Results 1  10
of
5,059
Neural network modeling for small datasets
, 2005
"... Neural network modeling for small datasets can be justified from a theoretical point of view according to some of Bartlett’s results showing that the generalization performance of a multilayer perceptron (MLP) depends more on the L1 norm �c�1 of the weights between the hidden layer and the output la ..."
Abstract

Cited by 2 (1 self)
 Add to MetaCart
Neural network modeling for small datasets can be justified from a theoretical point of view according to some of Bartlett’s results showing that the generalization performance of a multilayer perceptron (MLP) depends more on the L1 norm �c�1 of the weights between the hidden layer and the output
Bootstrapping statistical parsers from small datasets
 In Proceedings of the EACL
, 2003
"... We present a practical cotraining method for bootstrapping statistical parsers using a small amount of manually parsed training material and a much larger pool of raw sentences. Experimental results show that unlabelled sentences can be used to improve the performance of statistical parsers. In add ..."
Abstract

Cited by 65 (7 self)
 Add to MetaCart
We present a practical cotraining method for bootstrapping statistical parsers using a small amount of manually parsed training material and a much larger pool of raw sentences. Experimental results show that unlabelled sentences can be used to improve the performance of statistical parsers
Learning SinglyRecursive Relations from Small Datasets
 In F. Bergadano, L. De Raedt, S. Matwin, & S. Muggleton (Eds.), IJCAI93WS
, 1993
"... The inductive logic programming system LOPSTER was created to demonstrate the advantage of basing induction on logical implication rather than `subsumption. LOPSTER's subunification procedures allow it to induce recursive relations using a minimum number of examples, whereas inductive logic ..."
Abstract

Cited by 3 (1 self)
 Add to MetaCart
The inductive logic programming system LOPSTER was created to demonstrate the advantage of basing induction on logical implication rather than `subsumption. LOPSTER's subunification procedures allow it to induce recursive relations using a minimum number of examples, whereas inductive logic programming algorithms based on `subsumption require many more examples to solve induction tasks. However, LOPSTER's input examples must be carefully chosen; they must be along the same inverse resolution path. We hypothesize that an extension of LOPSTER can efficiently induce recursive relations without this requirement. We introduce a generalization of LOPSTER named CRUSTACEAN that has this capability and empirically evaluate its ability to induce recursive relations. 1 Introduction Several wellknown inductive logic programming (ILP) algorithms are based on ` subsumption (e.g., Muggleton & Feng, 1990; Quinlan, 1991; Pazzani & Kibler, 1992). When inducing recursive clauses, the input exam...
Summarization as Feature Selection for Document Categorization on Small Datasets
"... Abstract. Most common feature selection techniques for document categorization are supervised and require lots of training data in order to accurately capture the descriptive and discriminative information from the defined categories. Considering that training sets are extremely small in many classi ..."
Abstract
 Add to MetaCart
Abstract. Most common feature selection techniques for document categorization are supervised and require lots of training data in order to accurately capture the descriptive and discriminative information from the defined categories. Considering that training sets are extremely small in many
Irrelevant Features and the Subset Selection Problem
 MACHINE LEARNING: PROCEEDINGS OF THE ELEVENTH INTERNATIONAL
, 1994
"... We address the problem of finding a subset of features that allows a supervised induction algorithm to induce small highaccuracy concepts. We examine notions of relevance and irrelevance, and show that the definitions used in the machine learning literature do not adequately partition the features ..."
Abstract

Cited by 757 (26 self)
 Add to MetaCart
We address the problem of finding a subset of features that allows a supervised induction algorithm to induce small highaccuracy concepts. We examine notions of relevance and irrelevance, and show that the definitions used in the machine learning literature do not adequately partition the features
Learning generative visual models from few training examples: an incremental Bayesian approach tested on 101 object categories
, 2004
"... Abstract — Current computational approaches to learning visual object categories require thousands of training images, are slow, cannot learn in an incremental manner and cannot incorporate prior information into the learning process. In addition, no algorithm presented in the literature has been te ..."
Abstract

Cited by 784 (16 self)
 Add to MetaCart
tested on more than a handful of object categories. We present an method for learning object categories from just a few training images. It is quick and it uses prior information in a principled way. We test it on a dataset composed of images of objects belonging to 101 widely varied categories. Our
SMOTE: Synthetic Minority Oversampling Technique
 Journal of Artificial Intelligence Research
, 2002
"... An approach to the construction of classifiers from imbalanced datasets is described. A dataset is imbalanced if the classification categories are not approximately equally represented. Often realworld data sets are predominately composed of ``normal'' examples with only a small percentag ..."
Abstract

Cited by 634 (27 self)
 Add to MetaCart
An approach to the construction of classifiers from imbalanced datasets is described. A dataset is imbalanced if the classification categories are not approximately equally represented. Often realworld data sets are predominately composed of ``normal'' examples with only a small
Activity recognition from userannotated acceleration data
, 2004
"... In this work, algorithms are developed and evaluated to detect physical activities from data acquired using five small biaxial accelerometers worn simultaneously on different parts of the body. Acceleration data was collected from 20 subjects without researcher supervision or observation. Subjects ..."
Abstract

Cited by 515 (7 self)
 Add to MetaCart
In this work, algorithms are developed and evaluated to detect physical activities from data acquired using five small biaxial accelerometers worn simultaneously on different parts of the body. Acceleration data was collected from 20 subjects without researcher supervision or observation
Estimating the Support of a HighDimensional Distribution
, 1999
"... Suppose you are given some dataset drawn from an underlying probability distribution P and you want to estimate a "simple" subset S of input space such that the probability that a test point drawn from P lies outside of S is bounded by some a priori specified between 0 and 1. We propo ..."
Abstract

Cited by 783 (29 self)
 Add to MetaCart
Suppose you are given some dataset drawn from an underlying probability distribution P and you want to estimate a "simple" subset S of input space such that the probability that a test point drawn from P lies outside of S is bounded by some a priori specified between 0 and 1. We
PCA versus LDA
 IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
, 2001
"... In the context of the appearancebased paradigm for object recognition, it is generally believed that algorithms based on LDA (Linear Discriminant Analysis) are superior to those based on PCA (Principal Components Analysis) . In this communication we show that this is not always the case. We present ..."
Abstract

Cited by 472 (16 self)
 Add to MetaCart
present our case first by using intuitively plausible arguments and then by showing actual results on a face database. Our overall conclusion is that when the training dataset is small, PCA can outperform LDA, and also that PCA is less sensitive to different training datasets. Keywords: face recognition
Results 1  10
of
5,059