Combination Of Multiple Classifiers Research Articles

A framework for probabilistic combination of multiple classifiers at an abstract level

In previous approaches, the combination of multiple classifiers depends heavily on one of the three classification results; measurement scores (measurement level), ranking (rank level), and top choice (abstract level). For a more general combination of multiple classifiers, it is desirable that combination methods should be developed at the abstract level. In combining multiple classifiers at this level, most studies have assumed that classifiers behave independently. Such an assumption degrades and biases the classification performance, in cases where highly dependent classifiers are added. In order to overcome such weaknesses, it should be possible to combine multiple classifiers in a probabilistic framework, using a Bayesian formalism. A probabilistic combination of multiple decisions of K classifiers needs a ( K + 1)st-order probability distribution. However, it is well known that such a distribution will become unmanageable to store and estimate, even for a small K. In this paper, a framework is proposed to optimally identify a product set of kth-order dependencies, where 1≤ k⪯- K for the product approximation of the ( K + 1)st-order probability distribution from training samples, and to probabilistically combine multiple decisions by the identified product set, using the Bayesian formalism. This framework was tested and evaluated using a standardized CENPARMI data base. The results showed superior performance over other combination methods.

THE COMBINATION OF MULTIPLE CLASSIFIERS BY A NEURAL NETWORK APPROACH

Due to different writing styles and various kinds of noise, the recognition of handwritten numerals is an extremely complicated problem. Recently, a new trend has emerged to tackle this problem by the use of multiple classifiers. This method combines individual classification decisions to derive the final decisions. This is called "Combination of Multiple Classifiers" (CME). In this paper, a novel approach to CME is developed and discussed in detail. It contains two steps: data transformation and data classification. In data transformation, the output values of each classifier are first transformed into a form of likeness measurement. The larger a likeness measurement is, the more probable the corresponding class has the input. In data classification, neural networks have been found very suitable to aggregate the transformed output to produce the final classification decisions. Some strategies for further improving the performance of neural networks have also been proposed in this paper. Experiments with several data transformation functions and data classification approaches have been performed on a large number of handwritten samples. The best result among them is achieved by using both the proposed data transformation function and the multi-layer perceptron neural net, which increased the recognition rate of three individual classifications considerably.

Chemical mediation in bone resorption stimulated by orthodontic forces

In this paper, we present an ensemble learning approach for better understanding ridesplitting behavior of passengers of ridesourcing companies who provide prearranged and on-demand transportation services. An ensemble learning model is a weighted combination of multiple classification models or week classifiers to form a strong classification model. The goal of ensemble learning is to combine decisions or predictions of several base classifiers to improve prediction, generalizability, and robustness over a single classifier. This paper employs the Boosting ensemble by growing individual decision trees sequentially and then assembling these trees to produce a powerful classification model. To improve the prediction accuracy of ridesplitting choices, we explored real-world individual level data extracted from the on-demand ride service platform of DiDi in Hangzhou, China. Over one million trips of the four service types, i.e., Taxi Hailing Service, Express, Private Car Service, and Hitch, are explored with descriptive statistics. A variety of features that may impact ridesplitting behavior are ranked and selected by using the ReliefF algorithm, such as trip travel time, trip costs, trip length, waiting time fee, travel time reliability of origins/destinations and so on. The Boosting ensemble trees with full features and selected features are trained and validated using two independent datasets. This paper also verifies that ensemble learning is particularly useful and powerful in the ridesplitting analysis and outperforms three other widely used classifiers. This paper is one of the first quantitative studies that empirically reveal the real-world demand and supply pattern by exploring the city-wide data of an on-demand ride service platform.