Gregor Stiglic
University of Maribor, Faculty of Health Sciences, Faculty Member
ABSTRACT Increasing demand for digitalization of Electronic Health Records results in increased demand for effective data mining solutions. In this study we enhance the classical Support Vector Machine-Recursive Feature Elimination... more
ABSTRACT Increasing demand for digitalization of Electronic Health Records results in increased demand for effective data mining solutions. In this study we enhance the classical Support Vector Machine-Recursive Feature Elimination (SVM-RFE) approach to optimally estimate disease risk from hospital discharge record data.
Purpose Classification is an important and widely used machine learning technique in bioinformatics. Researchers and other end-users of machine learning software often prefer to work with comprehensible models where knowledge extraction... more
Purpose Classification is an important and widely used machine learning technique in bioinformatics. Researchers and other end-users of machine learning software often prefer to work with comprehensible models where knowledge extraction and explanation of reasoning behind the classification model are possible. Methods This paper presents an extension to an existing machine learning environment and a study on visual tuning of decision tree classifiers.
Abstract Smart phones are increasingly being used to track and recognize different types of activity. However, the task of using smart phones to infer the intensity of sport activities has not received a lot of attention yet. Therefore,... more
Abstract Smart phones are increasingly being used to track and recognize different types of activity. However, the task of using smart phones to infer the intensity of sport activities has not received a lot of attention yet. Therefore, we study how off-the-shelf smart phones with built-in accelerometers can be used to estimate the intensity of recreational sport activities.
Summary: Often, the most informative genes have to be selected from different gene sets and several computer gene ranking algorithms have been developed to cope with the problem. To help researchers decide which algorithm to use, we... more
Summary: Often, the most informative genes have to be selected from different gene sets and several computer gene ranking algorithms have been developed to cope with the problem. To help researchers decide which algorithm to use, we developed the analysis of gene ranking algorithms (AGRA) system that offers a novel technique for comparing ranked lists of genes. The most important feature of AGRA is that no previous knowledge of gene ranking algorithms is needed for their comparison. Using the text mining system finding-associated concepts with text analysis. AGRA defines what we call biomedical concept space (BCS) for each gene list and offers a comparison of the gene lists in six different BCS categories. The uploaded gene lists can be compared using two different methods. In the first method, the overlap between each pair of two gene lists of BCSs is calculated. The second method offers a text field where a specific biomedical concept can be entered. AGRA searches for this concept in each gene lists' BCS, highlights the rank of the concept and offers a visual representation of concepts ranked above and below it.Availability and Implementation: Available at http://agra.fzv.uni-mb.si/, implemented in Java and running on the Glassfish server.Contact: simon.kocbek/at/uni-mb.si<!-- try{initUnObscureEmail ("e_id562354", '<a href="' + reverseAndReplaceString('is.bm\-inu/ta/kebcok.nomis:otliam', '/at/', '@') + '">' + reverseAndReplaceString('is.bm\-inu/ta/kebcok.nomis', '/at/','@') + '</a>')}catch(e){} //-->
Research Interests:
Research Interests:
Research Interests: Knowledge Representation, Gene expression, Sample Preparation, Quality Control, Next generation sequencing, and 18 moreBiological Sciences, New Technology, Mathematical Sciences, Statistical Techniques in Spatial Analysis, Stability Analysis, BMC Bioinformatics, Decision Tree, Gene Selection, Knowledge Extraction, Very high throughput, Reproducibility of Results, Gene Set Enrichment Analysis, Genome sequence, Gene Expression Analysis, Microarray Data, Gene expression profiling, Gene Expression Data, and Statistical techniques
Research Interests:
Research Interests: Artificial Intelligence, Biomedical Engineering, Machine Learning, Cardiac arrest, Humans, and 11 moreSupervised Learning, Intelligence Analysis, Cardiopulmonary Resuscitation, Theoretical Models, Sex Factors, Survival Rate, Prognosis, Age Factors, Advanced Life Support, Computer Methods, and Ventricular Fibrillation
Research Interests:
Research Interests:
Many advanced machine learning and statistical methods have recently been employed in classification of gene expression measurements. Although many of these methods can achieve high accuracy, they generally lack comprehensibility of the... more
Many advanced machine learning and statistical methods have recently been employed in classification of gene expression measurements. Although many of these methods can achieve high accuracy, they generally lack comprehensibility of the classification process. In this paper a new method for interpretation of small ensembles of classifiers is used on gene expression data from real-world dataset. It was shown that interactive interpretation systems that were developed for classical machine learning problems also give a great range of ...
Research Interests:
Research Interests:
Research Interests:
Research Interests:
... about data such as probability in probability theory or grade of membership in fuzzy set theory. ... Using rough sets method we extracted a large number of rules from which only a few ... of large number of samples in the database we... more
... about data such as probability in probability theory or grade of membership in fuzzy set theory. ... Using rough sets method we extracted a large number of rules from which only a few ... of large number of samples in the database we also tried sampling method which can be used in ...
Research Interests:
Abstract In this paper we propose a new method inspired by a multi-agent based system that was initially used for identification of significant genes in microarray databases. Gene subset selection is a common problem in the filed of... more
Abstract In this paper we propose a new method inspired by a multi-agent based system that was initially used for identification of significant genes in microarray databases. Gene subset selection is a common problem in the filed of bioinformatics. If we regard the software measurements values of a software module as a genome of that module, and the real world dynamic characteristic of that module as its phenotype (ie failures as a disease symptoms) we can borrow the established bioinformatics methods in the manner first to predict the ...
Research Interests:
The possibility for an expert to verify and evaluate a decision tree is the major advantage of using this machine learning method, especially for medical diagnostics. In this paper we explore the use of a machine learning method based on... more
The possibility for an expert to verify and evaluate a decision tree is the major advantage of using this machine learning method, especially for medical diagnostics. In this paper we explore the use of a machine learning method based on decision trees using CART for verifying clinically established diagnostic criteria and also for seeking new criteria in different autopsy-confirmed Parkinsonian disorders. Since differentiating various types of Parkinsonian disorders can often present great difficulties due to the overlapping of ...
Research Interests:
Abstract In usual modern hospital architectural design we try to separate the clean and dirty corridor that should facilitate traffic flow of clean and dirty items. However that is not the case in the hospitals with older architectural... more
Abstract In usual modern hospital architectural design we try to separate the clean and dirty corridor that should facilitate traffic flow of clean and dirty items. However that is not the case in the hospitals with older architectural design, where clean and dirty pathways flow through a single corridor systems. The most common solution which enables separation of clean and dirty pathways in such systems is introduction of time allocation for dirty and clean items transportation. Currently such schedules are fixed and do not allow quick changes in case ...
Research Interests:
Ensembles of classifiers have the ability to boost classification accuracy comparing to single classifiers and are a commonly used method in the field of machine learning. However in some cases ensemble construction algorithms do not... more
Ensembles of classifiers have the ability to boost classification accuracy comparing to single classifiers and are a commonly used method in the field of machine learning. However in some cases ensemble construction algorithms do not improve the classification accuracy. Mostly ensembles are constructed using specific machine learning method or a combination of methods, the drawback being that the combination of methods or selection of the appropriate method for a specific problem must be made by the user. To overcome this ...
Research Interests:
Abstract Early and accurate diagnosing of various diseases has proved to be of vital importance in many health care processes. In recent years intelligent systems have been often used for decision support and classification in many... more
Abstract Early and accurate diagnosing of various diseases has proved to be of vital importance in many health care processes. In recent years intelligent systems have been often used for decision support and classification in many scientific and engineering disciplines including health care. However, in many cases the proposed treatment, prediction or diagnose can differ from one intelligent system to another, similar to the real world where different medical specialists may have different opinions. Indeed, in real ...
Research Interests:
Abstract As thethird age&amp;amp;amp;#x27;of human life becomes noticeably longer, the opportunity for elderly to obtain new skills reduces the tendency to consider this period of life as being disadvantaged. Hence, the fundamental... more
Abstract As thethird age&amp;amp;amp;#x27;of human life becomes noticeably longer, the opportunity for elderly to obtain new skills reduces the tendency to consider this period of life as being disadvantaged. Hence, the fundamental aim of the project PRIMER-ICT was to educate older people in four participating countries (Slovenia, Ireland, UK and Austria) in Information and Communication Technologies (ICT) skills/practice by using an inter-generational and multisectoral approach empowering elderly to use ICT on everyday basis to improve their ...
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Abstract The problem of feature selection is an important part of data processing prior to applying a learning algorithm. By adding only the most relevant features to the subset, we can improve the results of the learning algorithm. There... more
Abstract The problem of feature selection is an important part of data processing prior to applying a learning algorithm. By adding only the most relevant features to the subset, we can improve the results of the learning algorithm. There are two common approaches: a wrapper uses learning algorithm to evaluate the relevance of selected features, while a filter evaluates features according to heuristics based on characteristics of the data. In our proposed method we search for the most relevant features from the subset of features that ...
Research Interests:
Abstract This paper presents an overview of past papers published at the CBMS symposiums from a content analysis point of view. A simple, yet effective word counting using Harvard Psycho-Social dictionary was used to estimate different... more
Abstract This paper presents an overview of past papers published at the CBMS symposiums from a content analysis point of view. A simple, yet effective word counting using Harvard Psycho-Social dictionary was used to estimate different aspects of sentiment that can be present even in scientific papers. Using simple statistics we uncover some of the very interesting trends in the last five CBMS symposiums. Additional to pure statistics we used some of the most advanced classification techniques to see if there are any ...
Abstract A lot of research has been done in the field of assembling classifiers in ensembles and on the other hand selecting the most appropriate single classifiers for a given problem which was solved by meta-learning techniques. This... more
Abstract A lot of research has been done in the field of assembling classifiers in ensembles and on the other hand selecting the most appropriate single classifiers for a given problem which was solved by meta-learning techniques. This paper presents application of recently proposed ensemble of classifiers called Rotation Forest to Grading meta-learning scheme, where it is used as one of the base classifiers and meta-level classifier at the same time. Our proposed Grading variation is compared to four widely used classifiers on 14 datasets ...
Research Interests:
Research Interests:
Research Interests:
Abstract Current software quality estimation models often involve the use of supervised learning methods for building a software fault prediction models. In such models, dependent variable usually represents a software quality measurement... more
Abstract Current software quality estimation models often involve the use of supervised learning methods for building a software fault prediction models. In such models, dependent variable usually represents a software quality measurement indicating the quality of a module by risk-basked class membership, or the number of faults. Independent variables include various software metrics as McCabe, Error Count, Halstead, Line of Code, etc... In this paper we present the use of advanced tool for data mining called Multimethod on the ...
Research Interests:
Random Forests, Support Vector Machines and k-Nearest Neighbors are successful and proven classification techniques that are widely used for different kinds of classification problems. One of them is classification of genomic and... more
Random Forests, Support Vector Machines and k-Nearest Neighbors are successful and proven classification techniques that are widely used for different kinds of classification problems. One of them is classification of genomic and proteomic data that is known as a problem with extremely high dimensionality and therefore demands suited classification techniques. In this domain they are usually combined with gene selection techniques to provide optimal classification accuracy rates. Another reason for reducing the ...