Quantitative Evaluation of Dependence among Outputs in ECOC Classifiers Using Mutual Information Based Measures
نویسندگان
چکیده
In previous work, it has been experimentally shown that the implementation of Error Correcting Output Coding (ECOC) classification methods with an ensemble of parallel and independent non linear dichotomizers (ECOC PND) outperforms the implementation with a single monolithic multi layer perceptron (ECOC MLP). This result was ascribed to the higher effectiveness of error correcting output coding methods for low correlation of the errors on different code bits. In this paper, we quantitatively evaluate the dependence of output errors in ECOC learning machines using mutual information based measures, and we study the relation between dependence of output errors and classification performances.
منابع مشابه
Dependence among Codeword Bits Errors in ECOC Learning Machines: An Experimental Analysis
One of the main factors affecting the effectiveness of ECOC methods for classification is the dependence among the errors of the computed codeword bits. We present an extensive experimental work for evaluating the dependence among output errors of the decomposition unit of ECOC learning machines. In particular, we compare the dependence between ECOC Multi Layer Perceptrons (ECOC monolithic), ma...
متن کاملAn experimental analysis of the dependence among codeword bit errors in ECOC learning machines
One of the main factors affecting the effectiveness of Error Correcting Output Coding (ECOC) methods for classification is the dependence among the errors of the computed codeword bits. We present an extensive experimental work for evaluating the dependence among output errors of the decomposition unit in ECOC learning machines. In particular, we apply measures based on mutual information to co...
متن کاملPerformance Evaluation of Closed Ended Mutual Funds in Pakistan
Mutual funds are the best tool to mobilize savings and investments in an economy and Pakistan is the pioneer in South Asia, but this industry is not as much mature in comparison to its age in Pakistan. This paper examines the performance of closed ended mutual funds in Pakistan by using five different ranking measures during a period of January 2009 to December 2013 and the sample consists of o...
متن کاملDeep Error-correcting Output Codes
Existing deep networks are generally initialized with unsupervised methods, such as random assignments and greedy layerwise pre-training. This may result in the whole training process (initialization/pre-training + fine-tuning) to be very timeconsuming. In this paper, we combine the ideas of ensemble learning and deep learning, and present a novel deep learning framework called deep error-corre...
متن کاملA comprehensive experimental comparison of the aggregation techniques for face recognition
In face recognition, one of the most important problems to tackle is a large amount of data and the redundancy of information contained in facial images. There are numerous approaches attempting to reduce this redundancy. One of them is information aggregation based on the results of classifiers built on selected facial areas being the most salient regions from the point of view of classificati...
متن کامل