Melanie Hilario | Université de Genève (original) (raw)

Papers by Melanie Hilario

Mass spectrometry …, Jan 1, 2006

Among the many applications of mass spectrometry, biomarker pattern discovery from protein mass s... more Among the many applications of mass spectrometry, biomarker pattern discovery from protein mass spectra has aroused considerable interest in the past few years. While research efforts have raised hopes of early and less invasive diagnosis, they have also brought to light the many issues to be tackled before mass-spectra-based proteomic patterns become routine clinical tools. Known issues cover the entire pipeline leading from sample collection through mass spectrometry analytics to biomarker pattern extraction, validation, and interpretation. This study focuses on the data-analytical phase, which takes as input mass spectra of biological specimens and discovers patterns of peak masses and intensities that discriminate between different pathological states. We survey current work and investigate computational issues concerning the different stages of the knowledge discovery process: exploratory analysis, quality control, and diverse transforms of mass spectra, followed by further dimensionality reduction, classification, and model evaluation. We conclude after a brief discussion of the critical biomedical task of analyzing discovered discriminatory patterns to identify their component proteins as well as interpret and validate their biological implications. # 2006 Wiley Periodicals, Inc., Mass Spec Rev 25:409-449, 2006

At the crossroads of symbolic and neural processing, researchers have been actively investigating... more At the crossroads of symbolic and neural processing, researchers have been actively investigating the synergies that might be obtained from combining the strengths of these two paradigms. Neurosymbolic integration comes in two avors: unifed and hybrid. Uni ed approaches strive to attain full symbol-processing functionalities using neural techniques alone while hybrid approaches blend symbolic reasoning and representational models with neural networks. This papers attempts to clarify and compare the objectives, mechanisms, variants and underlying assumptions of these major integration approaches.

Knowledge and information systems, Jan 1, 2007

With the proliferation of extremely high-dimensional data, feature selection algorithms have beco... more With the proliferation of extremely high-dimensional data, feature selection algorithms have become indispensable components of the learning process. Strangely, despite extensive work on the stability of learning algorithms, the stability of feature selection algorithms has been relatively neglected. This study is an attempt to fill that gap by quantifying the sensitivity of feature selection algorithms to variations in the training set. We assess the stability of feature selection algorithms based on the stability of the feature preferences that they express in the form of weights-scores, ranks, or a selected feature subset. We examine a number of measures to quantify the stability of feature preferences and propose an empirical way to estimate them. We perform a series of experiments with several feature selection algorithms on a set of proteomics datasets. The experiments allow us to explore the merits of each stability measure and create stability profiles of the feature selection algorithms. Finally we show how stability profiles can support the choice of a feature selection algorithm.

Artificial Intelligence in …, Jan 1, 2006

Objective: An important problem that arises in hospitals is the monitoring and detection of nosoc... more Objective: An important problem that arises in hospitals is the monitoring and detection of nosocomial or hospital acquired infections (NIs). This paper describes a retrospective analysis of a prevalence survey of NIs done in the Geneva University Hospital. Our goal is to identify patients with one or more NIs on the basis of clinical and other data collected during the survey. Methods and material: Standard surveillance strategies are time-consuming and cannot be applied hospital-wide; alternative methods are required. In NI detection viewed as a classification task, the main difficulty resides in the significant imbalance between positive or infected (11%) and negative (89%) cases. To remedy class imbalance, we explore two distinct avenues: (1) a new resampling approach in which both oversampling of rare positives and undersampling of the noninfected majority rely on synthetic cases (prototypes) generated via class-specific subclustering, and (2) a support vector algorithm in which asymmetrical margins are tuned to improve recognition of rare positive cases. Results and conclusion: Experiments have shown both approaches to be effective for the NI detection problem. Our novel resampling strategies perform remarkably better than classical random resampling. However, they are outperformed by asymmetrical soft margin support vector machines which attained a sensitivity rate of 92%, significantly better than the highest sensitivity (87%) obtained via prototype-based resampling. #

Proteomics, Jan 1, 2003

We addressed the problem of discriminating between 24 diseased and 17 healthy specimens on the ba... more We addressed the problem of discriminating between 24 diseased and 17 healthy specimens on the basis of protein mass spectra. To prepare the data, we performed mass to charge ratio (m/z) normalization, baseline elimination, and conversion of absolute peak height measures to height ratios. After preprocessing, the major difficulty encountered was the extremely large number of variables (1676 m/z values) versus the number of examples (41). Dimensionality reduction was treated as an integral part of the classification process; variable selection was coupled with model construction in a single ten-fold cross-validation loop. We explored different experimental setups involving two peak height representations, two variable selection methods, and six induction algorithms, all on both the original 1676-mass data set and on a prescreened 124-mass data set. Highest predictive accuracies (1-2 off-sample misclassifications) were achieved by a multilayer perceptron and Naïve Bayes, with the latter displaying more consistent performance (hence greater reliability) over varying experimental conditions. We attempted to identify the most discriminant peaks (proteins) on the basis of scores assigned by the two variable selection methods and by neural network based sensitivity analysis. These three scoring schemes consistently ranked four peaks as the most relevant discriminators: 11683, 1403, 17350 and 66107.

…, Jan 1, 2004

In this paper we try to identify potential biomarkers for early stroke diagnosis using surfaceenh... more In this paper we try to identify potential biomarkers for early stroke diagnosis using surfaceenhanced laser desorption/ionization mass spectrometry coupled with analysis tools from machine learning and data mining. Data consist of 42 specimen samples, i.e., mass spectra divided in two big categories, stroke and control specimens. Among the stroke specimens two further categories exist that correspond to ischemic and hemorrhagic stroke; in this paper we limit our data analysis to discriminating between control and stroke specimens. We performed two suites of experiments. In the first one we simply applied a number of different machine learning algorithms; in the second one we have chosen the best performing algorithm as it was determined from the first phase and coupled it with a number of different feature selection methods. The reason for this was 2-fold, first to establish whether feature selection can indeed improve performance, which in our case it did not seem to confirm, but more importantly to acquire a small list of potentially interesting biomarkers. Of the different methods explored the most promising one was support vector machines which gave us high levels of sensitivity and specificity. Finally, by analyzing the models constructed by support vector machines we produced a small set of 13 features that could be used as potential biomarkers, and which exhibited good performance both in terms of sensitivity, specificity and model stability.

Machine Learning, Jan 1, 2004

In this paper we address two symmetrical issues, the discovery of similarities among classificati... more In this paper we address two symmetrical issues, the discovery of similarities among classification algorithms, and among datasets. Both on the basis of error measures, which we use to define the error correlation between two algorithms, and determine the relative performance of a list of algorithms. We use the first to discover similarities between learners, and both of them to discover similarities between datasets. The latter sketch maps on the dataset space. Regions within each map exhibit specific patterns of error correlation or relative performance. To acquire an understanding of the factors determining these regions we describe them using simple characteristics of the datasets. Descriptions of each region are given in terms of the distributions of dataset characteristics within it.

Tools with Artificial Intelligence, 2000. …, Jan 1, 2000

The selection of an appropriate inducer is crucial for performing effective classification. In pr... more The selection of an appropriate inducer is crucial for performing effective classification. In previous work we presented a system called NOEMON which relied on a mapping between dataset characteristics and inducer performance to propose inducers for specific datasets. Instancebased learning was used to create that mapping. Here we extend and refine the set of data characteristics; we also use a wider range of base-level inducers and a much larger collection of datasets to create the meta-models. We compare the performance of meta-models produced by instancebased learners, decision trees and boosted decision trees. The results show that decision trees and boosted decision trees models enhance the performance of the system.

Briefings in bioinformatics, Jan 1, 2008

Mass-spectra based proteomic profiles have received widespread attention as potential tools for b... more Mass-spectra based proteomic profiles have received widespread attention as potential tools for biomarker discovery and early disease diagnosis. A major data-analytical problem involved is the extremely high dimensionality (i.e. number of features or variables) of proteomic data, in particular when the sample size is small. This article reviews dimensionality reduction methods that have been used in proteomic biomarker studies. It then focuses on the problem of selecting the most appropriate method for a specific task or dataset, and proposes method combination as a potential alternative to single-method selection. Finally, it points out the potential of novel dimension reduction techniques, in particular those that incorporate domain knowledge through the use of informative priors or causal inference.

Proceedings International Symposium on …

MIX is an ESPRIT project aimed at developing strategies and tools for integrating symbolic and ne... more MIX is an ESPRIT project aimed at developing strategies and tools for integrating symbolic and neural methods in hybrid systems. The project arose from the observation that current hybrid systems are generally small-scale experimental systems which couple one symbolic and one connectionist model, often in an ad hoc fashion. Hence the objective of building a versatile testbed for the design, prototyping and assessment of a variety of hybrid models or architectures, in particular those which combine diverse neural network models with rule/model-based, cased-based, and fuzzy reasoning. A multiagent approach has been chosen to facilitate modular implementation of these hybrid models, which will be tested in the context of real-world applications in the steel and automobile industries.

Advances in Knowledge Discovery and Data …, Jan 1, 2001

Abstract. The selection of an appropriate inducer is crucial for per-forming effective classifica... more Abstract. The selection of an appropriate inducer is crucial for per-forming effective classification. In previous work we presented a system called NOEMON which relied on a mapping between dataset character-istics and inducer performance to propose inducers for ...

With the proliferation of extremely high-dimensional data, feature selection algorithms have beco... more With the proliferation of extremely high-dimensional data, feature selection algorithms have become indispensable components of the learning process. Strangely, despite extensive work on the stability of learning algorithms, the stability of feature selection algorithms has been relatively neglected. This study is an attempt to fill that gap by quantifying the sensitivity of feature selection algorithms to variations in the training set. We assess the stability of feature selection algorithms based on the stability of the feature preferences that they express in the form of weightsscores, ranks, or a selected feature subset. We examine a number of measures to quantify the stability of feature preferences and propose an empirical way to estimate them. We perform a series of experiments with several feature selection algorithms on a set of proteomics datasets. The experiments allow us to explore the merits of each stability measure and create stability profiles of the feature selection algorithms. Finally we show how stability profiles can support the choice of a feature selection algorithm.

MACHINE LEARNING-INTERNATIONAL …, Jan 1, 2003

To address the problem of algorithm selection for the classification task, we equip a relational ... more To address the problem of algorithm selection for the classification task, we equip a relational case base with new similarity measures that are able to cope with multirelational representations. The proposed approach builds on notions from clustering and is closely related to ideas developed in similarity-based relational learning. The results provide evidence that the relational representation coupled with the appropriate similarity measure can improve performance. The ideas presented are pertinent not only for meta-learning representational issues, but for all domains with similar representation requirements.

Principles of Data Mining and Knowledge …, Jan 1, 2001

Abstract. Meta-learning for model selection, as reported in the sym-bolic machine learning commun... more Abstract. Meta-learning for model selection, as reported in the sym-bolic machine learning community, can be described as follows. First, it is cast as a purely data-driven predictive task. Second, it typically relies on a mapping of dataset characteristics to some measure ...

Knowledge Discovery in …, Jan 1, 2004

Abstract. Protein fingerprints are groups of conserved motifs which can be used as diagnostic sig... more Abstract. Protein fingerprints are groups of conserved motifs which can be used as diagnostic signatures to identify and characterize collections of protein sequences. These fingerprints are stored in the prints database after time-consuming annotation by domain experts who ...

Proceedings of the 24th …, Jan 1, 2007

The k-Nearest Neighbors algorithm can be easily adapted to classify complex objects (e.g. sets, g... more The k-Nearest Neighbors algorithm can be easily adapted to classify complex objects (e.g. sets, graphs) as long as a proper dissimilarity function is given over an input space. Both the representation of the learning instances and the dissimilarity employed on that representation should be determined on the basis of domain knowledge. However, even in the presence of domain knowledge, it can be far from obvious which complex representation should be used or which dissimilarity should be applied on the chosen representation. In this paper we present a framework that allows to combine different complex representations of a given learning problem and/or different dissimilarities defined on these representations. We build on ideas developed previously on metric learning for vectorial data. We demonstrate the utility of our method in domains in which the learning instances are represented as sets of vectors by learning how to combine different set distance measures.

ENGINEERING INTELLIGENT SYSTEMS FOR …, Jan 1, 2000

We propose the use of learning algorithm profiles to address the model selection problem in knowl... more We propose the use of learning algorithm profiles to address the model selection problem in knowledge discovery systems. These profiles consist of metalevel feature-value vectors which describe learning algorithms from the point of view of their representation and functionality, efficiency, resilience, and practicality. Values for these features are assigned on the basis of author specifications, expert consensus or previous empirical studies. We review past evaluations of the better known learning algorithms and suggest an experimental strategy for building algorithm profiles on more quantitative grounds. Preliminary experiments have disconfirmed expert judgments on certain algorithm features, thus showing the need to build and refine such profiles via controlled experiments.

World Conference on the …, Jan 1, 1995

The ultimate|if implicit|goal of arti cial intelligence (AI) research is to model the full range ... more The ultimate|if implicit|goal of arti cial intelligence (AI) research is to model the full range of human cognitive capabilities. Symbolic AI and connectionism, the major AI paradigms, have each tried|and failed|to attain this goal. In the meantime, the idea has gained ground that this goal might still be within reach if we could harness the respective strengths of these two paradigms in integrated neurosymbolic models. This paper attempts to lay a cognitive basis for neurosymbolic integration and describes the di erent strategies that have been adopted to date. Uni ed approaches strive to attain symbol-processing capabilities using neural network techniques alone, while hybrid approaches blend symbolic and neural models in novel architectures with the hope of gleaning the best of both paradigms.

PAKDD, Jan 1, 2005

In this paper we present a novel and general framework based on concepts of relational algebra fo... more In this paper we present a novel and general framework based on concepts of relational algebra for kernel-based learning over relational schema. We exploit the notion of foreign keys to define a new attribute that we call instanceset and we use this type of attributes to define a tree like structured representation of the learning instances. We define kernel functions over relational schemata which are instances of R-Convolution kernels and use them as a basis for a relational instance-based learning algorithm. These kernels can be considered as being defined over typed and unordered trees where elementary kernels are used to compute the graded similarity between nodes. We investigate their formal properties and evaluate the performance of the relational instance-based algorithm on a number of relational benchmark datasets.

Mass spectrometry …, Jan 1, 2006

Among the many applications of mass spectrometry, biomarker pattern discovery from protein mass s... more Among the many applications of mass spectrometry, biomarker pattern discovery from protein mass spectra has aroused considerable interest in the past few years. While research efforts have raised hopes of early and less invasive diagnosis, they have also brought to light the many issues to be tackled before mass-spectra-based proteomic patterns become routine clinical tools. Known issues cover the entire pipeline leading from sample collection through mass spectrometry analytics to biomarker pattern extraction, validation, and interpretation. This study focuses on the data-analytical phase, which takes as input mass spectra of biological specimens and discovers patterns of peak masses and intensities that discriminate between different pathological states. We survey current work and investigate computational issues concerning the different stages of the knowledge discovery process: exploratory analysis, quality control, and diverse transforms of mass spectra, followed by further dimensionality reduction, classification, and model evaluation. We conclude after a brief discussion of the critical biomedical task of analyzing discovered discriminatory patterns to identify their component proteins as well as interpret and validate their biological implications. # 2006 Wiley Periodicals, Inc., Mass Spec Rev 25:409-449, 2006

At the crossroads of symbolic and neural processing, researchers have been actively investigating... more At the crossroads of symbolic and neural processing, researchers have been actively investigating the synergies that might be obtained from combining the strengths of these two paradigms. Neurosymbolic integration comes in two avors: unifed and hybrid. Uni ed approaches strive to attain full symbol-processing functionalities using neural techniques alone while hybrid approaches blend symbolic reasoning and representational models with neural networks. This papers attempts to clarify and compare the objectives, mechanisms, variants and underlying assumptions of these major integration approaches.

Knowledge and information systems, Jan 1, 2007

With the proliferation of extremely high-dimensional data, feature selection algorithms have beco... more With the proliferation of extremely high-dimensional data, feature selection algorithms have become indispensable components of the learning process. Strangely, despite extensive work on the stability of learning algorithms, the stability of feature selection algorithms has been relatively neglected. This study is an attempt to fill that gap by quantifying the sensitivity of feature selection algorithms to variations in the training set. We assess the stability of feature selection algorithms based on the stability of the feature preferences that they express in the form of weights-scores, ranks, or a selected feature subset. We examine a number of measures to quantify the stability of feature preferences and propose an empirical way to estimate them. We perform a series of experiments with several feature selection algorithms on a set of proteomics datasets. The experiments allow us to explore the merits of each stability measure and create stability profiles of the feature selection algorithms. Finally we show how stability profiles can support the choice of a feature selection algorithm.

Artificial Intelligence in …, Jan 1, 2006

Objective: An important problem that arises in hospitals is the monitoring and detection of nosoc... more Objective: An important problem that arises in hospitals is the monitoring and detection of nosocomial or hospital acquired infections (NIs). This paper describes a retrospective analysis of a prevalence survey of NIs done in the Geneva University Hospital. Our goal is to identify patients with one or more NIs on the basis of clinical and other data collected during the survey. Methods and material: Standard surveillance strategies are time-consuming and cannot be applied hospital-wide; alternative methods are required. In NI detection viewed as a classification task, the main difficulty resides in the significant imbalance between positive or infected (11%) and negative (89%) cases. To remedy class imbalance, we explore two distinct avenues: (1) a new resampling approach in which both oversampling of rare positives and undersampling of the noninfected majority rely on synthetic cases (prototypes) generated via class-specific subclustering, and (2) a support vector algorithm in which asymmetrical margins are tuned to improve recognition of rare positive cases. Results and conclusion: Experiments have shown both approaches to be effective for the NI detection problem. Our novel resampling strategies perform remarkably better than classical random resampling. However, they are outperformed by asymmetrical soft margin support vector machines which attained a sensitivity rate of 92%, significantly better than the highest sensitivity (87%) obtained via prototype-based resampling. #

Proteomics, Jan 1, 2003

We addressed the problem of discriminating between 24 diseased and 17 healthy specimens on the ba... more We addressed the problem of discriminating between 24 diseased and 17 healthy specimens on the basis of protein mass spectra. To prepare the data, we performed mass to charge ratio (m/z) normalization, baseline elimination, and conversion of absolute peak height measures to height ratios. After preprocessing, the major difficulty encountered was the extremely large number of variables (1676 m/z values) versus the number of examples (41). Dimensionality reduction was treated as an integral part of the classification process; variable selection was coupled with model construction in a single ten-fold cross-validation loop. We explored different experimental setups involving two peak height representations, two variable selection methods, and six induction algorithms, all on both the original 1676-mass data set and on a prescreened 124-mass data set. Highest predictive accuracies (1-2 off-sample misclassifications) were achieved by a multilayer perceptron and Naïve Bayes, with the latter displaying more consistent performance (hence greater reliability) over varying experimental conditions. We attempted to identify the most discriminant peaks (proteins) on the basis of scores assigned by the two variable selection methods and by neural network based sensitivity analysis. These three scoring schemes consistently ranked four peaks as the most relevant discriminators: 11683, 1403, 17350 and 66107.

…, Jan 1, 2004

In this paper we try to identify potential biomarkers for early stroke diagnosis using surfaceenh... more In this paper we try to identify potential biomarkers for early stroke diagnosis using surfaceenhanced laser desorption/ionization mass spectrometry coupled with analysis tools from machine learning and data mining. Data consist of 42 specimen samples, i.e., mass spectra divided in two big categories, stroke and control specimens. Among the stroke specimens two further categories exist that correspond to ischemic and hemorrhagic stroke; in this paper we limit our data analysis to discriminating between control and stroke specimens. We performed two suites of experiments. In the first one we simply applied a number of different machine learning algorithms; in the second one we have chosen the best performing algorithm as it was determined from the first phase and coupled it with a number of different feature selection methods. The reason for this was 2-fold, first to establish whether feature selection can indeed improve performance, which in our case it did not seem to confirm, but more importantly to acquire a small list of potentially interesting biomarkers. Of the different methods explored the most promising one was support vector machines which gave us high levels of sensitivity and specificity. Finally, by analyzing the models constructed by support vector machines we produced a small set of 13 features that could be used as potential biomarkers, and which exhibited good performance both in terms of sensitivity, specificity and model stability.

Machine Learning, Jan 1, 2004

In this paper we address two symmetrical issues, the discovery of similarities among classificati... more In this paper we address two symmetrical issues, the discovery of similarities among classification algorithms, and among datasets. Both on the basis of error measures, which we use to define the error correlation between two algorithms, and determine the relative performance of a list of algorithms. We use the first to discover similarities between learners, and both of them to discover similarities between datasets. The latter sketch maps on the dataset space. Regions within each map exhibit specific patterns of error correlation or relative performance. To acquire an understanding of the factors determining these regions we describe them using simple characteristics of the datasets. Descriptions of each region are given in terms of the distributions of dataset characteristics within it.

Tools with Artificial Intelligence, 2000. …, Jan 1, 2000

The selection of an appropriate inducer is crucial for performing effective classification. In pr... more The selection of an appropriate inducer is crucial for performing effective classification. In previous work we presented a system called NOEMON which relied on a mapping between dataset characteristics and inducer performance to propose inducers for specific datasets. Instancebased learning was used to create that mapping. Here we extend and refine the set of data characteristics; we also use a wider range of base-level inducers and a much larger collection of datasets to create the meta-models. We compare the performance of meta-models produced by instancebased learners, decision trees and boosted decision trees. The results show that decision trees and boosted decision trees models enhance the performance of the system.

Briefings in bioinformatics, Jan 1, 2008

Mass-spectra based proteomic profiles have received widespread attention as potential tools for b... more Mass-spectra based proteomic profiles have received widespread attention as potential tools for biomarker discovery and early disease diagnosis. A major data-analytical problem involved is the extremely high dimensionality (i.e. number of features or variables) of proteomic data, in particular when the sample size is small. This article reviews dimensionality reduction methods that have been used in proteomic biomarker studies. It then focuses on the problem of selecting the most appropriate method for a specific task or dataset, and proposes method combination as a potential alternative to single-method selection. Finally, it points out the potential of novel dimension reduction techniques, in particular those that incorporate domain knowledge through the use of informative priors or causal inference.

Proceedings International Symposium on …

MIX is an ESPRIT project aimed at developing strategies and tools for integrating symbolic and ne... more MIX is an ESPRIT project aimed at developing strategies and tools for integrating symbolic and neural methods in hybrid systems. The project arose from the observation that current hybrid systems are generally small-scale experimental systems which couple one symbolic and one connectionist model, often in an ad hoc fashion. Hence the objective of building a versatile testbed for the design, prototyping and assessment of a variety of hybrid models or architectures, in particular those which combine diverse neural network models with rule/model-based, cased-based, and fuzzy reasoning. A multiagent approach has been chosen to facilitate modular implementation of these hybrid models, which will be tested in the context of real-world applications in the steel and automobile industries.

Advances in Knowledge Discovery and Data …, Jan 1, 2001

Abstract. The selection of an appropriate inducer is crucial for per-forming effective classifica... more Abstract. The selection of an appropriate inducer is crucial for per-forming effective classification. In previous work we presented a system called NOEMON which relied on a mapping between dataset character-istics and inducer performance to propose inducers for ...

With the proliferation of extremely high-dimensional data, feature selection algorithms have beco... more With the proliferation of extremely high-dimensional data, feature selection algorithms have become indispensable components of the learning process. Strangely, despite extensive work on the stability of learning algorithms, the stability of feature selection algorithms has been relatively neglected. This study is an attempt to fill that gap by quantifying the sensitivity of feature selection algorithms to variations in the training set. We assess the stability of feature selection algorithms based on the stability of the feature preferences that they express in the form of weightsscores, ranks, or a selected feature subset. We examine a number of measures to quantify the stability of feature preferences and propose an empirical way to estimate them. We perform a series of experiments with several feature selection algorithms on a set of proteomics datasets. The experiments allow us to explore the merits of each stability measure and create stability profiles of the feature selection algorithms. Finally we show how stability profiles can support the choice of a feature selection algorithm.

MACHINE LEARNING-INTERNATIONAL …, Jan 1, 2003

To address the problem of algorithm selection for the classification task, we equip a relational ... more To address the problem of algorithm selection for the classification task, we equip a relational case base with new similarity measures that are able to cope with multirelational representations. The proposed approach builds on notions from clustering and is closely related to ideas developed in similarity-based relational learning. The results provide evidence that the relational representation coupled with the appropriate similarity measure can improve performance. The ideas presented are pertinent not only for meta-learning representational issues, but for all domains with similar representation requirements.

Principles of Data Mining and Knowledge …, Jan 1, 2001

Abstract. Meta-learning for model selection, as reported in the sym-bolic machine learning commun... more Abstract. Meta-learning for model selection, as reported in the sym-bolic machine learning community, can be described as follows. First, it is cast as a purely data-driven predictive task. Second, it typically relies on a mapping of dataset characteristics to some measure ...

Knowledge Discovery in …, Jan 1, 2004

Abstract. Protein fingerprints are groups of conserved motifs which can be used as diagnostic sig... more Abstract. Protein fingerprints are groups of conserved motifs which can be used as diagnostic signatures to identify and characterize collections of protein sequences. These fingerprints are stored in the prints database after time-consuming annotation by domain experts who ...

Proceedings of the 24th …, Jan 1, 2007

The k-Nearest Neighbors algorithm can be easily adapted to classify complex objects (e.g. sets, g... more The k-Nearest Neighbors algorithm can be easily adapted to classify complex objects (e.g. sets, graphs) as long as a proper dissimilarity function is given over an input space. Both the representation of the learning instances and the dissimilarity employed on that representation should be determined on the basis of domain knowledge. However, even in the presence of domain knowledge, it can be far from obvious which complex representation should be used or which dissimilarity should be applied on the chosen representation. In this paper we present a framework that allows to combine different complex representations of a given learning problem and/or different dissimilarities defined on these representations. We build on ideas developed previously on metric learning for vectorial data. We demonstrate the utility of our method in domains in which the learning instances are represented as sets of vectors by learning how to combine different set distance measures.

ENGINEERING INTELLIGENT SYSTEMS FOR …, Jan 1, 2000

We propose the use of learning algorithm profiles to address the model selection problem in knowl... more We propose the use of learning algorithm profiles to address the model selection problem in knowledge discovery systems. These profiles consist of metalevel feature-value vectors which describe learning algorithms from the point of view of their representation and functionality, efficiency, resilience, and practicality. Values for these features are assigned on the basis of author specifications, expert consensus or previous empirical studies. We review past evaluations of the better known learning algorithms and suggest an experimental strategy for building algorithm profiles on more quantitative grounds. Preliminary experiments have disconfirmed expert judgments on certain algorithm features, thus showing the need to build and refine such profiles via controlled experiments.

World Conference on the …, Jan 1, 1995

The ultimate|if implicit|goal of arti cial intelligence (AI) research is to model the full range ... more The ultimate|if implicit|goal of arti cial intelligence (AI) research is to model the full range of human cognitive capabilities. Symbolic AI and connectionism, the major AI paradigms, have each tried|and failed|to attain this goal. In the meantime, the idea has gained ground that this goal might still be within reach if we could harness the respective strengths of these two paradigms in integrated neurosymbolic models. This paper attempts to lay a cognitive basis for neurosymbolic integration and describes the di erent strategies that have been adopted to date. Uni ed approaches strive to attain symbol-processing capabilities using neural network techniques alone, while hybrid approaches blend symbolic and neural models in novel architectures with the hope of gleaning the best of both paradigms.

PAKDD, Jan 1, 2005

In this paper we present a novel and general framework based on concepts of relational algebra fo... more In this paper we present a novel and general framework based on concepts of relational algebra for kernel-based learning over relational schema. We exploit the notion of foreign keys to define a new attribute that we call instanceset and we use this type of attributes to define a tree like structured representation of the learning instances. We define kernel functions over relational schemata which are instances of R-Convolution kernels and use them as a basis for a relational instance-based learning algorithm. These kernels can be considered as being defined over typed and unordered trees where elementary kernels are used to compute the graded similarity between nodes. We investigate their formal properties and evaluate the performance of the relational instance-based algorithm on a number of relational benchmark datasets.