Model Inference and Averaging

Hastie, Trevor; Friedman, J.; Tibshirani, Robert

doi:10.1007/978-0-387-21606-5_8

Cited by 799 publications

(1,337 citation statements)

References 0 publications

Supporting

Mentioning

1,315

Contrasting

Unclassified

Order By: Relevance

“…We want the model M to be successful in the prediction of the classes of unseen cases taken from the same distribution as D. A Bayesian view of the success of a classifier defines that the optimal classifier M Bayes maximizes the probability of predicting the correct class value g ∈ G for a given case x [11].…”

Section: The Problemmentioning

confidence: 99%

“…The prediction given by the set of resulting models for one example e is done by averaging the predictions of the different models. Bagging has the effect of improving the results of an unstable classifier by reducing its variance [11]. Domingos [9] suggests that, in the case of decision trees, bagging works because it increases the probability of choosing more complex models.…”

Section: Bagging Association Rules For Classificationmentioning

confidence: 99%

See 1 more Smart Citation

An Experiment with Association Rules and Classification: Post-Bagging and Conviction

Jorge

Azevedo

2005

Discovery Science

View full text Add to dashboard Cite

Abstract. In this paper we study a new technique we call post-bagging, which consists in resampling parts of a classification model rather then the data. We do this with a particular kind of model: large sets of classification association rules, and in combination with ordinary best rule and weighted voting approaches. We empirically evaluate the effects of the technique in terms of classification accuracy. We also discuss the predictive power of different metrics used for association rule mining, such as confidence, lift, conviction and χ 2 . We conclude that, for the described experimental conditions, post-bagging improves classification results and that the best metric is conviction.

show abstract

Section: The Problemmentioning

confidence: 99%

Section: Bagging Association Rules For Classificationmentioning

confidence: 99%

An Experiment with Association Rules and Classification: Post-Bagging and Conviction

Jorge

Azevedo

2005

Discovery Science

View full text Add to dashboard Cite

show abstract

“…In unsupervised methods, no drug combination labels are available, and the hidden structure of the given data is aimed to be extrapolated [59].…”

Section: Unsupervised Methodsmentioning

confidence: 99%

In-Silico Methodologies for Cancer Multidrug Optimization

Hasan¹,

Eldin

Khedr

et al. 2018

IJCT

View full text Add to dashboard Cite

Drug combinations is considered as an effective strategy designed to control complex diseases like cancer. Combinations of drugs can effectively decrease side effects and enhance adaptive resistance. Therefore, increasing the likelihood of defeating complex diseases in a synergistic way. This is due to overcoming factors such as off-target activities, network robustness, bypass mechanisms, cross-talk across compensatory escape pathways and the mutational heterogeneity which results in alterations within multiple molecular pathways. The plurality of effective drug combinations used in clinic were found out through experience. The molecular mechanisms underlying these drug combinations are often not clear. It is not easy to suggest new drug combinations. Computational approaches are proposed to reduce the search space for defining the most promising combinations and prioritizing their experimental evaluation. In this paper, we review methods, techniques and hypotheses developed for in silico methodologies for drug combination discovery in cancer and discuss the limitations and challenges of these methods.

show abstract

“…Now we present examples of granules obtained by application of a tolerance relation (i.e., reflexive and symmetric relation; for more information see, e.g. [43], and [11] for clustering methods based on similarity).…”

Section: Sets Of Granulesmentioning

confidence: 99%

Approximate Reasoning in Distributed Environments

Skowron

2004

Intelligent Technologies for Information Analysis

View full text Add to dashboard Cite

Abstract. Information sources provide us with granules of information that must be transformed, analyzed and built into structures that support problem solving. Lotfi A. Zadeh has recently pointed out to the need to develop a new research branch called Computing with Words. One way to achieve Computing with Words is through Granular Computing (GC). The main concepts of GC are related to information granule calculi. One of the main goals of information granule calculi is to develop algorithmic methods for construction of complex information granules from elementary ones by means of available operations and inclusion (closeness) measures. These constructions can also be interpreted as approximate schemes of reasoning (AR-schemes). These constructed complex granules represent a form of information fusion. Such constructed granules should satisfy some constraints like quality criteria or/and degrees of granule inclusion in (closeness to) a given information granule. In the paper we discuss the idea of the rough neurocomputing paradigm for inducing AR-schemes based on rough sets and, in particular, on rough mereology. Information granule decomposition methods are important components of methods for AR-schemes induced from data and background knowledge. We report some recent results on information granule decomposition.

show abstract

Model Inference and Averaging

Cited by 799 publications

References 0 publications

An Experiment with Association Rules and Classification: Post-Bagging and Conviction

An Experiment with Association Rules and Classification: Post-Bagging and Conviction

In-Silico Methodologies for Cancer Multidrug Optimization

Approximate Reasoning in Distributed Environments

Contact Info

Product

Resources

About