2019 IEEE 31st International Conference on Tools With Artificial Intelligence (ICTAI) 2019
DOI: 10.1109/ictai.2019.00141
|View full text |Cite
|
Sign up to set email alerts
|

Transfer Learning on Decision Tree with Class Imbalance

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
13
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
5
2

Relationship

1
6

Authors

Journals

citations
Cited by 11 publications
(13 citation statements)
references
References 17 publications
0
13
0
Order By: Relevance
“…For example, Al-Stouhi et al previously proposed that transfer learning can be used to solve class imbalance problems with inadequate data and provided theoretical and empirical validation on healthcare and text classification applications [ 10 ]. Minvielle et al explored the impact of class imbalance using transfer learning on decision trees [ 33 ]. However, only a few studies have been carried out on AMR so far.…”
Section: Discussionmentioning
confidence: 99%
See 1 more Smart Citation
“…For example, Al-Stouhi et al previously proposed that transfer learning can be used to solve class imbalance problems with inadequate data and provided theoretical and empirical validation on healthcare and text classification applications [ 10 ]. Minvielle et al explored the impact of class imbalance using transfer learning on decision trees [ 33 ]. However, only a few studies have been carried out on AMR so far.…”
Section: Discussionmentioning
confidence: 99%
“…Park et al used meta-transfer learning to explore the data heterogeneity and extremely small sample size problem based on single cell data [ 30 ]. Transfer learning is also widely used in the medical area with an imbalanced label [ 10 , 31 , 32 , 33 , 34 ]. For example, Gao et al used deep transfer learning to reduce healthcare disparities arising from imbalanced biomedical data [ 35 ].…”
Section: Introductionmentioning
confidence: 99%
“…If topology changes are considered, they can be included in the input features [16,32], such as those indicating whether transmission lines are available or not [16]. Alternatively, transfer learning on decision trees [33,34] can be adopted to modify the tree structure or thresholds.…”
Section: Model-based Solution Of Multiple Corrective Sced Runsmentioning
confidence: 99%
“…of testing instances × 100%. (33) Results are summarized in Table 5. CART has a TPR of 98.87% and a TIR of 94.17%, and 139 unpredicted contingencies from 85 instances (a contingency c may be counted multiple times as unpredicted from different instances).…”
Section: No Of Instances With All Predictedmentioning
confidence: 99%
“…Liu et al 27 treated synthetic minority samples as additional source data and extracted knowledge from them to enrich the target minority class. Minvielle et al 28 analyzed the impact of class imbalance on two TL methods based on decision trees, and proposed an improvement for each of them, termed as SER* ${\text{SER}}^{* }$ and STRUT* ${\text{STRUT}}^{* }$. Motivated by the superior performance of TL in handling imbalanced data sets, we present a transfer weighted‐ELM (TWELM)‐based class imbalance learning, in which a WELM output matrix is learned from the labeled data in source domain and treated as a regularization term to improve the performance of a target classifier.…”
Section: Introductionmentioning
confidence: 99%