Academic literature on the topic 'C4.5 decision tree algorithm'

The iris dataset will be classified using the support vector machine and decision tree algorithms. flower dataset identifies the pattern and classifies it. The dataset has 150 rows and 5 attributes, which contains 50 samples from each species. There are three species in this dataset. Iris flower classification can be performed using support vector machines and decision tree algorithms. SVM stands for Support Vector Machine, and is a supervised machine learning technique that can be used for classification and regression. The Decision Tree algorithm is a simple approach mainly used for classification and prediction. The sample size has been determined to be 20 for both the groups using G Power 80%. The Support Vector Machine algorithm provides a mean accuracy of 98.09% when compared to the Decision Tree algorithm, with a mean accuracy of 95.55%. A statistically insignificant difference was observed between the Decision Tree and the Support Vector Machine, p = 0.92 (> 0.05) based on 2-tailed analysis. In the classification of Iris flowers, the Support Vector Machine outperformed the Decision Tree Algorithm.

This chapter presents a comprehensive comparative analysis of five distinct machine learning algorithms, namely Random Forest, Linear Regression, Bagging Regressor, AdaBoost Regressor, and Decision Tree Regressor, applied to the challenging task of stock price prediction. In an era marked by increasing reliance on data-driven decision-making, accurate stock price forecasting is of paramount importance for investors, traders, and financial analysts. The primary objective of this study was to identify the machine learning algorithm that exhibits superior predictive accuracy in the context of stock price forecasting. To achieve this goal, a meticulously curated dataset comprising historical stock price and relevant financial features was collected and preprocessed. Through a series of rigorous experiments and evaluations, this chapter examines and compares the algorithms' performance in terms of predictive accuracy, robustness, and computational efficiency.

Lenke classification has shown fair reliability when used by surgeons. Rule based algorithms have shown to improve AIS classification using King classification. No such algorithms have yet been developed for Lenke classification. A computer classifier using an algorithm based on a decision tree was developed to classify curve types of AIS spines according to Lenke classification strictly from Cobb angle measurements. A simplified and clinically usable diagram of that decision tree was designed and introduced to surgical residents, nurses and orthopaedic surgeons. They were asked to classify 36 scoliotic curves strictly from radiographic measurements, using the Lenke classification description from the original article and then using the decision tree diagram as well. Wilkoxon ranking test was used for statistical analysis of gain in accuracy and assessment time. Instant classification by the computer classifier achieved 99% accuracy on a 603 patient database. When comparing accuracy from classifying curve types using the Lenke system description alone and in complement with the decision tree diagram, classification accuracy mean (SD) of 81% (24%) and 94% (6%) were achieved respectively. That difference was statistically significant (p=0.01). Classification times mean (SD) without and with decision tree diagram were 31.66 (8.99) and 23.33 (9.69) seconds per item respectively. It was not statistically significant (p=0.10). Rule based algorithms and decision trees can improve AIS classification. The development of such algorithms can highlight ambiguous cases in classification description. Adaptation of algorithms developed for computer application to the clinical setting can increase clinicians' classification accuracy.

Recently machine learning based intrusion detection system developments have been subjected to extensive researches because they can detect both misuse detection and anomaly detection. In this paper, we propose an AdaBoost based algorithm for network intrusion detection system with single weak classifier. In this algorithm, the classifiers such as Bayes Net, Naïve Bayes and Decision tree are used as weak classifiers. KDDCup99 dataset is used in these experiments to demonstrate that boosting algorithm can greatly improve the classification accuracy of weak classification algorithms. Our approach achieves higher detection rate with low false alarm rates and is scalable for large datasets, resulting in an effective intrusion detection system.

With climate change, large, unpredictable and difficult to suppress forest fires are increasingly frequent. To increase the ability to anticipate and respond to these extreme events it is necessary to characterize the meteorological conditions associated with the risk levels of these events. The main objective of this work is to identify those conditions characterizing extreme forest fires in Portugal in the period 2001-2020 with at least 100ha burned area (90% percentile). The conditions characterizing the extreme fires are elicited by applying unsupervised fuzzy clustering and predictive methods to forest fire data and corresponding fire risk indices, namely the Canadian Forest Fire Risk Index (FWI), and subindices, as well as the Continuous Haines Index (CHI), provided by the Portuguese Institute of Sea and Atmosphere (IPMA). The dates and localization of fires are obtained from the shapefiles provided by the Portuguese Institute for Nature Conservation and Forests (ICNF), and complemented with data from the MODIS Global Burned Area Product MCD64A1 downloaded from the University of Maryland repository. The unsupervised fuzzy clustering algorithm (fuzzy c-means) is used for data classification and segmentation, and of the predictive model (decision trees), for weather characterization and extraction of rules. The fuzzy c-means was used to segment the data into 5 or 7 clusters, and to each cluster it is assigned the fire risk scale class of the cluster’s prototype, respectively the EEFIS scale (European-Forest-Fire Information System) for 5 clusters and IPMA fire risk scale for 7 clusters. Using the data from the 2001-2018, decision trees were induced and tested with the data from 2019 and 2020. To ensure the quality of its results, metrics and validation techniques such as cross-validation and bootstrapping are applied. From the experimental study, it is concluded that both the fuzzy c-means algorithm and the decision trees were effective in addressing the problem at hand. From the meteorological conditions, described by the fire risk indices, it was found that these were not always in agreement with the reference forest fire risk prediction scales, revealing the importance of adapting the indices values according to the region in question and taking into account several factors (forest fire risk indices) in the analysis of the conditions associated with the level of risk of an extreme forest fire. The proposed approach proved to be a proof of concept to test the applicability of this type of algorithm in this domain and to compare the results with the two fire risk scales used by IPMA and EEFIS.

Classification plays an important role in medicine, especially for medical diagnosis. Real-world medical applications often require classifiers that minimize the total cost, including costs for wrong diagnosis (misclassifications costs) and diagnostic test costs (attribute costs). There are indeed many reasons for considering costs in medicine, as diagnostic tests are not free and health budgets are limited. In this chapter, the authors have defined strategies for cost-sensitive learning. They have developed an algorithm for decision tree induction that considers various types of costs, including test costs, delayed costs and costs associated with risk. Then they have applied their strategy to train and to evaluate cost-sensitive decision trees in medical data. Generated trees can be tested following some strategies, including group costs, common costs, and individual costs. Using the factor of “risk” it is possible to penalize invasive or delayed tests and obtain patient-friendly decision trees.

Here, the researchers are trying to prepare a model for identifying whether a patient is diabetic or not. The Pima Indian Dataset has been used in this case study. There are two types of diabetes. The research consists of two stages. The first is data pre-processing, and the other is classifier construction. After pre-processing, the data classifier will be constructed which will predict whether the patient is diabetic or not. Here the researchers plan to use decision tree classifier and random tree classifier. After studying the dataset, the researchers handled the missing values in optimum ways. All the types of proposed algorithm have been described in this article.

The paper presents a database of published Y chromosome deletions and the results of analyzing the database with data mining and other heuristic techniques with the goal of developing a diagnostic test for male infertility. The database describes 382 patients for which 177 markers were tested. Two data mining techniques, clustering and decision tree induction were used, as well as a heuristic set cover algorithm. Clustering was used to group markers according to their appearance across patients, while a heuristic set covering algorithm was used to select as small a set of markers that cover as many patients with deletions as possible. This algorithm created a diagnostic set of 13 markers that cover more than 90% of the patients with deletions. Finally, decision tree induction was used to relate deletion patterns to the severity of the clinical phenotype. A decision tree induced from the data uses 5 markers, all of which are also in the diagnostic set of 13 markers, to show relations between the severity of the clinical phenotype and deletion patterns which have not been known previously.

Because of its relevance in the growth of tremendous applications in the medical area, data mining is a hugely important domain for exploration. When it comes to fatalities throughout the world, heart disease appears to be the leading cause of death. Recognizing a person’s risk of heart disease, it is a difficult assignment for health professionals since it takes very much time and extensive medical testing in the existing models. For the early prediction of cardiac disease, an improved version of the K-Nearest Neighbors (KNN) and Decision Tree (DT) classifier are applied, which ensures greater accuracy than existing models. Enhanced KNN is used for the correct treatment of patients and preserve consistency in the heart disease prediction system. We have calculated feature relevance, which assigns a score to independent factors based on how well they predict the difference walking. The variables difference walking, stroke and diabetes were the most relevant aspects, as shown in figure 2. We have compared to other existing models, the proposed model outperforms the existing models in terms of heart disease prediction. And the suggested work’s KNN algorithm hasthe greatest accuracy of 90.97 percentage and the highest ROC Curve of 69.97 percentage in predicting heart disease in comparison to Decision Tree (DT). The expenses and duration of treatment are reduced when the condition is detected early.