Results 1  10
of
218
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 594 (23 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 into useful categories of relevance. We present definitions for irrelevance and for two degrees of relevance. These definitions improve our understanding of the behavior of previous subset selection algorithms, and help define the subset of features that should be sought. The features selected should depend not only on the features and the target concept, but also on the induction algorithm. We describe a method for feature subset selection using crossvalidation that is applicable to any induction algorithm, and discuss experiments conducted with ID3 and C4.5 on artificial and real datasets.
Bayesian Model Averaging for Linear Regression Models
 Journal of the American Statistical Association
, 1997
"... We consider the problem of accounting for model uncertainty in linear regression models. Conditioning on a single selected model ignores model uncertainty, and thus leads to the underestimation of uncertainty when making inferences about quantities of interest. A Bayesian solution to this problem in ..."
Abstract

Cited by 184 (13 self)
 Add to MetaCart
We consider the problem of accounting for model uncertainty in linear regression models. Conditioning on a single selected model ignores model uncertainty, and thus leads to the underestimation of uncertainty when making inferences about quantities of interest. A Bayesian solution to this problem involves averaging over all possible models (i.e., combinations of predictors) when making inferences about quantities of
Using Statistical Testing in the Evaluation of Retrieval Experiments
, 1993
"... The standard strategies for evaluation based on precision and recall are examined and their relative advantages and disadvantages are discussed. In particular, it is suggested that relevance feedback be evaluated from the perspective of the user. A number of different statistical tests are described ..."
Abstract

Cited by 172 (0 self)
 Add to MetaCart
The standard strategies for evaluation based on precision and recall are examined and their relative advantages and disadvantages are discussed. In particular, it is suggested that relevance feedback be evaluated from the perspective of the user. A number of different statistical tests are described for determining if differences in performance between retrieval methods are significant. These tests have often been ignored in the past because most are based on an assumption of normality which is not strictly valid for the standard performance measures. However, one can test this assumption using simple diagnostic plots, and if it is a poor approximation, there are a number of nonparametric alternatives.
A Comparison of Prediction Accuracy, Complexity, and Training Time of Thirtythree Old and New Classification Algorithms
, 2000
"... . Twentytwo decision tree, nine statistical, and two neural network algorithms are compared on thirtytwo datasets in terms of classication accuracy, training time, and (in the case of trees) number of leaves. Classication accuracy is measured by mean error rate and mean rank of error rate. Both cr ..."
Abstract

Cited by 167 (7 self)
 Add to MetaCart
. Twentytwo decision tree, nine statistical, and two neural network algorithms are compared on thirtytwo datasets in terms of classication accuracy, training time, and (in the case of trees) number of leaves. Classication accuracy is measured by mean error rate and mean rank of error rate. Both criteria place a statistical, splinebased, algorithm called Polyclass at the top, although it is not statistically signicantly dierent from twenty other algorithms. Another statistical algorithm, logistic regression, is second with respect to the two accuracy criteria. The most accurate decision tree algorithm is Quest with linear splits, which ranks fourth and fth, respectively. Although splinebased statistical algorithms tend to have good accuracy, they also require relatively long training times. Polyclass, for example, is third last in terms of median training time. It often requires hours of training compared to seconds for other algorithms. The Quest and logistic regression algor...
Motivational and selfregulated learning components of classroom academic performance
 Journal of Educational Psychology
, 1990
"... A correlational study examined relationships between motivational orientation, selfregulated learning, and classroom academic performance for 173 seventh graders from eight science and seven English classes. A selfreport measure of student selfefficacy, intrinsic value, test anxiety, selfregulat ..."
Abstract

Cited by 100 (1 self)
 Add to MetaCart
A correlational study examined relationships between motivational orientation, selfregulated learning, and classroom academic performance for 173 seventh graders from eight science and seven English classes. A selfreport measure of student selfefficacy, intrinsic value, test anxiety, selfregulation, and use of learning strategies was administered, and performance data were obtained from work on classroom assignments. Selfefficacy and intrinsic value were positively related to cognitive engagement and performance. Regression analyses revealed that, depending on the outcome measure, selfregulation, selfefficacy, and test anxiety emerged as the best predictors of performance. Intrinsic value did not have a direct influence on performance but was strongly related to selfregulation and cognitive strategy use, regardless of prior achievement level. The implications of individual differences in motivational orientation for cognitive engagement and selfregulation in the classroom are discussed. Selfregulation of cognition and behavior is an important aspect of student learning and academic performance in the classroom context (Corno & Mandinach, 1983; Corno & Rohrkemper, 1985). There are a variety of definitions of selfregulated learning, but three components seem especially important for classroom performance. First, selfregulated learning includes students ' metacognitive strategies for planning, monitoring, and modifying their cognition (e.g., Brown,
A MetaAnalysis of Rates of Return to Agricultural R&D  Ex Pede Herculem?
"... this report. Willis was a pioneer in the economic analysis of agricultural research and technical change, a teacher, and an inspirationas well as a friend and a good bloke ..."
Abstract

Cited by 71 (9 self)
 Add to MetaCart
this report. Willis was a pioneer in the economic analysis of agricultural research and technical change, a teacher, and an inspirationas well as a friend and a good bloke
Model Selection and Accounting for Model Uncertainty in Linear Regression Models
, 1993
"... We consider the problems of variable selection and accounting for model uncertainty in linear regression models. Conditioning on a single selected model ignores model uncertainty, and thus leads to the underestimation of uncertainty when making inferences about quantities of interest. The complete B ..."
Abstract

Cited by 47 (6 self)
 Add to MetaCart
We consider the problems of variable selection and accounting for model uncertainty in linear regression models. Conditioning on a single selected model ignores model uncertainty, and thus leads to the underestimation of uncertainty when making inferences about quantities of interest. The complete Bayesian solution to this problem involves averaging over all possible models when making inferences about quantities of interest. This approach is often not practical. In this paper we offer two alternative approaches. First we describe a Bayesian model selection algorithm called "Occam's "Window" which involves averaging over a reduced set of models. Second, we describe a Markov chain Monte Carlo approach which directly approximates the exact solution. Both these model averaging procedures provide better predictive performance than any single model which might reasonably have been selected. In the extreme case where there are many candidate predictors but there is no relationship between any of them and the response, standard variable selection procedures often choose some subset of variables that yields a high R² and a highly significant overall F value. We refer to this unfortunate phenomenon as "Freedman's Paradox" (Freedman, 1983). In this situation, Occam's vVindow usually indicates the null model as the only one to be considered, or else a small number of models including the null model, thus largely resolving the paradox.
Comparing Interactive Information Retrieval Systems Across Sites: The TREC6 Interactive Track Matrix Experiment
, 1998
"... This is a case study in the design and analysis of a 9site TREC6 experiment aimed at comparing the performance of 12 interactive information retrieval (IR) systems on a shared problem: a questionanswering task, 6 statements of information need, and a collection of 210,158 articles from the Financ ..."
Abstract

Cited by 42 (0 self)
 Add to MetaCart
This is a case study in the design and analysis of a 9site TREC6 experiment aimed at comparing the performance of 12 interactive information retrieval (IR) systems on a shared problem: a questionanswering task, 6 statements of information need, and a collection of 210,158 articles from the Financial Times of London 19911994. The study discusses the application of experimental design principles and the use of a shared control IR system in addressing the problems of comparing experimental interactive IR systems across sites: isolating the effects of topics, human searchers, and other sitespecific factors within an affordable design. The results confirm the dominance of the topic effect, show the searcher effect is almost as often absent as present, and indicate that for several sites the 2factor interactions are negligible. An analysis of variance found the system effect to be significant, but a multiple comparisons test found no significant pairwise differences. 1 Introduction T...
Contextprocessing deficits in schizophrenia: Converging evidence from three theoretically motivated cognitive tasks
 Journal of Abnormal Psychology
, 1999
"... Previous research on schizophrenia suggests that contextprocessing disturbances are one of the core cognitive deficits present in schizophrenia. However, it is not clear whether such deficits are specific to schizophrenia as compared with other psychotic disorders. To address this question, the aut ..."
Abstract

Cited by 39 (12 self)
 Add to MetaCart
Previous research on schizophrenia suggests that contextprocessing disturbances are one of the core cognitive deficits present in schizophrenia. However, it is not clear whether such deficits are specific to schizophrenia as compared with other psychotic disorders. To address this question, the authors administered a version of the AX Continuous Performance Test designed to assess context processing in a sample of healthy controls, patients with schizophrenia, and patients with other psychotic disorders. Participants were tested at index (when medication naive and experiencing their first contact with psychiatric services) and 4 weeks later, following medication treatment. At index, patients with schizophrenia and the psychotic comparison group demonstrated similar impairments in context processing. However, contextprocessing deficits improved in the psychotic comparison group at 4 weeks but did not improve in patients with schizophrenia. Editor’s Note. article.—TBB Gregory A. Miller served as the action editor for this
Feature subset selection as search with probabilistic estimates
 in Proc. AAAI Fall Symposium on Relevance
, 1994
"... Irrelevant features and weakly relevant features may reduce the comprehensibility and accuracy of concepts induced by supervised learning algorithms. We formulate search problem with probabilistic estimates. Searching a space using an evaluation function that is a random variable requires trading of ..."
Abstract

Cited by 37 (9 self)
 Add to MetaCart
Irrelevant features and weakly relevant features may reduce the comprehensibility and accuracy of concepts induced by supervised learning algorithms. We formulate search problem with probabilistic estimates. Searching a space using an evaluation function that is a random variable requires trading off accuracy of estimates for increased state exploration. We show how recent feature subset selection algorithms in the machine learning literature fit into this search problem as simple hill climbing approaches, and conduct a small experiment using a bestfirst search technique. 1