Sciweavers

Free Online Productivity Tools i2Speak i2Symbol i2OCR iTex2Img iWeb2Print iWeb2Shot i2Type iPdf2Split iPdf2Merge i2Bopomofo i2Arabic i2Style i2Image i2PDF iLatex2Rtf Sci2ools

13

PAKDD
2004
ACM

favoriteEmaildiscussreport

113views Data Mining» more PAKDD 2004»

Logistic Regression and Boosting for Labeled Bags of Instances

13 years 10 months ago

Logistic Regression and Boosting for Labeled Bags of Instances

Download www.cs.waikato.ac.nz

Abstract. In this paper we upgrade linear logistic regression and boosting to multi-instance data, where each example consists of a labeled bag of instances. This is done by connecting predictions for individual instances to a bag-level probability estimate by simple averaging and maximizing the likelihood at the bag level—in other words, by assuming that all instances contribute equally and independently to a bag’s label. We present empirical results for artiﬁcial data generated according to the underlying generative model that we assume, and also show that the two algorithms produce competitive results on the Musk benchmark datasets.

Xin Xu, Eibe Frank

Real-time Traffic

Bag-level Probability Estimate | Data Mining | Individual Instances | Linear Logistic Regression | PAKDD 2004 |

claim paper

Related Content

» MultipleInstance Regression with Structured Data

» Pedestrian detection via logistic multiple instance boosting

» EndUser Feature Labeling via Locally Weighted Logistic Regression

» An empirical comparison of supervised learning algorithms

» Inverse Multiple Instance Learning for Classifier Grids

» Boosting and Maximum Likelihood for Exponential Models

» An Evaluation of Machine LearningBased Methods for Detection of Phishing Sites

» MICCLLR MultipleInstance Learning Using Class Conditional Log Likelihood Ratio

» Learning InstanceSpecific Predictive Models

Post Info
More Details (n/a)

Added	02 Jul 2010
Updated	02 Jul 2010
Type	Conference
Year	2004
Where	PAKDD
Authors	Xin Xu, Eibe Frank

Comments (0)