A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf
.
Filters
Mining Minority-class Examples With Uncertainty Estimates
[article]
2021
arXiv
pre-print
A promising solution is to mine tail-class examples to balance the training dataset. However, mining tail-class examples is a very challenging task. ...
Substantial improvements in the minority-class mining and fine-tuned model's performance strongly corroborate the value of our proposed solution. ...
Conclusion This study presents a novel, simple, and effective framework to mine minorityclass examples using uncertainty estimates. ...
arXiv:2112.07835v1
fatcat:ht75jqqrs5cttbnb3q3ebbq7yq
Why label when you can search?
2010
Proceedings of the 16th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '10
An alternative way to deploy human resources for training-data acquisition is to have them "guide" the learning by searching explicitly for training examples of each class. ...
This paper analyses alternative techniques for deploying lowcost human resources for data acquisition for classifier induction in domains exhibiting extreme class imbalance-where traditional labeling strategies ...
additional minority-class examples. ...
doi:10.1145/1835804.1835859
dblp:conf/kdd/AttenbergP10
fatcat:3v6v2ojhozagrgh7hhjqvix23q
Robust Learning at Noisy Labeled Medical Images: Applied to Skin Lesion Classification
[article]
2019
arXiv
pre-print
Specifically, an online uncertainty sample mining method is proposed to eliminate the disturbance from noisy-labeled images. ...
If with the noisy-labeled images, the training procedure will immediately encounter difficulties, leading to a suboptimal classifier. ...
Specifically, an online uncertainty sample mining strategy is proposed to suppress the noisy samples, and an individual re-weighting module is developed to preserve the hard samples and minority class. ...
arXiv:1901.07759v2
fatcat:qafh4fspwzfjxdy4idwi6vhc6y
Practical learning from one-sided feedback
2007
Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '07
In many data mining applications, online labeling feedback is only available for examples which were predicted to belong to the positive class. ...
Experimental results show that these methods can be significantly more effective in practice than those using the Apple Tasting transformation, even on minority class problems. ...
Our initial tests showed that cost weighting is necessary with all of the methods on minority class problems. ...
doi:10.1145/1281192.1281258
dblp:conf/kdd/Sculley07
fatcat:wwymra5hhjdlzbb3i2fmz45p3y
Comprehensive Accounting for REDD+ Programs: A Pragmatic Approach as Exemplified in Guyana
2020
Forests
This approach can be scaled to other countries with other activities that results in greenhouse gas emissions from deforestation and forest degradation. ...
Since submitting its FREL in 2014, Guyana has made stepwise improvements to its emission estimates so that the country is now able to report on all deforestation and degradation activities resulting in ...
Pete Watt, Danny Donoghue, and Towana Smartt were integral in the development of emissions estimates, providing estimates for area change. ...
doi:10.3390/f11121265
fatcat:oqesa4756zbdti3ujqcwnkvxqm
Mining with rarity
2004
SIGKDD Explorations
This article discusses the role that rare classes and rare cases play in data mining. ...
These descriptions utilize examples from existing research, so that this article provides a good survey of the literature on rarity in data mining. ...
on the impact of small disjuncts and class distribution on data mining. ...
doi:10.1145/1007730.1007734
fatcat:sb2tk62wrffifcirx75kw22etq
Uncertainty based under-sampling for learning Naive Bayes classifiers under imbalanced data sets
2019
IEEE Access
class. ...
Afterwards, it iteratively teaches its base model with the instances that the model is most uncertain about and retrains it until some criteria are satisfied. ...
Thus, it starts using all minority examples and only one from majority. ...
doi:10.1109/access.2019.2961784
fatcat:2tsiu2nd25evpbija44ymajfvm
Class Imbalance and Active Learning
[chapter]
2013
Imbalanced Learning
the deleterious effects of class imbalance, (iii) how extreme class imbalance can prevent AL systems from selecting useful examples, and alternatives to AL in these cases. ...
While using more examples in the training will often result in a better informed, more accurate model; limits on computer memory and real-world costs associated with gathering labeled examples often constrain ...
SMOTE oversamples the minority class by creating synthetic examples rather than with replacement. ...
doi:10.1002/9781118646106.ch6
fatcat:co355x5bxvdavnq6xy2g6iygfi
Inactive learning?
2011
SIGKDD Explorations
For example, derived estimates of generalization performance could be arbitrarily inaccurate. ...
Figure 3 : Comparison of random sampling and uncertainty sampling and guided learning on the problem seen in Figure 1 . represent the minority class instances, with the value on the left vertical axis ...
doi:10.1145/1964897.1964906
fatcat:cry3cqlesbd67aooh72njm7fsu
A General Framework for Mining Concept-Drifting Data Streams with Skewed Distributions
[chapter]
2007
Proceedings of the 2007 SIAM International Conference on Data Mining
We formally show some interesting and important properties of the proposed framework, e.g., reliability of estimated probabilities on skewed positive class, accuracy of estimated probabilities, efficiency ...
In this paper, we propose a new approach to mine data streams by estimating reliable posterior probabilities using an ensemble of models to match the distribution over under-samples of negatives and repeated ...
In some real applications, the class distribution is highly skewed, there are insufficient examples for minority class. ...
doi:10.1137/1.9781611972771.1
dblp:conf/sdm/GaoFHY07
fatcat:6ips4ugs2nfwjncopqk27tgn34
Sampling Strategies to Evaluate the Performance of Unknown Predictors
[chapter]
2012
Proceedings of the 2012 SIAM International Conference on Data Mining
Our goal is to design strategies for choosing examples such that they can be used to evaluate accurately a large set of classification models or rules one may want to experiment with, and not just one ...
Here are a few examples: ...
Blind Labeling In the previous subsections, we showed that by sampling from the minority class, we reduce the estimation uncertainty for the evaluation measures. ...
doi:10.1137/1.9781611972825.43
pmid:24955293
pmcid:PMC4063531
dblp:conf/sdm/ValizadeganAH12
fatcat:hmpddbis3zej5haaf64425rt7m
Foundations of Imbalanced Learning
[chapter]
2013
Imbalanced Learning
However, much of this research has focused on methods for dealing with imbalanced data, without discussing exactly how or why such methods work-or what underlying issues they address. ...
border regions with minority-class examples, figuring that they may be the
result of noise [34]. ...
Once this is done multiple training sets with
the desired class distribution can be formed using all minority-class examples
and a subset of the majority-class examples. ...
doi:10.1002/9781118646106.ch2
fatcat:opqe7dy2onaadp2ckacz6bdaxq
A Hybrid Feature Selection Method to Improve Performance of a Group of Classification Algorithms
2013
International Journal of Computer Applications
The experiments also show that this method outperforms other feature selection methods with a lower cost. ...
In this paper a hybrid feature selection method is proposed which takes advantages of wrapper subset evaluation with a lower cost and improves the performance of a group of classifiers. ...
The minority class is over-sampled by taking each minority class sample and introducing synthetic examples along the line segments joining any of the k minority class nearest neighbors. ...
doi:10.5120/12065-8172
fatcat:a2ahn3olrvehdgkr3aa3xsto54
Handling Inter-class and Intra-class Imbalance in Class-imbalanced Learning
[article]
2022
arXiv
pre-print
., noise removal, borderline sampling, hard example mining) but are still confined to a specific factor and cannot generalize to broader scenarios, which raises an interesting question: how to handle both ...
It features explicit and efficient inter-\&intra-class balancing as well as easy extension with standardized APIs. Extensive experiments validate the effectiveness of DuBE. ...
Hard example mining. Let's first introduce the hard example mining (HEM) considered in DUBE . ...
arXiv:2111.12791v2
fatcat:eqbhbosb3fasjo4k4r24iyxhyi
Estimating change in areas of indigenous vegetation cover in New Zealand from the New Zealand Land Cover Database (LCDB)
2017
New Zealand Journal of Ecology
Change in areas were estimated to within plus or minus 10% of change for classes with a large change in area, and to within plus or minus 30% for the classes with a small change in area. ...
We anticipate similar uncertainties for estimated changes in area between the other dates and for other classes. ...
So we expect the uncertainty of classes with an area greater than one million hectares to have an uncertainty of less than ±5%. ...
doi:10.20417/nzjecol.41.5
fatcat:wqwvq6va7bfylim74mxdzuaqyq
« Previous
Showing results 1 — 15 out of 46,940 results