Sciweavers

56 search results - page 5 / 12
» New Transfer Learning Techniques for Disparate Label Sets
Sort
View
ICTAI
2006
IEEE
14 years 1 months ago
MI-Winnow: A New Multiple-Instance Learning Algorithm
We present MI-Winnow, a new multiple-instance learning (MIL) algorithm that provides a new technique to convert MIL data into standard supervised data. In MIL each example is a co...
Sharath R. Cholleti, Sally A. Goldman, Rouhollah R...
KDD
2004
ACM
164views Data Mining» more  KDD 2004»
14 years 8 months ago
Ordering patterns by combining opinions from multiple sources
Pattern ordering is an important task in data mining because the number of patterns extracted by standard data mining algorithms often exceeds our capacity to manually analyze the...
Pang-Ning Tan, Rong Jin
WEBI
2004
Springer
14 years 29 days ago
Co-training with a Single Natural Feature Set Applied to Email Classification
When dealing with information overload from the Internet, such as the classification of Web pages and the filtering of email spam, a new technique called cotraining has been shown...
Jason Chan, Irena Koprinska, Josiah Poon
ICIG
2009
IEEE
14 years 2 months ago
A PLSA-Based Semantic Bag Generator with Application to Natural Scene Classification under Multi-instance Multi-label Learning F
Classifying natural scenes into semantic categories has always been a challenging task. So far, many works in this field are primarily intended for single label classification, wh...
Shuangping Huang, Lianwen Jin
EMMCVPR
2011
Springer
12 years 7 months ago
Optimization of Robust Loss Functions for Weakly-Labeled Image Taxonomies: An ImageNet Case Study
The recently proposed ImageNet dataset consists of several million images, each annotated with a single object category. However, these annotations may be imperfect, in the sense t...
Julian John McAuley, Arnau Ramisa, Tibério ...