Vili Podgorelec

University of Maribor | UM · Institute of Informatics

This systematic literature review employs the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology to investigate recent applications of explainable AI (XAI) over the past three years. From an initial pool of 664 articles identified through the Web of Science database, 512 peer-reviewed journal articles met the in...

V prostrani, neukročeni divjini umetne inteligence se je pojavila in postavila v ospredje nova generativna vrsta: veliki jezikovni modeli. Ti orjaki s svojimi milijardami parametrov tavajo po podatkovni pokrajini, lačni vzorcev in vpogledov v obilje besedil. Toda izkoriščanje njihove moči ni enostaven podvig. Lahko so nepredvidljivi, nagnjeni k hal...

Microbiota analysis can provide valuable insights in various fields, including diet and nutrition, understanding health and disease, and in environmental contexts, such as understanding the role of microorganisms in different ecosystems. Based on the results, we can provide targeted therapies, personalized medicine, or detect environmental contamin...

Explainable artificial intelligence (XAI) refers to machine learning techniques, or general methods in artificial intelligence, for which the underlying decision logic and outcomes can be explained. It addresses the tradeoff between powerful but opaque machine learning models by shedding light into the black boxes. Thus, XAI are applicable only for...

Numerical association rule mining (NARM) is a popular method under the umbrella of data mining, focused on finding relationships between attributes in transaction databases. Numerical association rules for time series are a new paradigm that extends the applicability of NARM to the domain of time series. Association rule mining algorithms result in...

Numerical association rule mining offers a very efficient way of mining association rules, where algorithms can operate directly with categorical and numerical attributes. These methods are suitable for mining different transaction databases, where data are entered sequentially. However, little attention has been paid to the time series numerical a...

Uporabnik: Pripravljam prispevek za konferenco OTS 2023 na temo pomočnikov umetne inteligence (natančneje, velikih jezikovnih modelov), ter priložnosti in pasti, ki jih taka tehnologija prinaša v svet razvoja programske opreme. Prispevek je namenjen izkušenim inženirjem, ki so najverjetneje že dodobra preizkusili pomočnike umetne inteligence pri sv...

Programski jezik Julia je v zadnjih letih pridobil veliko pozornosti zaradi svoje edinstvene kombinacije zmožnosti programiranja na visoki ravni, zmogljivosti in enostavne uporabe. Dandanes programski jezik spada med tri najpomembnejše jezike za namene podatkovne znanosti poleg jezika Python in R. Namen tega članka je zagotoviti pregled ključnih la...

This study aimed to observe the impact of eight explainable AI (XAI) explanation techniques on user trust and satisfaction in the context of XAI-enhanced learning analytics while comparing two groups of STEM college students based on their Bologna study level, using various established feature relevance techniques, certainty, and comparison explana...

Nowadays, the issue of student drop-out is addressed not only through the prism of pedagogy, but also by technological practices. In this paper, we demonstrate how a student drop-out could be predicted through a student’s performance using different Machine Learning techniques, i.e., supervised learning and unsupervised learning. The results show t...

Association rule mining is intended for searching for the relationships between attributes in transaction databases. The whole process of rule discovery is very complex, and involves pre-processing techniques, a rule mining step, and post-processing, in which visualization is carried out. Visualization of discovered association rules is an essentia...

Numerical association rule mining offers a very efficient way of mining association rules, where algorithms can operate directly with categorical and numerical attributes. These methods are suitable for mining different transaction databases, where data are entered sequentially. However, little attention has been paid to the time series numerical a...

Learning analytics (LA) is data collection, analysis, and representation of data about learners in order to improve their learning and performance. Furthermore, LA opens the door to opportunities for self-regulated learning in higher education, a circular process in which learners activate and sustain behaviours that are oriented toward their perso...

A plethora of data on students’ activities and interactions is acquired through the use of virtual learning environments. The analysis of gathered data and the prediction of withdrawal and academic success has been an ongoing point of discussion in the field of Learning Analytics. The main aim of this paper is an early prediction of students‘ acade...

Pitna voda je redek in dragocen vir, zato je ustrezno upravljanje tega vira bistveno za družbeni in gospodarski razvoj vsake države ali regije, saj je čista pitna voda uporabljena v vseh ključnih sektorjih gospodarstva, kot so kmetijstvo, industrija, energetika in promet. Pametne naprave so spremenile skoraj vse vidike našega življenja – in tudi pr...

Podjetja dandanes vneto tekmujejo v zagotavljanju najboljših možnih storitev svojim strankam, pri čemer podjetja na trgu električne energije niso izjema. Glede na negotove razmere na področju zagotavljanja energetskih virov, v katerih se je znašel svet, vse večje potrebe po električni energiji in trend strme rasti cen energije je postala optimizaci...

Dostopnost velikih količin podatkov in relativno poceni in dostopne računske moči je v zadnjih nekaj letih pripomogla k enormnemu vzponu računske inteligence. Čeprav se mnoga podjetja in organizacije že dolga leta poslužujejo uporabe različnih tehnik matematične optimizacije, ki se najpogosteje uporabljajo za namen optimizacije poslovnih procesov,...

The advances of information technology have brought us to a period of increasingly rapid creation, sharing and exchange of information. In general, we do not perceive just how much information we are creating by using modern information technology devices. Still, the reality here is that we are creating data on a business and personal level while i...

Planning sport sessions automatically is becoming a very important aspect of improving an athlete’s fitness. So far, many Artificial Intelligence methods have been proposed for planning sport training sessions. These methods depend largely on test data, where Machine Learning models are built, yet evaluated later. However, one of the biggest concer...

Companies nowadays eagerly compete in providing their customers with the best possible services, where the companies in the electrical energy market are no exception. As artificial intelligence and machine learning are considered the fundamental multi-purpose technologies and the innovation entity with the most significant potential for disruption,...

The Covid-19 pandemic has shaken the world and brought us new challenges such as spreading information. In today’s social networks-dominated world, the public is commonly informed via Instagram, one of the fastestgrowing networks. In our research, we looked at the said network for the purpose of spreading information to the public during the pandem...

The growth of social media and its interactivity and different communication functions represents a big opportunity for companies to conduct successful business. The communication of B2B companies on social media offers a range of different ways of connecting companies with each other and their customers successfully. Companies need to be careful w...

Childhood pneumonia, the leading cause of children mortality globally, is most commonly diagnosed based on the radiographic data, which requires radiologic interpretation of X-ray images. With recent advancements in the field of deep learning, the convolutional neural networks (CNN) have proven to be able to achieve great performance in medical ima...

Nowadays, only a few papers exist dealing with Association Rule Mining with numerical attributes. When we are confronted with solving this problem using nature-inspired algorithms, two issues emerge: How to shrink the values of the upper and lower bounds of attributes properly, and How to define the evaluation function properly? This paper proposes...

For more than a year the COVID-19 epidemic is threatening people all over the world. Numerous researchers are looking for all possible insights into the new corona virus SARS-CoV-2. One of the possibilities is an in-depth analysis of X-ray images from COVID-19 patients, commonly conducted by a radiologist, which are due to high demand facing with o...

Background Considerable percentage of patients tested with a standard patch test series respond with a positive reaction to more than one allergen, and some associations between synchronous positive reactions to distinct standard patch allergens have been described in the literature. Objectives To evaluate the prevalence of sensibilization to hapt...

With the exponential growth of the presence of sport in the media, the need for effective classification of sports images has become crucial. The traditional approaches require carefully hand-crafted features, which make them impractical for massive-scale data and less accurate in distinguishing images that are very similar in appearance. As the de...

With the utilization of deep learning approaches, the key factors for a successful application are sufficient datasets with reliable ground truth, which are generally not easy to obtain, especially in the field of medicine. In recent years, this issue has been commonly addressed with the exploitation of transfer learning via fine-tuning, which enab...

Phishing stands for a fraudulent process, where an attacker tries to obtain sensitive information from the victim. Usually, these kinds of attacks are done via emails, text messages, or websites. Phishing websites, which are nowadays in a considerable rise, have the same look as legitimate sites. However, their backend is designed to collect sensit...

At this early stage in the COVID-19 epidemic, researchers are looking for all possible insights into the new corona virus SARS-CoV-2. One of the possibilities is an in-depth analysis of X-ray images from COVID-19 patients. We first developed a new adapted classification method that is able to identify COVID-19 patients based on a chest X-ray, and t...

With the advent of big data, interest for new data mining methods has increased dramatically. The main drawback of traditional data mining methods is the lack of comprehensibility. In this paper, the firefly algorithm was employed for standalone binary classification, where each solution is represented by two classification rules that are easy unde...

The protection of sensitive data against unauthorized access remains a primary concern of modern life. Over time, many different approaches have been introduced to tackle this problem, from substitution ciphers in classic cryptography to post-quantum cryptography as a representative of modern cryptography. In this paper, we focus on a polyalphabeti...

This paper outlines a short overview of swarm intelligence algorithms that are used within the software engineering area. Swarm intelligence algorithms have been used in many software engineering tasks, e.g., grammatical inference or mutation testing. However, their presence in the agile software development field is still awakening. As there are s...

Over the past years, the application of deep neural networks in a wide range of areas is noticeably increasing. While many state-of-the-art deep neural networks are providing the performance comparable or in some cases even superior to humans, major challenges such as parameter settings for learning deep neural networks and construction of deep lea...

Self-admitted technical debt (SATD) is annotated in source code comments by developers and has been recognized as a great source of discovering flawed software. To reduce manual effort, some recent studies have focused on automated detection of SATD using text classification methods. To train their classifier, these methods need labeled samples, wh...

The analysis of non-stationary signals commonly includes the signal segmentation process, dividing such signals into smaller time series, which are considered stationary and thus easier to process. Most commonly, the methods for signal segmentation utilize complex filtering, transformation and feature extraction techniques together with various kin...

Z razcvetom strojnega učenja in prodorom v vse veje gospodarstva, se nenehno povečuje tudi potreba po ustaljenih razvojnih procesih pri razvoju napovednih modelov strojnega učenja, ki bi olajšala njihovo vpeljavo in integracijo v obstoječe informacijske sisteme. Podobno kot je to že ustaljena praksa pri programskem inženirstvu, je potreba in želja...

Bat algorithm belongs to a class of swarm intelligence algorithms. Comparing to the other stochastic nature-inspired population-based algorithms, this has always been considered as computationally inexpensive. Due to its simplicity and effectiveness, it is very popular in scientific community by solving various optimization problems. However, not e...

In the standard process of creating classification decision trees with genetic programming, the evaluation process it the most time-consuming part of the whole evolution loop. Here we introduce a lazy evaluation approach of classification decision trees in the evolution process, that does not evaluate the whole population but evaluates only the ind...

Imbalanced data typically refers to class distribution skews and underrepresented data, which affect the performance of learning algorithms. Such data are well-known in real-life situations, such as behavior analysis, cancer malignancy grading, industrial systems’ monitoring and software defect prediction. In this paper, we present a W-PSO method,...

This book constitutes the proceedings of the 23rd European Conference on Advances in Databases and Information Systems, ADBIS 2019, held in Bled, Slovenia, in September 2019. The 27 full papers presented were carefully reviewed and selected from 103 submissions. The papers cover a wide range of topics from different areas of research in database an...

This book constitutes the thoroughly refereed short papers, workshops and doctoral consortium papers of the 23rd European Conference on Advances in Databases and Information Systems, ADBIS 2019, held in Bled, Slovenia, in September 2019. The 19 short research papers and the 5 doctoral consortium papers were carefully reviewed and selected from 103...

Different challenges arise while detecting deficient software source code. Usually a large number of potentially problematic entities are identified when an individual software metric or individual quality aspect is used for the identification of deficient program entities. Additionally, a lot of these entities quite often turn out to be false posi...

In this paper, we explore the correlation between cuisine and text-based information in recipes. The experiments are conducted on a real dataset consisting of 9,080 recipes with data science approaches focusing on enhancing cuisine prediction and providing a detailed insight on the characterization of food cultures. The findings suggest that inform...

Over last years the deep neural network has become one of the most popular classification methods with performance comparable and in some cases even superior to humans in the wide range of applications. However, there are still some major challenges regarding the deep neural networks. One of the biggest, with the huge impact on the classification p...

Analyzing sport data becomes, every year, more interesting for a wide spectrum of researchers in the sports domain. Recently, more and more data relating to sports have become available to researchers due to the huge progress of information technologies. New wearable devices enable athletes to track performance data that are saved into sport activi...

The increasingly rapid creation, sharing and exchange of information nowadays put researchers and data scientists ahead of a challenging task of data analysis and extracting relevant information out of data. To be able to learn from data, the dimensionality of the data should be reduced first. Feature selection (FS) can help to reduce the amount of...

Technical debt (TD) is a term used to describe a trade off between code quality and timely software release. Since technical debt has negative impact on software development, identification of such debt is an important task in the software engineering domain. Sometimes, technical debt is annotated in source code comments. This kind of debt is refer...

The analysis of biomedical signals, such as the EEGs for measuring brain activity, provides means for the diagnosis of various cognitive tasks and neural disorders. These signals are frequently transformed into visual representations such as spectrograms, which can reveal characteristic patterns and serve as a basis for classification, when extract...

Adaptive boosting (AdaBoost) is a method for building classification ensemble, which combines multiple classifiers built in an iterative process of reweighting instances. This method proves to be a very effective classification method, therefore it was the major part of our evolutionary inspired classification algorithm. In this paper, we introduce...

Every day, millions of short-texts are generated for which effective tools for organization and retrieval are required. Because of the short length of these documents and of their extremely sparse representations, the traditional text classification methods are not effective. We propose a new approach that uses DBpedia Spotlight annotation tools, t...

In last decades, the web and online services have revolutionized the modern world. However, by increasing our dependence on online services, as a result, online security threats are also increasing rapidly. One of the most common online security threats is a so-called Phishing attack, the purpose of which is to mimic a legitimate website such as on...

In this paper, we present a novel algorithm called STAPSO, which comprises Scrum task allocation and the Particle Swarm Optimization algorithm. The proposed algorithm aims to address one of the most significant problems in the agile software development, i.e., iteration planning. The actuality of the topic is not questionable, since nowadays, agile...

Analiza biomedicinskih signalov kot je EEG, za merjenje možganskih aktivnosti omogoča diagnosticiranje različnih kognitivnih nalog in nevroloških motenj. Pogosto so takšni signali pretvorjeni v vizualne predstavitve kot so spektrogrami, ki lahko razkrijejo karakteristične vzorce in služijo kot osnova za klasifikacijo. Za namen klasifikacije EEG sig...

Imbalanced data is a relatively new branch of machine learning that gained popularity in the past years. It typically refers to a class distribution skews and underrepresented data, which affect the performance of learning algorithms. Such problem is also known as an imbalanced learning problem. In this paper, we present a W-PSO method, which compr...

Supervised text classification methods are efficient when they can learn with reasonably sized labeled sets. On the other hand, when only a small set of labeled documents is available, semi-supervised methods become more appropriate. These methods are based on comparing distributions between labeled and unlabeled instances, therefore it is importan...

Knowledge management in organisation (KMO) became an important part of our economy far before the actual term 'knowledge management' emerged. Many studies have been written about different aspects and viewpoints of knowledge management, however in this study we used a bibliometric mapping approach to determine the main research topics and the conte...

This paper deals with the problem of selecting a suitable design pattern when necessary. The number of design patterns has been rapidly rising, but management and searching facilities appear to be lagging behind. In this paper we will present a platform, which is used to search for suitable design patterns and for design patterns knowledge exchange...

Harmonising the metadata format alone does not solve the issue of efficient access to relevant information in heterogeneous environments, when different systems use different content, contextual and semantic concepts for certain entities. One such type of heterogeneous systems are also Current Research Information Systems (CRIS), which store their...

In this paper, we present an alternative approach to software vulnerability prediction with modern machine learning methods-with deep learning methods. Deep learning methods are techniques where features in our case software metrics) are processed and sent through multiple layers where transformations and computations are done in sequence to form a...

In this chapter, a Semantic Web services-based knowledge management framework that enables holistic knowledge management in organizations is presented. As the economy is becoming one single global marketplace, where the best offer wins, organizations have to search for competitive advantage within themselves. With the growing awareness that key pot...

In this chapter, a Semantic Web services-based knowledge management framework that enables holistic knowledge management in organizations is presented. As the economy is becoming one single global marketplace, where the best offer wins, organizations have to search for competitive advantage within themselves. With the growing awareness that key pot...

Knowledge management became an important part of our economy far before than the actual term “knowledge management”. Many studies have been written about different aspects and viewpoints of knowledge management, however in this study we used a bibliometric mapping approach to determine the main research topics and the contexts in which they are emp...

Operating in a knowledge-based economy requires a wide range of additional skills and competences. An important part in the transfer of new skills and competences to the labor marker influence the design of courses offered at higher education institutions. To support students’ professional career development, the content of the courses should const...

When evaluating the process of building classification decision trees, it is necessary to assess the performance of constructed trees, as well as the speed and efficiency of the algorithm. Top-down induction algorithms are relatively simple and can quickly generate good solutions, however their deterministic nature often prevents them from finding...