Boosting by weighting critical and erroneous samples

Gómez-Verdejo, Vanessa; Ortega-Moral, Manuel; Arenas‐García, Jerónimo; Figueiras-Vidal, Anı́bal R.

doi:10.1016/j.neucom.2005.12.011

Cited by 41 publications

(34 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…It uses Newton stepping rather than exact optimization at each boosting iteration. Another variant of Real AdaBoost -that uses a weighted emphasis function -is presented in [13], called Emphasis Boost. Finally, the Modest AdaBoost algorithm is mentioned here [40].…”

Section: Binary Classificationmentioning

confidence: 99%

A novel confidence-based multiclass boosting algorithm for mobile physical activity monitoring

Reiss

Hendeby

Stricker

2014

Pers Ubiquit Comput

View full text Add to dashboard Cite

This paper addresses one of the main challenges in physical activity monitoring, as indicated by recent benchmark results: The difficulty of the complex classification problems exceeds the potential of existing classifiers. Therefore, this paper proposes the ConfAdaBoost.M1 algorithm. This algorithm is a variant of the AdaBoost.M1 that incorporates well-established ideas for confidence-based boosting. ConfAdaBoost.M1 is compared to the most commonly used boosting methods using benchmark datasets from the UCI machine learning repository. Moreover, it is evaluated on an activity recognition and an intensity estimation problem, including a large number of physical activities from the recently released PAMAP2 dataset. The presented results indicate that the proposed ConfAdaBoost.M1 algorithm significantly improves the classification performance on most of the evaluated datasets, especially for larger and more complex classification tasks. Finally, two empirical studies are designed and carried out to investigate the feasibility of ConfAdaBoost.M1 for physical activity monitoring applications in mobile systems

show abstract

Section: Binary Classificationmentioning

confidence: 99%

A novel confidence-based multiclass boosting algorithm for mobile physical activity monitoring

Reiss

Hendeby

Stricker

2014

Pers Ubiquit Comput

View full text Add to dashboard Cite

show abstract

“…6 The numbering of the activities in the table corresponds to the activity IDs as given in the PAMAP2 dataset. The results are averaged over 10 test runs, the overall accuracy is 77.78%.…”

Section: Resultsmentioning

confidence: 99%

“…It uses Newton stepping rather than exact optimization at each boosting iteration. Further variants of binary AdaBoost are Emphasis Boost [6] and Modest AdaBoost [21].…”

Section: Binary Classificationmentioning

confidence: 99%

Confidence-based multiclass AdaBoost for physical activity monitoring

Reiss

Hendeby

Stricker

2013

Proceedings of the 2013 International Symposium on Wearable Computers

View full text Add to dashboard Cite

Physical activity monitoring has recently become an important topic in wearable computing, motivated by e.g. healthcare applications. However, new benchmark results show that the difficulty of the complex classification problems exceeds the potential of existing classifiers. Therefore, this paper proposes the ConfAdaBoost.M1 algorithm. The proposed algorithm is a variant of the AdaBoost.M1 that incorporates well established ideas for confidence based boosting. The method is compared to the most commonly used boosting methods using benchmark datasets from the UCI machine learning repository and it is also evaluated on an activity recognition and an intensity estimation problem, including a large number of physical activities from the recently released PAMAP2 dataset. The presented results indicate that the proposed ConfAdaBoost.M1 algorithm significantly improves the classification performance on most of the evaluated datasets, especially for larger and more complex classification tasks.

show abstract

“…1 (b), Boundary classifier has zero at the region close to a threshold. Then, Note that the region close to the threshold has a low rate of confidence, meaning that features in the boundary region may be classified incorrectly, since they are close to this threshold [11][12][13][14][15][16]. According to AdaBoost algorithm, Boosting method can be presented as the following strong classifier which combines a lot of weak classifies.…”

Section: Boundary Adaboostmentioning

confidence: 99%

Distance Sensitive AdaBoost using Distance Weight Function

Lee¹,

Cheon²,

Hyun³

et al. 2012

International Journal of Fuzzy Logic and Intelligent Systems

View full text Add to dashboard Cite

This paper proposes a new method to improve performance of AdaBoost by using a distance weight function to increase the accuracy of its machine learning processes. The proposed distance weight algorithm improves classification in areas where the original binary classifier is weak. This paper derives the new algorithm's optimal solution, and it demonstrates how classifier accuracy can be improved using the proposed Distance Sensitive AdaBoost in a simulation experiment of pedestrian detection.

show abstract

Boosting by weighting critical and erroneous samples

Cited by 41 publications

References 14 publications

A novel confidence-based multiclass boosting algorithm for mobile physical activity monitoring

A novel confidence-based multiclass boosting algorithm for mobile physical activity monitoring

Confidence-based multiclass AdaBoost for physical activity monitoring

Distance Sensitive AdaBoost using Distance Weight Function

Contact Info

Product

Resources

About