China
Africa
Explore chapters and articles related to this topic
Identification of Emotion Parameters in Music to Modulate Human Affective States
Published in Wellington Pinheiro dos Santos, Juliana Carneiro Gomes, Valter Augusto de Freitas Barbosa, Swarm Intelligence Trends and Applications, 2023
Maíra A. Santana, Ingrid B. Nunes, Andressa L.Q. Ribeiro, Flávio S. Fonseca, Arianne S. Torcate, Amanda Suarez, Vanessa Marques, Nathália Córdula, Juliana C. Gomes, Giselle M.M. Moreno, Wellington P. Santos
Random Forest is an algorithm that works by combining decision trees. Each of these trees is composed by vectors of clustered samples, randomly obtained. The vectors have a similar distribution for all trees in the forest and select subsets of random samples from the initial data (Biau and Scornet, 2016; Breiman, 2001). This algorithm uses the supervised learning paradigm and are widely applied to solve classification and regression problems. However, for classification problems, the chosen prediction is given by the class most voted by the trees. In regression problems, the prediction is given by the average of the values obtained by each tree (Biau and Scornet, 2016). Due to the operating mode of this algorithm, RF can be applied in databases with large data volumes. In addition, it also adapts well to databases with missing data and does not tend to overfit the model, even with the increase in the number of trees (Biau and Scornet, 2016; Breiman, 2001; da Silva et al., 2021; de Lima et al., 2020).
Comparative Analysis for Moving Object Detection in Vibrant Background Using SVM, LS SVM and Random Forest Classifier
Published in Amit Kumar Tyagi, Ajith Abraham, A. Kaklauskas, N. Sreenath, Gillala Rekha, Shaveta Malik, Security and Privacy-Preserving Techniques in Wireless Robotics, 2022
Random Forest is a supervised learning algorithm. This algorithm creates multiple decision trees and makes it random. The forest it builds is a collection of Decision Trees as shown in Figure 6.1. Random forest builds multiple decision trees and merges them together to get a more accurate and stable prediction. Random Forest has nearly the same hyper-parameters as a decision tree. In Random Forest, only a random subset of the features is taken into consideration by the algorithm to split a node. Trees can be made more random, by additional use of random thresholds for each feature in spite of searching for the best possible thresholds (like a normal decision tree does).[6] Random forest algorithm works on the multiple random decision trees as shown in Figure 6.1; it automatically returns the output that appears in the maximum time on different random sets of decision trees. Here it shows that the node indicated with the red mark appears for three decision trees and node blue appears in one tree, so that the random forest algorithm will return the output that appears at red dot node.
Pixel-Based Classification of Land Use/Land Cover Built-Up and Non-Built-Up Areas Using Google Earth Engine in an Urban Region (Delhi, India)
Published in Mohamed Lahby, Utku Kose, Akash Kumar Bhoi, Explainable Artificial Intelligence for Smart Cities, 2021
A. Kumar, A. Jain, B. Agarwal, M. Jain, P. Harjule, R.A. Verma
Random Forest is a supervised machine learning model which is an extension of the decision tree model but in place of a single tree, a large number of trees are created to achieve higher accuracy. In this model, a forest of trees is created where for each tree features are selected randomly and the output is calculated by finding out the average or mean of all the trees. The Decision Trees are sensitive on the training data, and as such that a small change may result in a significant change in the data. Random Forest uses this property of decision trees and uses the technique of bagging, in which features are selected randomly which results in different trees each time. Random Forest works efficiently as each tree works individually and the random selection of features helps in avoiding overfitting and provides more accurate results. Also the error of an individual tree does not affect the output, as the final result is the average of the output of all the trees. It is observed that in case of increasing the number of trees in the Random Forest, the accuracy may increase upto a certain level and only computation cost is increased.
Taxi drivers’ traffic violations detection using random forest algorithm: A case study in China
Published in Traffic Injury Prevention, 2023
Ming Wan, Qian Wu, Lixin Yan, Junhua Guo, Wenxia Li, Wei Lin, Shan Lu
Random forest algorithm is widely adopted in the field of classification and recognition. Li et al. (2021) used a random forest algorithm to predict violation probability. It was proven to be the best traffic violation prediction model among logistic regression, Gaussian naive Bayes, and support vector machines. The area under the curve (AUC) and out-of-bag (OOB) error with IR = 1 reached 0.914 and 0.0787, showing the better performance of the random forest algorithm in dealing with imbalanced traffic violation data. This study adopted the RF model to predict taxi drivers’ traffic violations by involving multiple factors such as traffic violation time, road conditions, environment, and taxi companies. The experiments were set up using the scikit-learn library in Python. The dataset was randomly divided into a training set (70%) and a test set (30%) by using sklearn. model selection. train_test_split.
Integrating machine learning and network analytics to model project cost, time and quality performance
Published in Production Planning & Control, 2023
Shahadat Uddin, Stephen Ong, Haohui Lu, Petr Matous
Random forest (RF) uses a machine learning process that is built upon several decision trees (Breiman 2001). A tree-like structure is utilised where each node represents a test on the input attribute. The leaf or terminal nodes represent the decision outcomes. A classification outcome can be produced, which may be distinct and separate from the input vector. However, there may be a tendency for uneven weighting within this technique (Breiman 2001; Uddin et al. 2019). The advantages of random forest are that it is capable of effectively handling huge datasets, and the results are explainable. Figure 4 (d) shows three decision trees to illustrate the function of a random forest. The outcomes from trees 1, 2 and 3 are class B, class A and class A, respectively. According to the majority vote, the final prediction will be class A.
An interactive web-based solar energy prediction system using machine learning techniques
Published in Journal of Management Analytics, 2023
Priyanka Chawla, Jerry Zeyu Gao, Teng Gao, Chengchen Luo, Huimin Li, Yiqin We
The ensemble learning technique known as random forest is used for classification and regression applications. Numerous distinct decision trees make up a random forest. Each individual tree runs concurrently during training and prediction, using the class with the most votes as the final output of the model or the mean prediction (for regression) of the individual trees (Sönnichsen, 2020). One of the key benefits of random forest is that it generates trees that are not only trained on diverse datasets (with bagging) but also employ distinct features to build models, as opposed to training models on datasets with the same features. This ensures that random forest operates with high precision. As a result, random forest is effective in estimating solar radiation due to its regression capabilities and high accuracy while handling huge datasets. The random forest model trains many distinct trees using different features in the dataset after inputting six variables, and each model predicts different sun radiation levels. In order to have a more accurate result, we have taken average of thousands of predicted values and output this value as the final result.