Billy Kurniawan · DR-NTU (Nanyang Technological University) · 2020
The abundance of data along with the significant rise of computational power boost the popularity and usage of machine learning in recent decision making and forecasting. Random forest, being one of the current state of the art model, are well known of its high accuracy and efficiency for both classification and regression problems. \n \nIn this paper, Meta Random Forest is introduced to the random forest method to make it even more accurate both in classification and regression problems. Shortly, meta random forest is a method where simple random forests are used as the base for the main random forest. To compare the performance of Meta Random Forest, we use accuracy score for classification problems and R2 score for regression problems.
Cited by 0
Faye Orcales (12488403), Lucy Moctezuma Tan (20485841), Meris Johnson-Hagler (20485844) et al. · Figshare · 2024
Cited by 0Open access
Sonika Ahlawat · Apress eBooks · 2025
In this chapter, we will look at random forests – a versatile, non-linear model that can be used for both regression and classification. Random forests have an impressive mathematical and statistical pedigree to be labeled as advanced statistical models, and their use of decision trees as weak learners and their reliance on bagging of data for training an ensemble of decision trees makes them among the more widely adopted machine learning tools.
Cited by 0
Stephane De Cesco (9333819), John B. Davis (7545314), Paul E. Brennan (1290474) · Figshare · 2020
Cited by 0Open access
Emil D. Attanasi, Timothy C. Coburn · Encyclopedia of earth sciences series/Encyclopedia of earth sciences · 2021
Cited by 4
Subhas Halder, Sudipta Kumar Das, Subrata Kundu et al. · 2025
Constant liver illness may be a driving cause for passing around the world, influencing an expansive number of individuals worldwide. This illness in the liver is caused by a grouping of substances. For example, corpulence is an undiscovered form of hepatitis disease and liquor mishandle. Anomalous nerve work, blood anomalies, kidney problems, liver disease, jaundice, and liver malfunction. This prediction and determination are exceptionally costly and involve complications. Therefore, the aim of this work is to find the best fitted ML to assess the process of diverse machine learning calculations that decrease the problematic tall fetched of constant liver infection conclusion by expectation. This study utilized six calculations: Calculated Relapse, K-Nearest Neighbors, Choice Tree, Naïve Bayes and Arbitrary Timberland. Distinctive classification methods were assessed using different estimation strategies based on precision, exactness, review, and F-1 score. It was found that the correctnesses were 70.18%, 84%, 64.04%, 65.79%, and 71.05% for the methods mentioned above. The investigation revealed that the RF had the most elevated exactness. In addition, our current opinion primarily centers on the utilization of clinical information for foreseeing liver maladies and the examination investigates distinctive ways of showing such information.
Cited by 0
Nashmia Khalid (20496403), Sadaf Hina (20496406), Khurram Shabih Zaidi (20496409) et al. · Figshare · 2025
<div><p>The widespread use of wireless networks to transfer an enormous amount of sensitive information has caused a plethora of vulnerabilities and privacy issues. The management frames, particularly authentication and association frames, are vulnerable to cyberattacks and it is a significant concern. Existing research in Wi-Fi attack detection focused on obtaining high detection accuracy while neglecting modern traffic and attack scenarios such as key reinstallation or unauthorized decryption attacks. This study proposed a novel approach using the AWID 3 dataset for cyberattack detection. The retained features were analyzed to assess their transferability, creating a lightweight and cost-effective model. A decision tree with a recursive feature elimination method was implemented for the extraction of the reduced features subset, and an additional feature wlan_radio.signal_dbm was used in combination with the extracted feature subset. Several deep learning and machine learning models were implemented, where DT and CNN achieved promising classification results. Further, feature transferability and generalizability were evaluated, and their detection performance was analyzed across different network versions where CNN outperformed other classification models. The practical implications of this research are crucial for the secure automation of wireless intrusion detection frameworks and tools in personal and enterprise paradigms.</p></div>
Cited by 0Open access
Nashmia Khalid (20496403), Sadaf Hina (20496406), Khurram Shabih Zaidi (20496409) et al. · Figshare · 2025
<div><p>The widespread use of wireless networks to transfer an enormous amount of sensitive information has caused a plethora of vulnerabilities and privacy issues. The management frames, particularly authentication and association frames, are vulnerable to cyberattacks and it is a significant concern. Existing research in Wi-Fi attack detection focused on obtaining high detection accuracy while neglecting modern traffic and attack scenarios such as key reinstallation or unauthorized decryption attacks. This study proposed a novel approach using the AWID 3 dataset for cyberattack detection. The retained features were analyzed to assess their transferability, creating a lightweight and cost-effective model. A decision tree with a recursive feature elimination method was implemented for the extraction of the reduced features subset, and an additional feature wlan_radio.signal_dbm was used in combination with the extracted feature subset. Several deep learning and machine learning models were implemented, where DT and CNN achieved promising classification results. Further, feature transferability and generalizability were evaluated, and their detection performance was analyzed across different network versions where CNN outperformed other classification models. The practical implications of this research are crucial for the secure automation of wireless intrusion detection frameworks and tools in personal and enterprise paradigms.</p></div>
Cited by 0Open access
Chen, Yinchen · Figshare · 2024
Random forest model
Cited by 0Open access
Yinglin Xia, Jun Sun · 2026
Cited by 0
Nashmia Khalid (20496403), Sadaf Hina (20496406), Khurram Shabih Zaidi (20496409) et al. · Figshare · 2025
<div><p>The widespread use of wireless networks to transfer an enormous amount of sensitive information has caused a plethora of vulnerabilities and privacy issues. The management frames, particularly authentication and association frames, are vulnerable to cyberattacks and it is a significant concern. Existing research in Wi-Fi attack detection focused on obtaining high detection accuracy while neglecting modern traffic and attack scenarios such as key reinstallation or unauthorized decryption attacks. This study proposed a novel approach using the AWID 3 dataset for cyberattack detection. The retained features were analyzed to assess their transferability, creating a lightweight and cost-effective model. A decision tree with a recursive feature elimination method was implemented for the extraction of the reduced features subset, and an additional feature wlan_radio.signal_dbm was used in combination with the extracted feature subset. Several deep learning and machine learning models were implemented, where DT and CNN achieved promising classification results. Further, feature transferability and generalizability were evaluated, and their detection performance was analyzed across different network versions where CNN outperformed other classification models. The practical implications of this research are crucial for the secure automation of wireless intrusion detection frameworks and tools in personal and enterprise paradigms.</p></div>
Cited by 0Open access
Steven J. Rigatti · Journal of Insurance Medicine · 2017
For the task of analyzing survival data to derive risk factors associated with mortality, physicians, researchers, and biostatisticians have typically relied on certain types of regression techniques, most notably the Cox model. With the advent of more widely distributed computing power, methods which require more complex mathematics have become increasingly common. Particularly in this era of "big data" and machine learning, survival analysis has become methodologically broader. This paper aims to explore one technique known as Random Forest. The Random Forest technique is a regression tree technique which uses bootstrap aggregation and randomization of predictors to achieve a high degree of predictive accuracy. The various input parameters of the random forest are explored. Colon cancer data (n = 66,807) from the SEER database is then used to construct both a Cox model and a random forest model to determine how well the models perform on the same data. Both models perform well, achieving a concordance error rate of approximately 18%.
Cited by 1,611
Dothang Truong · 2024
Cited by 0
?????????, ????????? · Scholarworks@UNIST (Ulsan National Institute of Science and Technology) · 2020
Cited by 0
Nashmia Khalid (20496403), Sadaf Hina (20496406), Khurram Shabih Zaidi (20496409) et al. · Figshare · 2025
<div><p>The widespread use of wireless networks to transfer an enormous amount of sensitive information has caused a plethora of vulnerabilities and privacy issues. The management frames, particularly authentication and association frames, are vulnerable to cyberattacks and it is a significant concern. Existing research in Wi-Fi attack detection focused on obtaining high detection accuracy while neglecting modern traffic and attack scenarios such as key reinstallation or unauthorized decryption attacks. This study proposed a novel approach using the AWID 3 dataset for cyberattack detection. The retained features were analyzed to assess their transferability, creating a lightweight and cost-effective model. A decision tree with a recursive feature elimination method was implemented for the extraction of the reduced features subset, and an additional feature wlan_radio.signal_dbm was used in combination with the extracted feature subset. Several deep learning and machine learning models were implemented, where DT and CNN achieved promising classification results. Further, feature transferability and generalizability were evaluated, and their detection performance was analyzed across different network versions where CNN outperformed other classification models. The practical implications of this research are crucial for the secure automation of wireless intrusion detection frameworks and tools in personal and enterprise paradigms.</p></div>
Cited by 0Open access
Chen, Yinchen · Figshare · 2024
Random forest model
Cited by 0Open access
Emil D. Attanasi, Timothy C. Coburn · Encyclopedia of earth sciences series/Encyclopedia of earth sciences · 2023
Cited by 6
Nashmia Khalid (20496403), Sadaf Hina (20496406), Khurram Shabih Zaidi (20496409) et al. · Figshare · 2025
<div><p>The widespread use of wireless networks to transfer an enormous amount of sensitive information has caused a plethora of vulnerabilities and privacy issues. The management frames, particularly authentication and association frames, are vulnerable to cyberattacks and it is a significant concern. Existing research in Wi-Fi attack detection focused on obtaining high detection accuracy while neglecting modern traffic and attack scenarios such as key reinstallation or unauthorized decryption attacks. This study proposed a novel approach using the AWID 3 dataset for cyberattack detection. The retained features were analyzed to assess their transferability, creating a lightweight and cost-effective model. A decision tree with a recursive feature elimination method was implemented for the extraction of the reduced features subset, and an additional feature wlan_radio.signal_dbm was used in combination with the extracted feature subset. Several deep learning and machine learning models were implemented, where DT and CNN achieved promising classification results. Further, feature transferability and generalizability were evaluated, and their detection performance was analyzed across different network versions where CNN outperformed other classification models. The practical implications of this research are crucial for the secure automation of wireless intrusion detection frameworks and tools in personal and enterprise paradigms.</p></div>
Cited by 0Open access
Yanzi-cell · Zenodo (CERN European Organization for Nuclear Research) · 2026
import numpy as np from sklearn.ensemble import RandomForestClassifier from sklearn.metrics import accuracy_score from sklearn.impute import SimpleImputer from sklearn.model_selection import train_test_split # Import this import pandas as pd import matplotlib.pyplot as plt Path to your training Excel file training_file_path = r"C:\Users\Lenovo\Desktop\prediction 1.xlsx" Define column names column_names = ["category", "bacteria", "f1", "f2", "f3", "f4"] Load training data df = pd.read_excel(training_file_path, sheet_name='RAW4', header=None, names=column_names, skiprows=0) features = df[['f1', 'f2', 'f3', 'f4']] categories = df['category'] # Assuming 'category' column contains your categories Handle missing values imputer = SimpleImputer(strategy='mean') features = pd.DataFrame(imputer.fit_transform(features), columns=features.columns) Split the data into training and testing sets X_train, X_test, y_train, y_test = train_test_split(features, categories, test_size=0.2, random_state=42) Train a Random Forest Classifier model = RandomForestClassifier(random_state=42) model.fit(X_train, y_train) Predict the categories on the test set y_pred = model.predict(X_test) Calculate accuracy accuracy = accuracy_score(y_test, y_pred) print(f"Test Data Accuracy: {accuracy:.2f}") If you have a new dataset and want to predict categories prediction_file_path = r"C:\Users\Lenovo\Desktop\prediction 1.xlsx" prediction_df = pd.read_excel(prediction_file_path, sheet_name='Prediction4', header=None, names=column_names, skiprows=0) Define column names column_names = ["category", "bacteria", "f1", "f2", "f3", "f4"] prediction_features = prediction_df[['f1', 'f2', 'f3', 'f4']] Handle missing values in the prediction data prediction_features = pd.DataFrame(imputer.transform(prediction_features), columns=prediction_features.columns) Predict the categories for the new dataset predicted_categories = model.predict(prediction_features) prediction_df['Predicted Category'] = predicted_categories Output the predictions print(prediction_df[['bacteria', 'Predicted Category']])
Cited by 0Open access
Yanzi-cell · Zenodo (CERN European Organization for Nuclear Research) · 2026
import numpy as np from sklearn.ensemble import RandomForestClassifier from sklearn.metrics import accuracy_score from sklearn.impute import SimpleImputer from sklearn.model_selection import train_test_split # Import this import pandas as pd import matplotlib.pyplot as plt Path to your training Excel file training_file_path = r"C:\Users\Lenovo\Desktop\prediction 1.xlsx" Define column names column_names = ["category", "bacteria", "f1", "f2", "f3", "f4"] Load training data df = pd.read_excel(training_file_path, sheet_name='RAW4', header=None, names=column_names, skiprows=0) features = df[['f1', 'f2', 'f3', 'f4']] categories = df['category'] # Assuming 'category' column contains your categories Handle missing values imputer = SimpleImputer(strategy='mean') features = pd.DataFrame(imputer.fit_transform(features), columns=features.columns) Split the data into training and testing sets X_train, X_test, y_train, y_test = train_test_split(features, categories, test_size=0.2, random_state=42) Train a Random Forest Classifier model = RandomForestClassifier(random_state=42) model.fit(X_train, y_train) Predict the categories on the test set y_pred = model.predict(X_test) Calculate accuracy accuracy = accuracy_score(y_test, y_pred) print(f"Test Data Accuracy: {accuracy:.2f}") If you have a new dataset and want to predict categories prediction_file_path = r"C:\Users\Lenovo\Desktop\prediction 1.xlsx" prediction_df = pd.read_excel(prediction_file_path, sheet_name='Prediction4', header=None, names=column_names, skiprows=0) Define column names column_names = ["category", "bacteria", "f1", "f2", "f3", "f4"] prediction_features = prediction_df[['f1', 'f2', 'f3', 'f4']] Handle missing values in the prediction data prediction_features = pd.DataFrame(imputer.transform(prediction_features), columns=prediction_features.columns) Predict the categories for the new dataset predicted_categories = model.predict(prediction_features) prediction_df['Predicted Category'] = predicted_categories Output the predictions print(prediction_df[['bacteria', 'Predicted Category']])
Cited by 0Open access