Vote-Based: Ensemble Approach

Abstract

Vote-based is one of the ensembles learning methods in which the individual classifier is situated on numerous weighted categories of the training datasets. In designing a method, training, validation and test sets are applied in terms of an ensemble approach to developing an efficient and robust binary classification model. Similarly, ensemble learning is the most prominent and broad research area of Machine Learning (ML) and image recognition, which assists in enhancing the capability of performance. In most cases, the ensemble learning algorithm yields better performance than ML algorithms. In this regard, numerous approaches had been studied significantly and used to accomplish better yields from the existing literature; however, the outcomes of these methods are inadequate. Unlike existing methods, the proposed technique aggregates an ensemble classifier, known as vote-based, to employ and integrate the advantage of ML classifiers, which are Naive Bayes (NB), Artificial Neural Network (ANN) and Logistic Model Tree (LMT). This paper proposes an ensemble framework that aims to evaluate datasets from the UCI ML repository by adopting performance analysis. The experimental consequences reveal that the intended approach outperforms than the conventional approaches. Furthermore, the experimental outputs indicate that the suggested method provides more accurate results according to the base learner approaches in terms of accuracy rates, an area under the curve (AUC), recall, precision, and F-measure values. This method can be used for binary classification, image recognition and machine learning problems.

Machine Learning Methods for Quality Prediction in Thermoplastics Injection Molding

“…99% is not an unreasonable value since in identical works these results were also obtained [3] [19], but for the industry, this 1% still represents a high number of failures. To go even further and with the observation that each method classifies correctly one type of NOK but not the other one, we designed a voting-based ensemble method technique [24] (with window solution). Our proposal is to create a classifier with both the contribution of ANN and SVM previous approaches.…”

A combined approach of base and meta learners for hybrid system

“…In such a scenario, certain features can significantly or negatively affect algorithms for classification accuracy. Therefore, the data assessments are normalized to the [0,1] range using the min-max normalization technique [23][24]. For mapping a value, of a feature xi from the range [min(xi ), max(xi)] to a new range [minxnew, maxxnew ], the normalized feature xȋ is computed as Eq.…”

“…In these subsections, we define and present the experimental procedure, measurements of evaluation and results of the experiment. All experiments are performed on base and metalearners by using WEKA (Waikato Environment for Knowledge Analysis) ML toolkit and JAVA programming language [24]. We have utilized default parameter values for all the classifiers in WEKA.…”

“…Precision is a positive analytical value [22,24]. Precision defines how reliable measurements are, although they are farther from the accepted value.…”

“…The recall is the hit rate [13,16,22,24]. The recall is the reverse of precision; it calculates false negatives against true positives.…”