusters with the evaluation value of the model using the silhouette score is 0.470.
{"title":"Implementation of RFM Method and K-Means Algorithm for Customer Segmentation in E-Commerce with Streamlit","authors":"F. Alzami, Fikri Diva Sambasri, Mira Nabila, Rama Aria Megantara, Ahmad Akrom, R. A. Pramunendar, D. P. Prabowo, Puri Sulistiyawati","doi":"10.33096/ilkom.v15i1.1524.32-44","DOIUrl":"https://doi.org/10.33096/ilkom.v15i1.1524.32-44","url":null,"abstract":"E-commerce is selling and buying goods through an online or online system. One of the business models in which consumers sell products to other consumers is the Customer to Customer (C2C) business model. One thing that needs to be considered in the business model is knowing the level of customer loyalty. By knowing the level of customer loyalty, the company can provide several different treatments to its customers to maintain good relationships with customers and increase product purchase revenue. In this study, the author wants to segment customers on data in E-commerce companies in Brazil using the K-Means clustering algorithm using the RFM (Recency, Frequency, Monetary) feature and display it in the form of a dashboard using the Streamlit framework. Several stages of research must be carried out. Firstly, taking data from the open public data site (Kaggle), then merging the data to select some data that needs to be used, understanding data by displaying it in graphic form, and conducting data selection to select features/attributes. The step follows the proposed method, performs data preprocessing, creates a model to get the cluster, and finally displays it as a dashboard using Streamlit. Based on the results of the research that has been done, the number of clusters is 4 clusters with the evaluation value of the model using the silhouette score is 0.470.","PeriodicalId":33690,"journal":{"name":"Ilkom Jurnal Ilmiah","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49395471","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 : 2023-04-07DOI: 10.33096/ilkom.v15i1.1426.153-164
Z. Y. Lamasigi, Serwin Serwin, Yusrianto Malago
Gorontalo Province is one of the provinces that have fishery potential and has a large sea area that can be managed to support the economy and development of the province. Gorontalo is also one of the tuna-producing provinces in Indonesia, where tuna is also one of the mainstay fisheries commodities. This study aimed to combine transformation and texture feature extraction methods to improve the identification of the freshness level of tuna. This research used Discrete Cosine Transform as transformation detection and Gray Level Co-Occurrence Matrix as texture feature extraction. To find out the value of the proximity of the training data and image testing of tuna fish, the K-Nearest Neighbor classification method was employed. Then, the Confusion Matrix was used to calculate the accuracy level of the K-Nearest Neighbor classification. This research was carried out with 4 stages of testing, namely at angles of 0 , 45 , 90
{"title":"Identification of the Freshness Level of Tuna based on Discrete Cosine Transform on Feature Extraction of Gray Level Co-Occurrence Matrix using K-Nearest Neighbor","authors":"Z. Y. Lamasigi, Serwin Serwin, Yusrianto Malago","doi":"10.33096/ilkom.v15i1.1426.153-164","DOIUrl":"https://doi.org/10.33096/ilkom.v15i1.1426.153-164","url":null,"abstract":"Gorontalo Province is one of the provinces that have fishery potential and has a large sea area that can be managed to support the economy and development of the province. Gorontalo is also one of the tuna-producing provinces in Indonesia, where tuna is also one of the mainstay fisheries commodities. This study aimed to combine transformation and texture feature extraction methods to improve the identification of the freshness level of tuna. This research used Discrete Cosine Transform as transformation detection and Gray Level Co-Occurrence Matrix as texture feature extraction. To find out the value of the proximity of the training data and image testing of tuna fish, the K-Nearest Neighbor classification method was employed. Then, the Confusion Matrix was used to calculate the accuracy level of the K-Nearest Neighbor classification. This research was carried out with 4 stages of testing, namely at angles of 0 , 45 , 90 ","PeriodicalId":33690,"journal":{"name":"Ilkom Jurnal Ilmiah","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49552638","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 : 2023-04-07DOI: 10.33096/ilkom.v15i1.1532.124-131
Qurrota A’yuna Itsnaini, Mardhiya Hayaty, Andriyan Dwi Putra, N. Jabari
Automatic Text Summarization (ATS) is one of the utilizations of technological sophistication in terms of text processing assisting humans in producing a summary or key points of a document in large quantities. We use Indonesian language as objects because there are few resources in NLP research using Indonesian language. This paper utilized PLTMs (Pre-Trained Language Models) from the transformer architecture, namely T5 (Text-to-Text Transfer Transformer) which has been completed previously with a larger dataset. Evaluation in this study was measured through comparison of the ROUGE (Recall-Oriented Understudy for Gisting Evaluation) calculation results between the reference summary and the model summary. The experiments with the pre-trained t5-base model with fine tuning parameters of 220M for the Indonesian news dataset yielded relatively high ROUGE values, namely ROUGE-1 = 0.68, ROUGE-2 = 0.61, and ROUGE-L = 0.65. The evaluation value worked well, but the resulting model has not achieved satisfactory results because in terms of abstraction, the model did not work optimally. We also found several errors in the reference summary in the dataset used.
自动文本摘要(Automatic Text Summarization, ATS)是在文本处理方面利用复杂的技术来帮助人类大量生成文档的摘要或要点的一种方法。我们之所以选择印尼语作为研究对象，是因为目前使用印尼语的自然语言处理研究资源很少。本文利用了来自转换器架构的pltm(预训练语言模型)，即T5(文本到文本传输转换器)，该转换器之前已经完成了一个更大的数据集。本研究的评价是通过比较参考总结和模型总结的ROUGE (Recall-Oriented Understudy for Gisting Evaluation)计算结果来衡量的。对印尼新闻数据集使用预训练的t5基模型和微调参数为220M的实验得到了较高的ROUGE值，即ROUGE-1 = 0.68, ROUGE-2 = 0.61, ROUGE- l = 0.65。评价值工作得很好，但由于在抽象方面，模型没有达到最优的效果，所以最终的模型并没有取得令人满意的结果。我们还在使用的数据集中发现了参考摘要中的几个错误。
{"title":"Abstractive Text Summarization using Pre-Trained Language Model \"Text-to-Text Transfer Transformer (T5)\"","authors":"Qurrota A’yuna Itsnaini, Mardhiya Hayaty, Andriyan Dwi Putra, N. Jabari","doi":"10.33096/ilkom.v15i1.1532.124-131","DOIUrl":"https://doi.org/10.33096/ilkom.v15i1.1532.124-131","url":null,"abstract":"Automatic Text Summarization (ATS) is one of the utilizations of technological sophistication in terms of text processing assisting humans in producing a summary or key points of a document in large quantities. We use Indonesian language as objects because there are few resources in NLP research using Indonesian language. This paper utilized PLTMs (Pre-Trained Language Models) from the transformer architecture, namely T5 (Text-to-Text Transfer Transformer) which has been completed previously with a larger dataset. Evaluation in this study was measured through comparison of the ROUGE (Recall-Oriented Understudy for Gisting Evaluation) calculation results between the reference summary and the model summary. The experiments with the pre-trained t5-base model with fine tuning parameters of 220M for the Indonesian news dataset yielded relatively high ROUGE values, namely ROUGE-1 = 0.68, ROUGE-2 = 0.61, and ROUGE-L = 0.65. The evaluation value worked well, but the resulting model has not achieved satisfactory results because in terms of abstraction, the model did not work optimally. We also found several errors in the reference summary in the dataset used.","PeriodicalId":33690,"journal":{"name":"Ilkom Jurnal Ilmiah","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46911410","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：