Filtering for improved gene selection on microarray data.

Evaluating scalable fuzzy clustering.

Ant Clustering Using Ensembles of Partitions.

Automatic Red Tide Detection from MODIS Satellite Images.



A Scalable Framework For Segmenting Magnetic Resonance Images.
A fast, accurate and fully automatic method of segmenting magnetic resonance images of the human brain is introduced. The approach scales well allowing fast segmentations of fine resolution images. The approach is based on modifications of the soft clustering algorithm, fuzzy c-means, that enable it to scale to large data sets. Two types of modifications to create incremental versions of fuzzy c-means are discussed. They are much faster when compared to fuzzy c-means for medium to extremely large data sets because they work on just parts of the data. They are comparable in quality to application of fuzzy c-means to all of the data. The clustering algorithms coupled with inhomogeneity correction and smoothing are used to create a framework for automatically segmenting magnetic resonance images of the human brain. The framework is applied to a set of normal human brain volumes acquired from different magnetic resonance scanners using different head coils, acquisition parameters and field strengths. Results are compared to those from two widely used magnetic resonance image segmentation programs, SPM and FSL. The results are comparable to FSL while providing significant speed-up and better scalability to larger volumes of data. 1
Automatically countering imbalance and its empirical relationship to cost.
Automatically countering imbalance and its empirical
Using classifier ensembles to label spatially disjoint data.
12 September 2007 Disk Used 11 We describe an ensemble approach to learning from arbitrarily partitioned data. The partitioning comes from the distributed process-12 ing requirements of a large scale simulation. The volume of the data is such that classifiers can train only on data local to a given par-13 tition. As a result of the partition reflecting the needs of the simulation, the class statistics can vary from partition to partition. Some 14 classes will likely be missing from some partitions. We combine a fast ensemble learning algorithm with probabilistic majority voting 15 in order to learn an accurate classifier from such data. Results from simulations of an impactor bar crushing a storage canister and from 16 facial feature recognition show that regions of interest are successfully identified in spite of the class imbalance in the individual training 17 sets.
Multivariate Feature Selection using Random Subspace Classifiers for Gene Expression Data.

Ensembles of Fuzzy Classifiers.
Abstract — The use of bagging is explored to create an ensemble of fuzzy classifiers. The learning algorithm used was ANFIS (Adaptive Neuro-Fuzzy Inference Systems). We compare results from bagging to those of a single classifier using both crisp and fuzzy classifier combination methods. Results on 20 data sets show that bagging results in a significantly more accurate classifier. I.
Single Pass Fuzzy C Means.
Abstract. Users of web search systems often have difficulty determining the relevance of search results to their information needs. Clustering has been suggested as a method for making this task easier. However, this introduces new challenges such as naming the clusters, selecting multiple clusters, and re-sorting the search results based on the cluster information. To address these challenges, we have developed Concept Highlighter, a tool for visually exploring concept-based fuzzy clusters in web search results. This tool automatically generates a set of concepts related to the users ’ queries, and performs single-pass fuzzy c-means clustering on the search results using these concepts as the cluster centroids. A visual interface is provided for interactively exploring the search results. In this paper, we describe the features of Concept Highlighter and its use in finding relevant documents within the search results through concept selection and document surrogate highlighting. 1
Noise%2DBased Feature Perturbation as a Selection Method for Microarray Data.
Abstract. DNA microarrays can monitor the expression levels of thousands of genes simultaneously, providing the opportunity for the identification of genes that are differentially expressed across different conditions. Microarray datasets are generally limited to a small number of samples with a large number of gene expressions, therefore feature selection becomes a very important aspect of the microarray classification problem. In this paper, a new feature selection method, feature perturbation by adding noise, is proposed to improve the performance of classification. The experimental results on a benchmark colon cancer dataset indicate that the proposed method can result in more accurate class predictions using a smaller set of features when compared to the SVM-RFE feature selection method. Key words: feature perturbation, microarray gene expression data, gene selection, classification 1
Boosting Lite %2D Handling Larger Datasets and Slower Base Classifiers.

Clinical deployment of a medical expert system to increase accruals for clinical trials Challenges.

A fuzzy c means variant for clustering evolving data streams.

A Comparison of Decision Tree Ensemble Creation Techniques.

Horizon Detection Using Machine Learning Techniques.



Learning to Predict Salient Regions from Disjoint and Skewed Training Sets.
We present an ensemble learning approach that achieves accurate predictions from arbitrarily partitioned data. The partitions come from the distributed processing requirements of a large scale simulation where the volume of the data is such that classifiers can train only on data local to a given partition. As a result of the partition reflecting the need for efficient simulation analysis, rather than the needs of data mining, the class statistics vary across partitions; indeed some classes will likely be absent from some partitions. We combine a fast ensemble learning algorithm with majority voting to generate an accurate working model of the simulation. Results from several simulations show that regions of interest are successfully identified in spite of training set class imbalances. Accuracy is analyzed both at the level of nodes in the simulation data structure, and in terms of higher-level regions of interest. It is shown that over 98 % of salient regions are found in independent test sets. Hence, this approach will be a significant time saver for simulation users and developers. 1.
Predicting Juvenile Diabetes from Clinical Test Results.

A constrained genetic approach for computing material property of elastic objects.
Abstract—This paper presents a constrained genetic approach for reconstructing the material properties of elastic objects. The considered reconstruction problem is ill-posed and must be constrained properly so that a unique and stable numerical solution can be obtained. Qualitative prior information is incorporated using a rank-based scheme to constrain the admissible solutions. Experiments show that the proposed approach is robust when presented with noisy data and can reconstruct the elastic property accurately and reliably. In a comparison study with the deterministic Gauss–Newton methods, the constrained genetic approach also shows very consistent performance. Index Terms—Constrained genetic algorithm (CGA), elastic property, finite-element model, inverse problem. I.
Scalable Swarm Based Fuzzy Clustering.

Scalable Swarm Based Fuzzy Clustering.

Bit Reduction Support Vector Machine.

Ensembles of Classifiers from Spatially Disjoint Data.
Abstract. We describe an ensemble learning approach that accurately learns from data which has been partitioned according to the arbitrary spatial requirements of a large-scale simulation wherein classifiers may be trained only the data local to a given partition. As a result, the class statistics can vary from partition to partition; some classes may even be missing from some partitions. In order to learn from such data, we combine a fast ensemble learning algorithm with Bayesian decision theory to generate an accurate working model of the simulation. Results from a simulation of an impactor bar crushing a storage canister and from region recognition in face images show that regions of interest are successfully identified. 1
Ensemble diversity measures and their application to thinning.
Abstract. The diversity of an ensemble can be calculated in a variety of ways. Here a diversity metric and a means for altering the diversity of an ensemble, called “thinning”, are introduced. We evaluate thinning algorithms on ensembles created by several techniques on 22 publicly available datasets. When compared to other methods, our percentage correct diversity measure algorithm shows a greater correlation between the increase in voted ensemble accuracy and the diversity value. Also, the analysis of different ensemble creation methods indicates each has varying levels of diversity. Finally, the methods proposed for thinning again show that ensembles can be made smaller without loss in accuracy.
Active Learning to Recognize Multiple Types of Plankton.
Active learning has been applied with support vector machines to reduce the data labeling effort in pattern recognition domains. However, most of those applications only deal with two class problems. In this paper, we extend the active learning approach to multiple class support vector machines. The experimental results from a plankton recognition system indicate that our approach often requires significantly less labeled images to maintain the same accuracy level as random sampling. 1.
Using Probabilistic Methods to Optimize Data Entry in Accrual of Patients to Clinical Trials.

Active Learning to Recognize Multiple Types of Plankton.
Active learning has been applied with support vector machines to reduce the data labeling effort in pattern recognition domains. However, most of those applications only deal with two class problems. In this paper, we extend the active learning approach to multiple class support vector machines. The experimental results from a plankton recognition system indicate that our approach often requires significantly less labeled images to maintain the same accuracy level as random sampling. 1.
A Comparison of Ensemble Creation Techniques.
We experimentally evaluate bagging and six other randomization-based approaches to creating an ensemble of decision-tree classifiers. Bagging uses randomization to create multiple training sets. Other approaches, such as Randomized C4.5 apply randomization in selecting a test at a given node of a tree. Then there are approaches, such as random forests and random subspaces, that apply randomization in the selection of attributes to be used in building the tree. On the other hand boosting, as compared here, incrementally builds classifiers by focusing on examples misclassified by existing classifiers. Experiments were performed on 34 publicly available data sets. While each of the other six approaches has some strengths, we find that none of them is consistently more accurate than standard bagging when tested for statistical significance. 1
Learning a model from spatially disjoint data.
Abstract- Some large-scale simulations are distributed over thousands of processors and generate terabytes of data. The output may take weeks or months to debug and explore. Therefore, learning a model that allows users to quickly focus on interesting events would be a great timesaver. Training data will not fit in one physical memory and the most natural splitting of the data into tractable size subsets would be along lines of disk farms, causing data for simulated objects to be split across training sets. In general, the training sets will contain very few interesting examples. A k nearest centroids approach was developed to classify unseen data. ROC analysis on a set of face images (partitioned spatially) indicates that this is a promising approach for the larger problem.
Decision trees work better than feed%2Dforward back%2Dprop neural nets for a specific class of problems.

Selection of patients for clinical trials an interactive web%2Dbased system.

Recognizing plankton images from the shadow image particle profiling evaluation recorder.
Abstract—We present a system to recognize underwater plankton images from the shadow image particle profiling evaluation recorder (SIPPER). The challenge of the SIPPER image set is that many images do not have clear contours. To address that, shape features that do not heavily depend on contour information were developed. A soft margin support vector machine (SVM) was used as the classifier. We developed a way to assign probability after multiclass SVM classification. Our approach achieved approximately 90 % accuracy on a collection of plankton images. On another larger image set containing manually unidentifiable particles, it also provided 75.6 % overall accuracy. The proposed approach was statistically significantly more accurate on the two data sets than a C4.5 decision tree and a cascade correlation neural network. The single SVM significantly outperformed ensembles of decision trees created by bagging and random forests on the smaller data set and was slightly better on the other data set. The 15-feature subset produced by our feature selection approach provided slightly better accuracy than using all 29 features. Our probability model gave us a reasonable rejection curve on the larger data set. Index Terms—Feature selection, learning, plankton recognition, probabilistic output, support vector machine (SVM). I.
List of Reviewers.

Learning Ensembles from Bites A Scalable and Accurate Approach.
Bagging and boosting are two popular ensemble methods that typically achieve better accuracy than a single classifier. These techniques have limitations on massive datasets, as the size of the dataset can be a bottleneck. Voting many classifiers built on small subsets of data (&#034;pasting small votes&#034;) is a promising approach for learning from massive datasets, one that can utilize the power of boosting and bagging. We propose a framework for building hundreds or thousands of such classifiers on small subsets of data in a distributed environment. Experiments show this approach is fast, accurate, and scalable.
Comments on %22A Parallel Mixture of SVMs for Very Large Scale Problems%22.
Collobert et. al. recently introduced a novel approach to using a neural network to provide a class prediction from an ensemble of support vector machines (SVMs). This approach has the advantage that the required computation scales well to very large data sets. Experiments on the Forest Cover data set show that this parallel mixture is more accurate than a single SVM, with 90.72 % accuracy reported on an independent test set. While this accuracy is impressive, the referenced paper does not consider alternative types of classifiers. In this comment, we show that a simple ensemble of decision trees results in a higher accuracy, 94.75%, and is computationally efficient. This result is somewhat surprising and illustrates the general value of experimental comparisons using different types of classifiers. 1 1
Comparing Pure Parallel Ensemble Creation Techniques Against Bagging.
We experimentally evaluate randomization-based approaches to creating an ensemble of decision-tree classifiers. Unlike methods related to boosting, all of the eight approaches considered here create each classifier in an ensemble independently of the other classifiers. Experiments were performed on 28 publicly available datasets, using C4.5 release 8 as the base classifier. While each of the other seven approaches has some strengths, we find that none of them is consistently more accurate than standard bagging when tested for statistical significance.
A New Ensemble Diversity Measure Applied to Thinning Ensembles.
Abstract. We introduce a new way of describing the diversity of an ensemble of classifiers, the Percentage Correct Diversity Measure, and compare it against existing methods. We then introduce two new methods for removing classifiers from an ensemble based on diversity calculations. Empirical results for twelve datasets from the UC Irvine repository show that diversity is generally modeled by our measure and ensembles can be made smaller without loss in accuracy. 1
SMOTEBoost Improving Prediction of the Minority Class in Boosting.
Abstract. Many real world data mining applications involve learning from imbalanced data sets, where the particular events of interest may be very few when compared to the other classes. Learning from data sets that contain rare events usually produces biased classifiers that have a higher predictive accuracy over the majority class(es), but poorer predictive accuracy over the minority class of interest. SMOTE (Synthetic Minority Over-sampling Technique) is a recent approach that is specifically designed for learning with minority classes. Boosting is a promising ensemble-based learning algorithm that can improve the classification performance of any weak classifier. This paper presents a novel approach for learning from rare classes, based on a combination of the SMOTE algorithm and the boosting procedure. Unlike standard boosting where all misclassified examples are given equal weights, the novel SMOTEBoost approach creates synthetic examples from the minority class, thus indirectly changing the updating weights and compensating for skewed distributions. The SMOTEBoost algorithm applied to several highly and moderately imbalanced data sets shows improvement in prediction performance on the rare class compared to standard boosting and to the SMOTE algorithm used with a single classifier. 1.
Is Error%2DBased Pruning Redeemable%3F

Generalization Methods in Bioinformatics.
Protein secondary structure prediction and high-throughput drug screen data mining are two important applications in bioinformatics. The data is represented in sparse feature spaces and can be unrepresentative of future data. Supervised learners in this context will display their inherent bias toward certain solutions, generally solutions that fit the training set well. In this paper, we first describe an ensemble approach using subsampling that scales well with dataset size. A sufficient number of ensemble members using subsamples of the data can yield a more accurate classifier than a single classifier using the entire dataset. Experiments on several datasets demonstrate the effectiveness of the approach. We report results from the KDD Cup 2001 drug discovery dataset in which our approach yields a higher weighted accuracy than the winning entry. We then extend our ensemble approach to create an over-generalized classifier for prediction by reducing the individual subsample size. The ensemble strategy using small subsamples has the effect of averaging over a wider range of hypotheses. Weshow that both protein secondary structure prediction and drug discovery prediction can be improved by the use of overgeneralization, specifically through the use of ensembles of small subsamples.

Error%2DBased Pruning of Decision Trees Grown on Very Large Data Sets Can Work%21
It has been asserted that, using traditional pruning methods, growing decision trees with increasingly larger amounts of training data will result in larger tree sizes even when accuracy does not increase. With regard to error-based pruning, the experimental data used to illustrate this assertion have apparently been obtained using the default setting for pruning strength; in particular, using the default certainty factor of 25 in the C4.5 decision tree implementation. We show that, in general, an appropriate setting of the certainty factor for error-based pruning will cause decision tree size to plateau when accuracy is not increasing with more training data. 1.
Distributed Pasting of Small Votes.
Bagging and boosting are two popular ensemble methods that achieve better accuracy than a single classifier. These techniques have limitations on massive datasets, as the size of the dataset can be a bottleneck. Voting many classifiers built on small subsets of data (&#034;pasting small votes&#034;) is a promising approach for learning from massive datasets. Pasting small votes can utilize the power of boosting and bagging, and potentially scale up to massive datasets. We propose a framework for building hundreds or thousands of such classifiers on small subsets of data in a distributed environment. Experiments show this approach is fast, accurate, and scalable to massive datasets.

SMOTE Synthetic Minority Over%2Dsampling Technique.
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 real-world data sets are predominately composed of ``normal&#039;&#039; examples with only a small percentage of ``abnormal&#039;&#039; or ``interesting&#039;&#039; examples. It is also the case that the cost of misclassifying an abnormal (interesting) example as a normal example is often much higher than the cost of the reverse error. Under-sampling of the majority (normal) class has been proposed as a good means of increasing the sensitivity of a classifier to the minority class. This paper shows that a combination of our method of over-sampling the minority (abnormal) class and under-sampling the majority (normal) class can achieve better classifier performance (in ROC space) than only under-sampling the majority class. This paper also shows that a combination of our method of over-sampling the minority class and under-sampling the majority class can achieve better classifier performance (in ROC space) than varying the loss ratios in Ripper or class priors in Naive Bayes. Our method of over-sampling the minority class involves creating synthetic minority class examples. Experiments are performed using C4.5, Ripper and a Naive Bayes classifier. The method is evaluated using the area under the Receiver Operating Characteristic curve (AUC) and the ROC convex hull strategy.
Investigation of bagging%2Dlike effects and decision trees versus neural nets in protein secondary structure prediction.

Bagging Is a Small%2DData%2DSet Phenomenon.
Bagging forms a committee of classifiers by bootstrap aggregation of training sets from a pool of training data. A simple alternative to bagging is to partition the data into disjoint subsets. Experiments on various datasets show that, given the same size partitions and bags, the use of disjoint partitions results in better performance than the use of bags. Many applications (e.g., protein structure prediction) involve the use of datasets that are too large to handle in the memory of the typical computer. Our results indicate that, in such applications, the simple approach of creating a committee of classifiers from disjoint partitions is to be preferred over the more complex approach of bagging.
Creating Ensembles of Classifiers.
Ensembles of classifiers offer promise in increasing overall classification accuracy. Combining multiple
Creating Ensembles of Classifiers.
Ensembles of classifiers offer promise in increasing overall classification accuracy. Combining multiple
Report of research activities in fuzzy AI and medicine at USF CSE.

Report of research activities in fuzzy AI and medicine at USF CSE.

Automatic segmentation of non%2Denhancing brain tumors in magnetic resonance images.
Tumor segmentation from magnetic resonance #MR# images may aid in tumor treatmentby  tracking the progress of tumor growth and#or shrinkage. In this paper we present the #rst automatic  segmentation method which separates non-enhancing brain tumors from healthy tissues in  MR images to aid in the task of tracking tumor size over time. The MR feature images used for  the segmentation consist of three weighted images #T1, T2 and proton density# for each axial slice  through the head. An initial segmentation is computed using an unsupervised fuzzy clustering  algorithm. Then, integrated domain knowledge and image processing techniques contribute to  the #nal tumor segmentation. They are applied under the control of a knowledge-based system.  The system knowledge was acquired by training on two patientvolumes #14 images#. Testing has  shown successful tumor segmentations on four patient volumes #31 images#. Our results show  that we detected all six non-enhancing brain tumors, located tumor tissue in 35 of the 36 ground  truth #radiologist labeled# slices containing tumor and successfully separated tumor regions from  physically connected CSF regions in nine of nine slices. Quantitative measurements are promising  as correspondence ratios between ground truth and segmented tumor regions ranged between  0.368 and 0.871 per volume, with percent match ranging between 0.530 and 0.909 per volume.  Keywords  MRI, non-enhancing brain tumors, image processing, automatic tissue classi#cation, fuzzy,  clustering  I. 
SMCB goes all electronic.

Finding Green River in SeaWiFS Satellite Images.
Understanding oceanic primary production on a global scale can be enhanced by methods that are able to automat-ically track phytoplankton blooms from color satellite im-ages. In this paper, unsupervised clustering and rule learn-ing are combined to track green river, a plume of discolored water that forms every March-May offshore along the edge of the west Florida Shelf, from the Sea Viewing Wide Field of View Sensor which beganjying in late 1997. Spatial in-formation and sea surface temperature can be integrated into the approach to improve pe$ormance. Using cross-validation experiments over a series of 59 multi-spectral im-ages, it is shown that the developed system is able to reliably discriminate between images with green river from those with no phytoplankton blooms or other kinds of blooms. It is also effective in identihing the region which the green river covers. 1.
Knowledge%2DGuided Classification of Coastal Zone Color Images Off the West Florida Shelf.
A knowledge-guided approach to automatic classification of Coastal Zone Color images of the West Florida Shelf is described. The approach is used to identify red tides on the West Florida Shelf, as well as areas with high concentration of dissolved organic matter such as a river plume found seasonally along the West Florida coast over the middle of the shelf. The Coastal Zone Color images are initially segmented by the unsupervised Multistage Random Sampling Fuzzy-c Means algorithm. Then, a knowledge-guided system is applied to the centroid values of resultant clusters to label case I, case II waters, a dilute river plume (&#034;green river&#034;), and red tide. The domain knowledge base contains information on cluster distribution in feature space, as well as spatial information such as bathymetry data. Our knowledge base consists of a rule-guided system and an embedded neural network. From 60 images, after training the system, this procedure recognizes all 15 images which contained a river plume and 45 images without. The system can correctly classify 74% of the pixels that belong to the river plume, which provides a substantial advantage to users looking for o#shore extensions of riverine influence. Red tides are also successfully identified in a time series of images for which ground truth confirmed the presence of a harmful bloom.
Learning Rules from Distributed Data.
In this paper a concern about the accuracy (as a function of parallelism) of a certain class of distributed learning algorithms is raised, and one proposed improvement is illustrated. We focus on learning a single model from a set of disjoint data sets, which are distributed across a set of computers. The model is a set of rules. The distributed data sets may be disjoint for any of several reasons. In our approach, the first step is to construct a rule set (model) for each of the original disjoint data sets. Then rule sets are merged until an eventual final rule set is obtained which models the aggregate data. We show that this approach compares to directly creating a rule set from the aggregate data and promises faster learning. Accuracy can drop off as the degree of parallelism increases. However, an approach has been developed to reduce this problem.
Clustering with a genetically optimized approach.
This paper describes a genetically guided approach to optimizing the hard (J1) and fuzzy (Jm) c-means functionals used in cluster analysis. Our experiments show that a genetic algorithm ameliorates the difficulty of choosing an initialization for the c-means clustering algorithms. Experiments use six data sets, including the Iris data, magnetic resonance and color images. The genetic algorithm approach is generally able to find the lowest known Jm value or a Jm associated with a partition very similar to that associated with the lowest Jm value. On data sets with several local extrema, the GA approach always avoids the less desirable solutions. Degenerate partitions are always avoided by the GA approach, which provides an effiective method for optimizing clustering models whose objective function can be represented in terms of cluster centers. The time cost of genetic guided clustering is shown to make a series random initializations of fuzzy/hard c-means, where the partition a...
SQFDiag semi%2Dquantitative model%2Dbased fault monitoring and diagnosis via episodic fuzzy rules.

A Qualitative Expert System for Clinical Trial Assignment.

Chemical Process Fault Diagnosis Using Kernel Retrofitted Fuzzy Genetic Algorithm Based Learner %28FGAL%29 with a Hidden Markov Model.

Generation of Fuzzy Rules from Decision Trees.
This paper introduces two methods of developing fuzzy rules, using decision trees, from data with continuous valued inputs and outputs. A key problem is how to deal with continuous outputs. Here output classes are created. A crisp decision tree may then be created using a set of fuzzy output classes allowing each training example to partially belong to the classes. Alternatively, a discrete set of fuzzy outputs classes can be created which include a selected group of overlaps, such as classA.75/classB.25. The training examples can then be provided to a standard decision tree learning program, such as C4.5. In both cases fuzzy rules will be extracted from the resultant decision tree. Output classes will have to be created for the case in which examples belong to discrete, but overlapping multiple classes. We discuss the tradeoffs of the two approaches to output class creation. An example of the systems performance using a discrete set of overlapping classes on the the Box-Jenkins gas furnace prediction problem and a function approximation problem are given. The learned rules are able to provide effective control and function approximation.
Automatic Tumor Segmentation Using Knowledge%2DBased Clustering.

Generation of Fuzzy Rules from Decision Trees.
This paper introduces two methods of developing fuzzy rules, using decision trees, from data with continuous valued inputs and outputs. A key problem is how to deal with continuous outputs. Here output classes are created. A crisp decision tree may then be created using a set of fuzzy output classes allowing each training example to partially belong to the classes. Alternatively, a discrete set of fuzzy outputs classes can be created which include a selected group of overlaps, such as classA.75/classB.25. The training examples can then be provided to a standard decision tree learning program, such as C4.5. In both cases fuzzy rules will be extracted from the resultant decision tree. Output classes will have to be created for the case in which examples belong to discrete, but overlapping multiple classes. We discuss the tradeoffs of the two approaches to output class creation. An example of the systems performance using a discrete set of overlapping classes on the the Box-Jenkins gas furnace prediction problem and a function approximation problem are given. The learned rules are able to provide effective control and function approximation.
An investigation of mountain method clustering for large data sets.

Partially supervised clustering for image segmentation.

Using Fuzzy Information in Knowledge Guided Segmentation of Brain Tumors.

Learning Membership Functions in a Function%2DBased Object Recognition System.
Functionality-based object recognition systems recognize objects at the category level by reasoning about how well the objects support the expected function of the category. Such systems naturally associate a &#034;measure of goodness&#034; or &#034;membership value&#034; with a recognized object. This measure of goodness is the result of combining individual measures, or membership values, from potentially many primitive evaluations of different properties of the object&#039;s shape. A membership function is used to compute the membership value for a primitive evaluation of a particular physical property of an object. In previous versions of a recognition system known as GRUFF, the membership function for each of the primitive evaluations had to be hand-crafted by the system designer. In this paper, we provide a learning component for the GRUFF system, called Omlet, that automatically learns primitive evaluation membership functions given a set of example objects labeled with their desired category measure. T...
Genetic Algorithm Guided Clustering.

Learning Fuzzy Membership Functions in a Function%2DBased Object Recognition System.

Methods for Combination of Evidence in Function%2DBased 3%2DD Object Recognition.

A Hybrid Connectionist%2C Symbolic Learning System.

SC%2Dnet A hybrid connectionist%2C symbolic system.

A Hybrid%2FSymbolic Connectionist Production System.

Backpac A parallel goal%2Ddriven reasoning system.

A Hybrid Connectionist%2C Symbolic Learning System.

On the choice of ply operators for modus ponens generation in fuzzy intelligent systems.

On the derivation of memberships for fuzzy sets in expert systems.

Relational Knowledge Acquisition.

