/
info.json
15 lines (15 loc) · 988 Bytes
/
info.json
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
{
"abstract": "In binary classification problems it is common for\nthe two classes to be imbalanced: one case is very\nrare compared to the other. In this paper we consider\nthe infinitely imbalanced case where one class has a finite\nsample size and the other class's sample size grows without bound.\nFor logistic regression, the infinitely imbalanced case often\nhas a useful solution. Under mild conditions,\nthe intercept diverges as expected, but\nthe rest of the coefficient vector approaches a non trivial\nand useful limit.\nThat limit can be expressed in terms of exponential tilting\nand is the minimum of a convex objective function.\nThe limiting form of logistic regression suggests a computational\nshortcut for fraud detection problems.",
"authors": [
"Art B. Owen"
],
"id": "owen07a",
"issue": 27,
"pages": [
761,
773
],
"title": "Infinitely Imbalanced Logistic Regression",
"volume": "8",
"year": "2007"
}