On the use of classification reliability for improving performance of the one-per-class decomposition method

Iannello, Giulio; Percannella, Gennaro; Sansone, Carlo; Soda, Paolo

doi:10.1016/j.datak.2009.07.003

Cited by 18 publications

(8 citation statements)

References 39 publications

(67 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We present now two reconstruction rules for OpC: the first is a traditional implementation referred to as Hamming decoding (H d ), whereas the second is referred to as M DS rule and it has been introduced in [4]. In case of H d , the final decision is given by O(x) = ω s , with:…”

Section: Decomposition Methodsmentioning

confidence: 99%

“…This algorithm, intrinsically binary, requires the use of a decomposition method to cope with multiclass problems reducing these tasks into several binary subtasks. Then, a reconstruction rule provides the final classification [2][3][4]. Furthermore, it was proven that unn classifier can effectively estimates the posterior probability of each classification act [5].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Biomedical Images Classification by Universal Nearest Neighbours Classifier Using Posterior Probability

D'Ambrosio

Ali

Nock

et al. 2012

Machine Learning in Medical Imaging

Self Cite

View full text Add to dashboard Cite

Abstract. Universal Nearest Neighbours (unn) is a classifier recently proposed, which can also effectively estimates the posterior probability of each classification act. This algorithm, intrinsically binary, requires the use of a decomposition method to cope with multiclass problems, thus reducing their complexity in less complex binary subtasks. Then, a reconstruction rule provides the final classification. In this paper we show that the application of unn algorithm in conjunction with a reconstruction rule based on the posterior probabilities provides a classification scheme robust among different biomedical image datasets. To this aim, we compare unn performance with those achieved by Support Vector Machine with two different kernels and by a k Nearest Neighbours classifier, and applying two different reconstruction rules for each of the aforementioned classification paradigms. The results on one private and five public biomedical datasets show satisfactory performance.

show abstract

Section: Decomposition Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Biomedical Images Classification by Universal Nearest Neighbours Classifier Using Posterior Probability

D'Ambrosio

Ali

Nock

et al. 2012

Machine Learning in Medical Imaging

Self Cite

View full text Add to dashboard Cite

show abstract

“…= H(T ) H(T ;Y) , H(T ; Y) = − z p t (z) log 2 p y (z)22NI based on cross-entropy NI22 = H(Y) H(Y;T ) , H(Y; T ) = − z p y (z) log 2 p t (z)23 NI based on cross-entropy NI 23 = 1 cross-entropy NI24 = H(T )+H(Y) H(T ;Y)+H(Y;T ) Numerical examples in Binary Classifications(M1-M4 and M6: C 1 = 90, C 2 = 10; M5: C 1 = 95, C 2 = 5). (R)= ranking order for the model, where R = A,B, ..., in descending order from the top.…”

mentioning

confidence: 99%

Information-theoretic Measures for Objective Evaluation of Classifications

Hu¹,

He²,

Yuan³

2012

Acta Automatica Sinica

View full text Add to dashboard Cite

This work presents a systematic study of objective evaluations of abstaining classifications using Information-Theoretic Measures (ITMs). First, we define objective measures for which they do not depend on any free parameter. This definition provides technical simplicity for examining "objectivity" or "subjectivity" directly to classification evaluations. Second, we propose twenty four normalized ITMs, derived from either mutual information, divergence, or crossentropy, for investigation. Contrary to conventional performance measures that apply empirical formulas based on users' intuitions or preferences, the ITMs are theoretically more sound for realizing objective evaluations of classifications. We apply them to distinguish "error types" and "reject types" in binary classifications without the need for input data of cost terms. Third, to better understand and select the ITMs, we suggest three desirable features for classification assessment measures, which appear more crucial and appealing from the viewpoint of classification applications. Using these features as "meta-measures", we can reveal the advantages and limitations of ITMs from a higher level of evaluation knowledge. Numerical examples are given to corroborate our claims and compare the differences among the proposed measures. The best measure is selected in terms of the meta-measures, and its specific properties regarding error types and reject types are analytically derived.

show abstract

“…In activity classification, a false assignment could occur due to the unreliable nature of sensor data [17], incorrect execution of an activity [18], similar activities due to overlapping in features [19] or inability of a learning algorithm to assign the correct label [20]. In health care systems, the reliability of activity recognition models is extremely important, therefore along with the correct recognition of activities, a model should also be capable of detecting and avoiding false assignments [20]. Most of the activity recognition approaches while focusing on the segmentation and recognition may ignore false assignments [10,13,21].…”

Section: Introductionmentioning

confidence: 99%

“…Segmented activity instances are classified in activity classes using different learning models such as Hidden Markov Model (HMM) [10], Conditional Random Fields (CRF) [11], Naive Bayes (NB) [12], Support Vector Machine (SVM) [13], Artificial Neural Network (ANN) [14,15], and Decision Tree (DT) [16]. In activity classification, a false assignment could occur due to the unreliable nature of sensor data [17], incorrect execution of an activity [18], similar activities due to overlapping in features [19] or inability of a learning algorithm to assign the correct label [20]. In health care systems, the reliability of activity recognition models is extremely important, therefore along with the correct recognition of activities, a model should also be capable of detecting and avoiding false assignments [20].…”

Section: Introductionmentioning

confidence: 99%

Activity recognition in smart homes with self verification of assignments

2015

View full text Add to dashboard Cite

This is the accepted version of the paper.This version of the publication may differ from the final published version. Permanent repository link AbstractActivity recognition in smart homes provides valuable benefits in the field of health and elderly care by remote monitoring of patients. In health care, capabilities of both performing the correct recognition and reducing the wrong assignments are of high importance. The novelty of the proposed activity recognition approach lies in being able to assign a category to the incoming activity, while measuring the confidence score of the assigned category that reduces the false positives in the assignments. Multiple sensors deployed at different locations of a smart home are used for activity observations. For multi-class activity classification, we propose a binary solution using support vector machines, which simplifies the problem to correct/incorrect assignments. We obtain the confidence score of each assignment by estimating the activity distribution within each class such that the assignments with low confidence are separated for further investigation by a human operator. The proposed approach is evaluated using a comprehensive performance evaluation metrics. Experimental results obtained from nine publicly available smart home datasets demonstrate a better performance of the proposed approach compared to the state of the art.

show abstract

On the use of classification reliability for improving performance of the one-per-class decomposition method

Cited by 18 publications

References 39 publications

Biomedical Images Classification by Universal Nearest Neighbours Classifier Using Posterior Probability

Biomedical Images Classification by Universal Nearest Neighbours Classifier Using Posterior Probability

Information-theoretic Measures for Objective Evaluation of Classifications

Activity recognition in smart homes with self verification of assignments

Contact Info

Product

Resources

About