logo
    Performance Analysis of Enhanced Adaboost Framework in Multifacet medical dataset
    9
    Citation
    0
    Reference
    20
    Related Paper
    Citation Trend
    Abstract:
    Predictions that are made based on features are performed through machine learning (ML) algorithms. Machine learning allows systems to learn and develop on their own by gaining experience. In the field of artificial intelligence, machine learning is a sub-discipline. Supervised and unsupervised learning are the two prevalent categories under machine learning. Supervised ML is used for classification whereas unsupervised ML is used for clustering. Currently, machine learning is being employed in a plethora of fields. Biometric recognition, handwriting recognition, and medical diagnosis are some of the use cases of ML. A significant role is played by machine learning in the medical field: identify diseases based on a patient's characteristics. Software applications based on ML algorithms are helping doctors in diagnosing various diseases like cancer, cardiac arrest, etc. We employed an ensemble learning strategy to predict heart problems in this paper. Through the comparison of different evaluation parameters namely ROC, F-measure, recall, precision and accuracy, our paper describes the performance of ML algorithms. The study used a mix of machine learning classifiers to predict heart problems, including Naive Bayes (NB), Decision Tree (DT), Random Forest (RF) and Support Vector Machine (SVM) algorithms. It was observed that implementing Paretto Distribution enabled adaboost resulted in 98.61% accuracy. NB, DT, RF and SVM models were also trained and tested separately.
    Keywords:
    AdaBoost
    Ensemble Learning
    The establishment of dataset threshold is one among the first steps when comparing the performance of machine learning algorithms. It involves the use of different datasets with different sample sizes in relation to the number of attributes and the number of instances available in the dataset. Currently, there is no limit which has been set for those who are unfamiliar with machine learning experiments on the categorisation of these datasets, as either small or large, based on the two factors. In this paper we perform experiments in order to establish dataset threshold. The established dataset threshold will help unfamiliar supervised machine learning experimenters to categorize datasets based on the number of instances and attributes and then choose the appropriate performance estimation method. The experiments will involve the use of four different datasets from UCI machine learning repository and two performance estimators. The performance of the methods will be measured using f1-score.
    Sample (material)
    Supervised Learning
    Machine learning algorithm integrates with various features and methods for learning. The performance metric has a major role in determining the capacity and ability of any machine learning algorithm. The supervised approach supports learning based on predefined parameters. The classification approach classifies the set of data with labels for training and evaluation purposes. This paper applies method of supervised learning algorithm based on classification approach to analyzed in terms of performance metrics. Some algorithms were applied in the experiment process to determine the clear outcome from them. The analysis gives an outcome based on certain parameters to obtain the results to understand the ability of these algorithm and their implementation in any potential use-cases.
    Weighted Majority Algorithm
    Statistical classification
    Supervised Learning
    Instance-based learning
    One-class classification
    Machine learning is a method in which computers are given the competence to acquire without being unambiguously programmed. Machine learning discovers the learning and structuring of algorithms that can learn from the past data and make predictions on the same. Methods for relating two or more algorithms on a single dataset have been inspected in the current scenario, comparison of algorithms on multiple datasets is even more crucial for a typical machine learning studies. In this paper I have discussed about the Kappa and Accuracy Evaluations of Machine Learning Classifiers on multiple datasets. The objective of this paper is to compare and analyze the execution of these algorithms based on the efficiency of machine learning algorithms such as Classification and Regression Tree (CART), Linear Discriminant Analysis (LDA), k-Nearest Neighbor (kNN), Support Vector Machine (SVM) and Random Forest.
    Online machine learning
    Learning classifier system
    Instance-based learning
    Kappa
    Statistical classification
    Cohen's kappa
    Citations (31)
    The large volume of data and its complexity in educational institutions require the sakes from informative technologies. In order to facilitate this task, many researchers have focused on using machine learning to extract knowledge from the education database to support students and instructors in getting better performance. In prediction models, the challenging task is to choose the effective techniques which could produce satisfying predictive accuracy. Hence, in this work, we introduced a hybrid approach of principal component analysis (PCA) as conjunction with four machines learning (ML) algorithms: random forest (RF), C5.0 of decision tree (DT), and naïve Bayes (NB) of Bayes network and support vector machine (SVM), to improve the performances of classification by solving the misclassification problem. Three datasets were used to confirm the robustness of the proposed models. Through the given datasets, we evaluated the classification accuracy and root mean square error (RSME) as evaluation metrics of the proposed models. In this classification problem, 10-fold cross-validation was proposed to evaluate the predictive performance. The proposed hybrid models produced very prediction results which shown itself as the optimal prediction and classification algorithms.
    Robustness
    Educational data mining (EDM) has become a very interesting field of study in machine learning (ML), since it has enabled searchers to mine knowledge from educational databases for improvement in students' and instructors' performance. The most challenging task in prediction is to identify which features and algorithms to select which will give satisfactory results. In this research, a hybrid algorithm of weighted voting classifier (WVC) in conjunction with 10-fold cross validation (10-CV) and five other machine learning algorithms that are support vector machine (SVM), multi-layer perceptron (MLP), logistic regression (LR), k-nearest neighbor (KNN) and naive bayes (NB) were used. We evaluated our proposed model on the student grade prediction dataset taken from kaggle. In this paper, the metrics that were measured included: accuracy, precision, recall, f1-score and area under the curve (AUC). An accuracy of 97.6% was achieved. The proposed model was able to identify 634 students out of 650 as (Fair, Good, and Excellent), therefore recommending the model to the school for student performance prediction since it will devise mechanisms to alleviate student dropout rates and improve their performance.
    Perceptron
    Educational Data Mining
    Multilayer perceptron
    Machine learning is concerned with the development, the analysis and the applications of algorithms that allow computers to learn. Supervised learning is the machine learning task of inferring a function from labelled training data. In this paper, we present a survey of various supervised classification techniques. The goal of this survey is to provide an inclusive review of different supervised classification techniques such as decision tree, Support Vector Machine, Naive Bayes, K-Nearest Neighbour, Neural Network.
    Statistical classification
    Supervised Learning
    One-class classification
    Citations (45)
    One of the most common tasks in machine learning is data classification. Machine learning emerges as a key feature to derive information from corporate operating datasets to large databases. Machine Learning in medical health care is evolving as a significant research field for delivering prognosis and a deeper understanding on medical data. Most methods of machine learning depend on several features defining the behavior of the algorithm, influencing the output, and the complexity of the resulting models either directly or indirectly. Many machine learning methods have been used in the past to detect heart diseases. Neural network and logistic regression are some of the few popular machine learning methods used in heart disease diagnosis. They analyze multiple algorithms such as neural network, K-nearest neighbor, naive bayes, and logistic regression along with composite approaches incorporating the aforementioned heart disease diagnostic algorithms. The system was implemented and trained in the python platform by using the UCI machine learning repository benchmark dataset. For the new data collection, the framework can be extended.
    Python
    Online machine learning
    Benchmark (surveying)
    Feature (linguistics)
    This paper proposes a comparative performance of ten different machine learning algorithms, done on a credit card fraud detection application. The machine learning methods have been classified into two groups namely classification algorithms and ensemble learning group. Each group is comprised of five different algorithms. Besides, the 'Time' feature is introduced in the data set and performances of the algorithms are studied with and without the 'Time' feature. Two algorithms of the ensemble learning group have been found to perform better when the used dataset does not include the 'Time' feature. However, for the classification algorithms group, three classifiers are found to show better predictive accuracies when all attributes are included in the used dataset. The rest of the machine learning models have approximate similar scores between these datasets.
    Feature (linguistics)
    Credit card fraud
    Ensemble Learning
    Citations (62)
    Machine learning is a subdivision of Artificial Intelligence (AI) that is concerned with the design and development of intelligent algorithms that enables machines to learn from data without being programmed. Machine learning mainly focus on how to automatically recognize complex patterns among data and make intelligent decisions. In this paper, intelligent machine learning algorithms are used to classify the type of an eye disease based on ophthalmology data collected from patients of Mecca hospital in Sudan. Three machine-learning techniques are used to predict the severity of the eye that occurred during the investigation, which are Naïve Bayesian, SVM, and J48 decision tree. The obtained result showed that J48 classifier outperforms both Naïve Bayesian as well as SVM.
    C4.5 algorithm