Teuku Rizky Noviandy, Aga Maulana, Ghazi Mauer Idroes, Talha Bin Emran, Trina Ekawati Tallei, Zuchra Helwani, Rinaldi Idroes
This comprehensive review explores the pivotal role of ensemble machine learning techniques in Quantitative Structure-Activity Relationship (QSAR) modeling for drug discovery. It emphasizes the significance of accurate QSAR models in streamlining candidate compound selection and highlights how ensemble methods, including AdaBoost, Gradient Boosting, Random Forest, Extra Trees, XGBoost, LightGBM, and CatBoost, effectively address challenges such as overfitting and noisy data. The review presents recent applications of ensemble learning in both classification and regression tasks within QSAR, showcasing the exceptional predictive accuracy of these techniques across diverse datasets and target properties. It also discusses the key challenges and considerations in ensemble QSAR modeling, including data quality, model selection, computational resources, and overfitting. The review outlines future directions in ensemble QSAR modeling, including the integration of multi-modal data, explainability, handling imbalanced data, automation, and personalized medicine applications while emphasizing the need for ethical and regulatory guidelines in this evolving field.
{"title":"Ensemble Machine Learning Approach for Quantitative Structure Activity Relationship Based Drug Discovery: A Review","authors":"Teuku Rizky Noviandy, Aga Maulana, Ghazi Mauer Idroes, Talha Bin Emran, Trina Ekawati Tallei, Zuchra Helwani, Rinaldi Idroes","doi":"10.60084/ijds.v1i1.91","DOIUrl":"https://doi.org/10.60084/ijds.v1i1.91","url":null,"abstract":"This comprehensive review explores the pivotal role of ensemble machine learning techniques in Quantitative Structure-Activity Relationship (QSAR) modeling for drug discovery. It emphasizes the significance of accurate QSAR models in streamlining candidate compound selection and highlights how ensemble methods, including AdaBoost, Gradient Boosting, Random Forest, Extra Trees, XGBoost, LightGBM, and CatBoost, effectively address challenges such as overfitting and noisy data. The review presents recent applications of ensemble learning in both classification and regression tasks within QSAR, showcasing the exceptional predictive accuracy of these techniques across diverse datasets and target properties. It also discusses the key challenges and considerations in ensemble QSAR modeling, including data quality, model selection, computational resources, and overfitting. The review outlines future directions in ensemble QSAR modeling, including the integration of multi-modal data, explainability, handling imbalanced data, automation, and personalized medicine applications while emphasizing the need for ethical and regulatory guidelines in this evolving field.","PeriodicalId":486036,"journal":{"name":"Infolitika Journal of Data Science","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135866812","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}
Novi Reandy Sasmita, Siti Ramadeska, Reksi Utami, Zuhra Adha, Ulayya Putri, Risky Haezah Syarafina, La Ode Reskiaddin, Saiful Kamal, Yarmaliza Yarmaliza, Muliadi Muliadi, Arif Saputra
Infant mortality remains a public health problem in Aceh Province, Indonesia. Health services during pregnancy are an essential factor in reducing infant mortality. Studies examining factors such as maternal and child health services that have implications for infant mortality in Aceh province are still scarce. Therefore, this study aims to examine the correlation between maternal and child health services variables such as Blood-Supplementing Tablets (TTD), Coverage of the First Visit of Pregnant Women (K1), Coverage of the First Visit of Pregnant Women (K4), and management of Obstetric Complications to live births and to map the maternal and child health services obtained during pregnancy. A cross-sectional study was used as the research study. This study used descriptive statistics, such as measures of data centering and data dispersion. In this work, inferential statistical analysis was conducted using the Shapiro-Wilk test, Spearman test, and fuzzy c-means. The result of the Shapiro Wilk test stated that the live birth rate variable and all Maternal and Child Healthcare Services variables were not normally distributed (p-value < 0.05), all Maternal and Child Healthcare Services variables were positively correlated to live birth rate based on the Spearman test (p-value < 0.05). Based on the Silhouette Index with 0.555, the formation of 3 clusters is the optimal cluster. The clustering is based on the Maternal and Child Healthcare Services that have been provided, where the first, second, and third clusters consist of five districts/city, eight districts/city, and ten districts/city, respectively, as a result of Fuzzy C-Means Clustering.
{"title":"Maternal and Child Healthcare Services in Aceh Province, Indonesia: A Correlation and Clustering Analysis in Statistics","authors":"Novi Reandy Sasmita, Siti Ramadeska, Reksi Utami, Zuhra Adha, Ulayya Putri, Risky Haezah Syarafina, La Ode Reskiaddin, Saiful Kamal, Yarmaliza Yarmaliza, Muliadi Muliadi, Arif Saputra","doi":"10.60084/ijds.v1i1.88","DOIUrl":"https://doi.org/10.60084/ijds.v1i1.88","url":null,"abstract":"Infant mortality remains a public health problem in Aceh Province, Indonesia. Health services during pregnancy are an essential factor in reducing infant mortality. Studies examining factors such as maternal and child health services that have implications for infant mortality in Aceh province are still scarce. Therefore, this study aims to examine the correlation between maternal and child health services variables such as Blood-Supplementing Tablets (TTD), Coverage of the First Visit of Pregnant Women (K1), Coverage of the First Visit of Pregnant Women (K4), and management of Obstetric Complications to live births and to map the maternal and child health services obtained during pregnancy. A cross-sectional study was used as the research study. This study used descriptive statistics, such as measures of data centering and data dispersion. In this work, inferential statistical analysis was conducted using the Shapiro-Wilk test, Spearman test, and fuzzy c-means. The result of the Shapiro Wilk test stated that the live birth rate variable and all Maternal and Child Healthcare Services variables were not normally distributed (p-value < 0.05), all Maternal and Child Healthcare Services variables were positively correlated to live birth rate based on the Spearman test (p-value < 0.05). Based on the Silhouette Index with 0.555, the formation of 3 clusters is the optimal cluster. The clustering is based on the Maternal and Child Healthcare Services that have been provided, where the first, second, and third clusters consist of five districts/city, eight districts/city, and ten districts/city, respectively, as a result of Fuzzy C-Means Clustering.","PeriodicalId":486036,"journal":{"name":"Infolitika Journal of Data Science","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135866811","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}
Leukemia is a blood cancer in which blood cells become malignant and uncontrolled. It can cause damage to the function of the body's organs. Several machine learning methods have been used to automatically detect biomedical images, including blood cell images. In this study, we utilized a hybrid machine learning method, called a hybrid Convolutional Neural Network-eXtreme Gradient Boosting (CNN-XGBoost) method to detect leukemia in blood cells. The hybrid method combines two machine learning methods. We use CNN as the basic classifier and XGBoost as the main classification method. The aim of this methodology was to assess whether incorporating the basic classification method would lead to an enhancement in the performance of the main classification model. The experimental findings demonstrated that the utilization of XGBoost as the main classifier led to a marginal increase in accuracy, elevating it from 85.32% to 85.43% compared to the basic CNN classification. This research highlights the potential of hybrid machine learning approaches in biomedical image analysis and their role in advancing the early diagnosis of leukemia and potentially other medical conditions.
{"title":"Implementation of Hybrid CNN-XGBoost Method for Leukemia Detection Problem","authors":"Taufiq Hidayat, Edrian Hadinata, Irfan Sudahri Damanik, Zakial Vikki, Irvanizam Irvanizam","doi":"10.60084/ijds.v1i1.87","DOIUrl":"https://doi.org/10.60084/ijds.v1i1.87","url":null,"abstract":"Leukemia is a blood cancer in which blood cells become malignant and uncontrolled. It can cause damage to the function of the body's organs. Several machine learning methods have been used to automatically detect biomedical images, including blood cell images. In this study, we utilized a hybrid machine learning method, called a hybrid Convolutional Neural Network-eXtreme Gradient Boosting (CNN-XGBoost) method to detect leukemia in blood cells. The hybrid method combines two machine learning methods. We use CNN as the basic classifier and XGBoost as the main classification method. The aim of this methodology was to assess whether incorporating the basic classification method would lead to an enhancement in the performance of the main classification model. The experimental findings demonstrated that the utilization of XGBoost as the main classifier led to a marginal increase in accuracy, elevating it from 85.32% to 85.43% compared to the basic CNN classification. This research highlights the potential of hybrid machine learning approaches in biomedical image analysis and their role in advancing the early diagnosis of leukemia and potentially other medical conditions.","PeriodicalId":486036,"journal":{"name":"Infolitika Journal of Data Science","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135926595","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}
This study aims to evaluate the implementation and effectiveness of the Adaptive Neuro-Fuzzy Inference System (ANFIS) based Quantitative Structure Retention Relationship (QSRR) to predict the Kovats retention index of compounds in gas chromatography. The model was trained using 340 essential oil compounds and their molecular descriptors. The evaluation of the ANFIS models revealed promising results, achieving an R2 of 0.974, an RMSE of 48.12, and an MAPE of 3.3% on the testing set. These findings highlight the ANFIS approach as remarkably accurate in its predictive capacity for determining the Kovats retention index in the context of gas chromatography. This study provides valuable perspectives on the efficiency of retention index prediction through ANFIS-based QSRR methods and the potential practicality in compound analysis and chromatographic optimization.
{"title":"ANFIS-Based QSRR Modelling for Kovats Retention Index Prediction in Gas Chromatography","authors":"Rinaldi Idroes, Teuku Rizky Noviandy, Aga Maulana, Rivansyah Suhendra, Novi Reandy Sasmita, Muslem Muslem, Ghazi Mauer Idroes, Raudhatul Jannah, Razief Perucha Fauzie Afidh, Irvanizam Irvanizam","doi":"10.60084/ijds.v1i1.73","DOIUrl":"https://doi.org/10.60084/ijds.v1i1.73","url":null,"abstract":"This study aims to evaluate the implementation and effectiveness of the Adaptive Neuro-Fuzzy Inference System (ANFIS) based Quantitative Structure Retention Relationship (QSRR) to predict the Kovats retention index of compounds in gas chromatography. The model was trained using 340 essential oil compounds and their molecular descriptors. The evaluation of the ANFIS models revealed promising results, achieving an R2 of 0.974, an RMSE of 48.12, and an MAPE of 3.3% on the testing set. These findings highlight the ANFIS approach as remarkably accurate in its predictive capacity for determining the Kovats retention index in the context of gas chromatography. This study provides valuable perspectives on the efficiency of retention index prediction through ANFIS-based QSRR methods and the potential practicality in compound analysis and chromatographic optimization.","PeriodicalId":486036,"journal":{"name":"Infolitika Journal of Data Science","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135098988","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}
Diabetes is a chronic condition characterized by elevated blood glucose levels which leads to organ dysfunction and an increased risk of premature death. The global prevalence of diabetes has been rising, necessitating an accurate and timely diagnosis to achieve the most effective management. Recent advancements in the field of machine learning have opened new possibilities for improving diabetes detection and management. In this study, we propose a fine-tuned XGBoost model for diabetes detection. We use the Pima Indian Diabetes dataset and employ a random search for hyperparameter tuning. The fine-tuned XGBoost model is compared with six other popular machine learning models and achieves the highest performance in accuracy, precision, sensitivity, and F1-score. This study demonstrates the potential of the fine-tuned XGBoost model as a robust and efficient tool for diabetes detection. The insights of this study advance medical diagnostics for efficient and personalized management of diabetes.
{"title":"Machine Learning Approach for Diabetes Detection Using Fine-Tuned XGBoost Algorithm","authors":"Aga Maulana, Farassa Rani Faisal, Teuku Rizky Noviandy, Tatsa Rizkia, Ghazi Mauer Idroes, Trina Ekawati Tallei, Mohamed El-Shazly, Rinaldi Idroes","doi":"10.60084/ijds.v1i1.72","DOIUrl":"https://doi.org/10.60084/ijds.v1i1.72","url":null,"abstract":"Diabetes is a chronic condition characterized by elevated blood glucose levels which leads to organ dysfunction and an increased risk of premature death. The global prevalence of diabetes has been rising, necessitating an accurate and timely diagnosis to achieve the most effective management. Recent advancements in the field of machine learning have opened new possibilities for improving diabetes detection and management. In this study, we propose a fine-tuned XGBoost model for diabetes detection. We use the Pima Indian Diabetes dataset and employ a random search for hyperparameter tuning. The fine-tuned XGBoost model is compared with six other popular machine learning models and achieves the highest performance in accuracy, precision, sensitivity, and F1-score. This study demonstrates the potential of the fine-tuned XGBoost model as a robust and efficient tool for diabetes detection. The insights of this study advance medical diagnostics for efficient and personalized management of diabetes.","PeriodicalId":486036,"journal":{"name":"Infolitika Journal of Data Science","volume":"262 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135621245","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：