Pub Date : 2022-07-06DOI: 10.1109/ICoDSA55874.2022.9862851
A. Kurniati, G. Wisudiawan, G. Kusuma
Clinical pathway analysis is an important analysis in the healthcare domain. This approach learns from historical data of patient pathways of clinical treatments and finds patterns to be used for further purposes, including treatment recommendations and precision medicine. Indonesia has the opportunity for clinical pathway analysis by using the Indonesia national health insurance data samples provided by the Social Security Administrator-Healthcare. The data samples are representative of the Indonesian population and potentially useful for initial explorations of clinical pathways in Indonesia. This study applied an exploratory data analysis using process mining for clinical pathway analysis. Process mining is a promising approach to learn from time-stamped datasets to find sequenced clinical pathway patterns. We examine the data samples carefully to define the minimum components of process mining for clinical pathway analysis and provide samples of the results of the clinical pathway analysis. Contributions of this study are two folds: to promote process mining for clinical pathway analysis and to present a case study of clinical pathway analysis using the Indonesia National Health Insurance data samples. The contributions of this paper are to promote clinical pathway analysis to improving health services using real data from BPJS Kesehatan system, and to propose a method for clinical pathway analysis based on process mining. The results of this study are disease trajectory visualization through a process model and statistics evaluating the performance of the results using process mining techniques.
{"title":"Potentials of Clinical Pathway Analysis Using Process Mining on the Indonesia National Health Insurance Data Samples: an Exploratory Data Analysis","authors":"A. Kurniati, G. Wisudiawan, G. Kusuma","doi":"10.1109/ICoDSA55874.2022.9862851","DOIUrl":"https://doi.org/10.1109/ICoDSA55874.2022.9862851","url":null,"abstract":"Clinical pathway analysis is an important analysis in the healthcare domain. This approach learns from historical data of patient pathways of clinical treatments and finds patterns to be used for further purposes, including treatment recommendations and precision medicine. Indonesia has the opportunity for clinical pathway analysis by using the Indonesia national health insurance data samples provided by the Social Security Administrator-Healthcare. The data samples are representative of the Indonesian population and potentially useful for initial explorations of clinical pathways in Indonesia. This study applied an exploratory data analysis using process mining for clinical pathway analysis. Process mining is a promising approach to learn from time-stamped datasets to find sequenced clinical pathway patterns. We examine the data samples carefully to define the minimum components of process mining for clinical pathway analysis and provide samples of the results of the clinical pathway analysis. Contributions of this study are two folds: to promote process mining for clinical pathway analysis and to present a case study of clinical pathway analysis using the Indonesia National Health Insurance data samples. The contributions of this paper are to promote clinical pathway analysis to improving health services using real data from BPJS Kesehatan system, and to propose a method for clinical pathway analysis based on process mining. The results of this study are disease trajectory visualization through a process model and statistics evaluating the performance of the results using process mining techniques.","PeriodicalId":339135,"journal":{"name":"2022 International Conference on Data Science and Its Applications (ICoDSA)","volume":"94 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126242489","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 : 2022-07-06DOI: 10.1109/ICoDSA55874.2022.9862831
Lathifa Rizqi Andhary, H. Nuha, T. Haryanti
In no time, 4G will be replaced by 5G. Furthermore, 5G still uses the Orthogonal Frequency Division Multiple Access schemes. However, Orthogonal Frequency Division Multiple Access has shortcomings in terms of massive connectivity, so a new scheme is needed to replace Orthogonal Frequency Division Multiple Access for 5G. So, this paper discusses the performance evaluation of the Sparse Code Multiple Access for 5G. In addition to the Orthogonal Frequency Division Multiple Access, a message-passing algorithm can also be used to decode Sparse Code Multiple Access messages so they can deal with large amounts of traffic. If the amount of traffic is large, it can accommodate big number of users in the use of 5G. The message-passing algorithm can meet 5G specifications, so it is suitable for 5G use. The performance of Sparse Code Multiple Access will confirm its usability for the 5G system. So, the message-passing algorithm can detect the user information on Sparse Code Multiple Access, to increasing complexity and to find bit error rate in the algorithm. This paper uses the message-passing algorithm to test the bit error rate on six types of codebooks. From all the codebooks, the codebook with smallest bit error rate is best codebook to choose for Sparse Code Multiple Access.
{"title":"Sparse Code Multiple Access Decoding Using Message-Passing Algorithm","authors":"Lathifa Rizqi Andhary, H. Nuha, T. Haryanti","doi":"10.1109/ICoDSA55874.2022.9862831","DOIUrl":"https://doi.org/10.1109/ICoDSA55874.2022.9862831","url":null,"abstract":"In no time, 4G will be replaced by 5G. Furthermore, 5G still uses the Orthogonal Frequency Division Multiple Access schemes. However, Orthogonal Frequency Division Multiple Access has shortcomings in terms of massive connectivity, so a new scheme is needed to replace Orthogonal Frequency Division Multiple Access for 5G. So, this paper discusses the performance evaluation of the Sparse Code Multiple Access for 5G. In addition to the Orthogonal Frequency Division Multiple Access, a message-passing algorithm can also be used to decode Sparse Code Multiple Access messages so they can deal with large amounts of traffic. If the amount of traffic is large, it can accommodate big number of users in the use of 5G. The message-passing algorithm can meet 5G specifications, so it is suitable for 5G use. The performance of Sparse Code Multiple Access will confirm its usability for the 5G system. So, the message-passing algorithm can detect the user information on Sparse Code Multiple Access, to increasing complexity and to find bit error rate in the algorithm. This paper uses the message-passing algorithm to test the bit error rate on six types of codebooks. From all the codebooks, the codebook with smallest bit error rate is best codebook to choose for Sparse Code Multiple Access.","PeriodicalId":339135,"journal":{"name":"2022 International Conference on Data Science and Its Applications (ICoDSA)","volume":"104 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121398802","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 : 2022-07-06DOI: 10.1109/ICoDSA55874.2022.9862910
Derry Alamsyah, Samsuryadi, Wijang Widhiarsho, S. Hasan
Handwriting analysis is an approach to get information through the handwriting. It extremely useful information, for instance in personality traits identification. The information came from the feature extracted from the handwriting. This feature can be size, slantness, pressure, and so forth. In this research, handwriting analysis is through the AND dataset that provide handwriting dataset along with feature label while most public dataset has nothing with it. By using the Coonvolutional Neural Networks (CNN) potentiality in capturing and recognizing global features, there are 15 models had built in this research in accordance with each feature and divided into three group by its number of types. After built a simple CNN architecture by only conduct two convolution layer, overall result show fair enough performance where the highest rate of accuracy is 80.88%. Furthermore, there are three best features had recognized, which is "entry stroke ‘A’", "size", and "slantness", where the last two is naturally global features. However, the fact that handwriting image data cannot be oversampled which can lead to the bias result, than the imbalance data becomes a problem in this research that reduced the model performance.
{"title":"Handwriting Analysis for Personality Trait Features Identification using CNN","authors":"Derry Alamsyah, Samsuryadi, Wijang Widhiarsho, S. Hasan","doi":"10.1109/ICoDSA55874.2022.9862910","DOIUrl":"https://doi.org/10.1109/ICoDSA55874.2022.9862910","url":null,"abstract":"Handwriting analysis is an approach to get information through the handwriting. It extremely useful information, for instance in personality traits identification. The information came from the feature extracted from the handwriting. This feature can be size, slantness, pressure, and so forth. In this research, handwriting analysis is through the AND dataset that provide handwriting dataset along with feature label while most public dataset has nothing with it. By using the Coonvolutional Neural Networks (CNN) potentiality in capturing and recognizing global features, there are 15 models had built in this research in accordance with each feature and divided into three group by its number of types. After built a simple CNN architecture by only conduct two convolution layer, overall result show fair enough performance where the highest rate of accuracy is 80.88%. Furthermore, there are three best features had recognized, which is \"entry stroke ‘A’\", \"size\", and \"slantness\", where the last two is naturally global features. However, the fact that handwriting image data cannot be oversampled which can lead to the bias result, than the imbalance data becomes a problem in this research that reduced the model performance.","PeriodicalId":339135,"journal":{"name":"2022 International Conference on Data Science and Its Applications (ICoDSA)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132815932","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 : 2022-07-06DOI: 10.1109/ICoDSA55874.2022.9862872
H. Nuha, Abdi T. Abdalla
Seismic Data, an exploration method of sending energy or sound waves into the earth and recording the wave reflections to reveal essential subsurface rock information including type, size, shape and depth. Seismic data acquisition typically produces a significant data size. Seismic files within a survey may include useless noisy traces that increase the file size. Noisy traces have some noticeable features which can be exploited to aid the denoising process. In this work, the features were formulated based on the Principal Component Analysis (PCA) to automatically distinguish excellent traces from noisy traces. PCA projects the seismic trace features to a lower dimension with only two features. To classify and detect noisy traces, we first select the dataset and generate Gaussian noise, then add the noise to the selected dataset and then normalize the traces before extracting the features: threshold algorithm, histogram algorithm, and zero-crossing algorithm and finally apply the PCA to obtain the projected data. In this work, two types of artificial noises were generated. It is shown that PCA is able to separate two types of noisy seismic traces. PCA projections show that at high noise contamination, the method is unable to separate the noisy and clean seismic traces.
{"title":"Noisy Seismic Traces Classification Using Principal Component Analysis","authors":"H. Nuha, Abdi T. Abdalla","doi":"10.1109/ICoDSA55874.2022.9862872","DOIUrl":"https://doi.org/10.1109/ICoDSA55874.2022.9862872","url":null,"abstract":"Seismic Data, an exploration method of sending energy or sound waves into the earth and recording the wave reflections to reveal essential subsurface rock information including type, size, shape and depth. Seismic data acquisition typically produces a significant data size. Seismic files within a survey may include useless noisy traces that increase the file size. Noisy traces have some noticeable features which can be exploited to aid the denoising process. In this work, the features were formulated based on the Principal Component Analysis (PCA) to automatically distinguish excellent traces from noisy traces. PCA projects the seismic trace features to a lower dimension with only two features. To classify and detect noisy traces, we first select the dataset and generate Gaussian noise, then add the noise to the selected dataset and then normalize the traces before extracting the features: threshold algorithm, histogram algorithm, and zero-crossing algorithm and finally apply the PCA to obtain the projected data. In this work, two types of artificial noises were generated. It is shown that PCA is able to separate two types of noisy seismic traces. PCA projections show that at high noise contamination, the method is unable to separate the noisy and clean seismic traces.","PeriodicalId":339135,"journal":{"name":"2022 International Conference on Data Science and Its Applications (ICoDSA)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122507217","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 : 2022-07-06DOI: 10.1109/ICoDSA55874.2022.9862934
Cagri Karahan, Ş. Öğüdücü
In recent years, machine learning and deep learning techniques have been frequently used in Algorithmic Trading. Algorithmic Trading means trading Forex, stock market, commodities, and many markets with the help of computers using systems created with various technical analysis indicators. The BTC/USD market is a market that allows buying and selling of products. People aim to profit by buying and selling in the Bitcoin market. Reinforcement Learning (RL) was also helpful in achieving those kinds of goals. Reinforcement learning is a sub-topic of machine learning. RL addresses the problem of a computational agent learning to make decisions by trial and error. For our application, it is aimed to make as much profit as possible. This study focuses on developing a novel tool to automate currency trading like a BTC/USD in a simulated market with maximum profit and minimum loss. RL technique with a modified version of the Collective Decision Optimization Algorithm is used to implement the proposed model. Feature engineering is also performed to create features that improve the result.
{"title":"Cryptocurrency Trading based on Heuristic Guided Approach with Feature Engineering","authors":"Cagri Karahan, Ş. Öğüdücü","doi":"10.1109/ICoDSA55874.2022.9862934","DOIUrl":"https://doi.org/10.1109/ICoDSA55874.2022.9862934","url":null,"abstract":"In recent years, machine learning and deep learning techniques have been frequently used in Algorithmic Trading. Algorithmic Trading means trading Forex, stock market, commodities, and many markets with the help of computers using systems created with various technical analysis indicators. The BTC/USD market is a market that allows buying and selling of products. People aim to profit by buying and selling in the Bitcoin market. Reinforcement Learning (RL) was also helpful in achieving those kinds of goals. Reinforcement learning is a sub-topic of machine learning. RL addresses the problem of a computational agent learning to make decisions by trial and error. For our application, it is aimed to make as much profit as possible. This study focuses on developing a novel tool to automate currency trading like a BTC/USD in a simulated market with maximum profit and minimum loss. RL technique with a modified version of the Collective Decision Optimization Algorithm is used to implement the proposed model. Feature engineering is also performed to create features that improve the result.","PeriodicalId":339135,"journal":{"name":"2022 International Conference on Data Science and Its Applications (ICoDSA)","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129390418","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 : 2022-07-06DOI: 10.1109/ICoDSA55874.2022.9862914
Yiqing Zhou, Rui Liao, Yong-hong Chen
In the paper, according to the original data and the value of the sensor at different moments, the box diagram method is used to process the data, and divides the normal value and outliers. The two types of outliers were distinguished based on the persistence of the outliers in the longitudinal time of the data and the linkage of the lateral sensors, and the clustering algorithm was used to reclassify the data. Then, persistence and linkage were calculated within each class, dividing the sum of persistence and linkage by the result of the maximum number of possible anomalies as the risk coefficient, and then defining a threshold to distinguish between risk-specific and non-risk anomalies. Later, a comprehensive evaluation model of anomaly degree was established through quantitative score, principal component analysis and 0,1 planning. Finally, this quantitative evaluation method is evaluated objectively.
{"title":"Study on Optimization of Data-Driven Anomaly Detection","authors":"Yiqing Zhou, Rui Liao, Yong-hong Chen","doi":"10.1109/ICoDSA55874.2022.9862914","DOIUrl":"https://doi.org/10.1109/ICoDSA55874.2022.9862914","url":null,"abstract":"In the paper, according to the original data and the value of the sensor at different moments, the box diagram method is used to process the data, and divides the normal value and outliers. The two types of outliers were distinguished based on the persistence of the outliers in the longitudinal time of the data and the linkage of the lateral sensors, and the clustering algorithm was used to reclassify the data. Then, persistence and linkage were calculated within each class, dividing the sum of persistence and linkage by the result of the maximum number of possible anomalies as the risk coefficient, and then defining a threshold to distinguish between risk-specific and non-risk anomalies. Later, a comprehensive evaluation model of anomaly degree was established through quantitative score, principal component analysis and 0,1 planning. Finally, this quantitative evaluation method is evaluated objectively.","PeriodicalId":339135,"journal":{"name":"2022 International Conference on Data Science and Its Applications (ICoDSA)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116435641","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}
YouTube is often used in public and private universities as branding for first-year students. YouTube facilitates users to interact by giving likes or dislikes, adding viewers to the video, and responding to videos through comment pages that can analyze by public feedback for branding. In doing branding, many alumni and college students discuss Telkom University as the best private university in content uploaded on YouTube. That can trigger the public to give positive, negative, or neutral comments to Telkom University. In this research, sentiment analysis focuses on the scientific context of branding the slogan "Number 1 Best Private University" to find out the perspectives and opinions of the public that can be used as evaluation material for the university to improve its reputation. Dataset takes from user opinions on YouTube regarding content that discusses Telkom University's branding slogan using the Term Frequency–Inverse Document Frequency (TF-IDF) feature extraction and Naïve Bayes as a classification. The final results of this research show that the ratio of 90:10 normalized then combined with the unigram-bigram token and Naïve Bayes with alpha 0.6 brings out the best performance, with an average accuracy of 85.27%, the precision of 91.41%, recall of 62.45%, and the F1-Score of 64.78%.
{"title":"Telkom University Slogan Analysis on YouTube Using Naïve Bayes","authors":"Rahma Fadhila Moenggah, Donni Richasdy, Mahendra Dwifebri Purbolaksono","doi":"10.1109/ICoDSA55874.2022.9862818","DOIUrl":"https://doi.org/10.1109/ICoDSA55874.2022.9862818","url":null,"abstract":"YouTube is often used in public and private universities as branding for first-year students. YouTube facilitates users to interact by giving likes or dislikes, adding viewers to the video, and responding to videos through comment pages that can analyze by public feedback for branding. In doing branding, many alumni and college students discuss Telkom University as the best private university in content uploaded on YouTube. That can trigger the public to give positive, negative, or neutral comments to Telkom University. In this research, sentiment analysis focuses on the scientific context of branding the slogan \"Number 1 Best Private University\" to find out the perspectives and opinions of the public that can be used as evaluation material for the university to improve its reputation. Dataset takes from user opinions on YouTube regarding content that discusses Telkom University's branding slogan using the Term Frequency–Inverse Document Frequency (TF-IDF) feature extraction and Naïve Bayes as a classification. The final results of this research show that the ratio of 90:10 normalized then combined with the unigram-bigram token and Naïve Bayes with alpha 0.6 brings out the best performance, with an average accuracy of 85.27%, the precision of 91.41%, recall of 62.45%, and the F1-Score of 64.78%.","PeriodicalId":339135,"journal":{"name":"2022 International Conference on Data Science and Its Applications (ICoDSA)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116312713","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 : 2022-07-06DOI: 10.1109/icodsa55874.2022.9862819
{"title":"ICoDSA 2022 TOC","authors":"","doi":"10.1109/icodsa55874.2022.9862819","DOIUrl":"https://doi.org/10.1109/icodsa55874.2022.9862819","url":null,"abstract":"","PeriodicalId":339135,"journal":{"name":"2022 International Conference on Data Science and Its Applications (ICoDSA)","volume":"200 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122427062","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 : 2022-07-06DOI: 10.1109/ICoDSA55874.2022.9862827
Akbar Ridwan, H. Nuha, Ramanti Dharayani
Indonesia is 6th largest population affected by floods in the world, which is 640.000 people every year. Indonesia areas that often experience floods due to high-intensity rainfall and tropical climate. Recently, there was a flood in South Kalimantan on January 14, 2021. From this incident, few netizen expressed their opinions about the natural flood disaster through Twitter social media. In this study, the author will classify netizen views regarding the natural flood disaster so that the netizen is aware of the incident and they can prevent flood causes. We will divide the tweet into relevant and irrelevant categories to categorize the incident using the Naïve Bayes Classifier. This research implements N-gram features to consider the most efficient method for determining a classification. We use Naïve Bayes because it assumes all variables are unique and provides weight to the text data using N-Gram. The importance of text data could be used to create a Naïve Bayes Classification model to calculate the probability. The naïve Bayes method can be implemented in classifying natural flood disasters. The tweet within the result using bigram will give higher accuracy than unigram or trigram. According this study the goverment can have plan for future mitigation action.
{"title":"Sentiment Analysis of Floods on Twitter Social Media Using the Naive Bayes Classifier Method with the N-Gram Feature","authors":"Akbar Ridwan, H. Nuha, Ramanti Dharayani","doi":"10.1109/ICoDSA55874.2022.9862827","DOIUrl":"https://doi.org/10.1109/ICoDSA55874.2022.9862827","url":null,"abstract":"Indonesia is 6th largest population affected by floods in the world, which is 640.000 people every year. Indonesia areas that often experience floods due to high-intensity rainfall and tropical climate. Recently, there was a flood in South Kalimantan on January 14, 2021. From this incident, few netizen expressed their opinions about the natural flood disaster through Twitter social media. In this study, the author will classify netizen views regarding the natural flood disaster so that the netizen is aware of the incident and they can prevent flood causes. We will divide the tweet into relevant and irrelevant categories to categorize the incident using the Naïve Bayes Classifier. This research implements N-gram features to consider the most efficient method for determining a classification. We use Naïve Bayes because it assumes all variables are unique and provides weight to the text data using N-Gram. The importance of text data could be used to create a Naïve Bayes Classification model to calculate the probability. The naïve Bayes method can be implemented in classifying natural flood disasters. The tweet within the result using bigram will give higher accuracy than unigram or trigram. According this study the goverment can have plan for future mitigation action.","PeriodicalId":339135,"journal":{"name":"2022 International Conference on Data Science and Its Applications (ICoDSA)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127944161","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 : 2022-07-06DOI: 10.1109/ICoDSA55874.2022.9862837
Mochammad Ivan Adhyaksa Pradana, A. Kurniati, G. Wisudiawan
Process Mining is a method to collect data about business processes and produce insight from those processes. This method can be applied to many sectors, including healthcare. One of the government’s programs to provide health services for the citizens is the Indonesia Health Social Security Agency (Badan Penyelenggara Jaminan Sosial Kesehatan/ BPJS Kesehatan). Currently, the services provided in this program are still unsatisfying, with one main concern in the waiting time. We analyze BPJS Kesehatan data samples using the Inductive Miner algorithm to mine event logs of treatment, frequent treatments, and health facility usage, with a focus on respiratory disease. Initial steps were needed in preprocessing to prepare the event logs. The produced process models are then evaluated based on their fitness, precision, generalization, and simplicity. Then, we replay the model toward the event logs for performance analysis. We test different types of Inductive Miner and found that the Inductive Miner Infrequent variant achieves the highest average score among other variants. We find eight treatment procedures that can be improved in terms of efficiency. We also find out that the most frequently used health facility is Public Health Center, followed by First Clinic and Hospital. The results are analyzed from the perspective of previously done treatment, recurring treatment, and facility usage process. Inductive Miner is a good algorithm that can produce an accurate process model and allow suggestions for improving the healthcare process.
流程挖掘是一种收集有关业务流程的数据并从这些流程中产生洞察力的方法。这种方法可以应用于许多部门，包括医疗保健。为公民提供医疗服务的政府项目之一是印度尼西亚卫生社会保障局(Badan Penyelenggara Jaminan Social Kesehatan/ BPJS Kesehatan)。目前，该计划提供的服务仍不令人满意，主要的一个问题是等待时间。我们使用归纳Miner算法分析BPJS Kesehatan数据样本，以挖掘治疗、频繁治疗和卫生设施使用的事件日志，重点是呼吸系统疾病。在预处理过程中需要初始步骤来准备事件日志。然后根据其适合度、精度、泛化和简单性对生成的流程模型进行评估。然后，我们对事件日志重放模型以进行性能分析。我们测试了不同类型的归纳Miner，发现归纳Miner infrequency变体在其他变体中获得了最高的平均分。我们发现有八种治疗方法可以提高效率。我们还发现，最常使用的卫生设施是公共卫生中心，其次是第一诊所和医院。从既往治疗、重复治疗和设施使用过程三个方面对结果进行分析。电感式Miner是一种很好的算法，它可以生成准确的流程模型，并为改进医疗保健流程提供建议。
{"title":"Inductive Miner Implementation to Improve Healthcare Efficiency on Indonesia National Health Insurance Data","authors":"Mochammad Ivan Adhyaksa Pradana, A. Kurniati, G. Wisudiawan","doi":"10.1109/ICoDSA55874.2022.9862837","DOIUrl":"https://doi.org/10.1109/ICoDSA55874.2022.9862837","url":null,"abstract":"Process Mining is a method to collect data about business processes and produce insight from those processes. This method can be applied to many sectors, including healthcare. One of the government’s programs to provide health services for the citizens is the Indonesia Health Social Security Agency (Badan Penyelenggara Jaminan Sosial Kesehatan/ BPJS Kesehatan). Currently, the services provided in this program are still unsatisfying, with one main concern in the waiting time. We analyze BPJS Kesehatan data samples using the Inductive Miner algorithm to mine event logs of treatment, frequent treatments, and health facility usage, with a focus on respiratory disease. Initial steps were needed in preprocessing to prepare the event logs. The produced process models are then evaluated based on their fitness, precision, generalization, and simplicity. Then, we replay the model toward the event logs for performance analysis. We test different types of Inductive Miner and found that the Inductive Miner Infrequent variant achieves the highest average score among other variants. We find eight treatment procedures that can be improved in terms of efficiency. We also find out that the most frequently used health facility is Public Health Center, followed by First Clinic and Hospital. The results are analyzed from the perspective of previously done treatment, recurring treatment, and facility usage process. Inductive Miner is a good algorithm that can produce an accurate process model and allow suggestions for improving the healthcare process.","PeriodicalId":339135,"journal":{"name":"2022 International Conference on Data Science and Its Applications (ICoDSA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129133221","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：