About
81
Publications
43,571
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
507
Citations
Introduction
Additional affiliations
April 2018 - present
September 2011 - June 2013
July 2013 - March 2018
Education
January 2005 - September 2008
October 2000 - December 2004
October 1995 - April 2000
Publications
Publications (81)
Protein interaction networks (PINs) are argued to be the richest source of hidden knowledge of the intrinsic physical and/or functional meanings of the involved proteins. We propose a novel method for computational protein function prediction based on semantic homogeneity optimization in PIN (SHOPIN). The SHOPIN method creates graph representations...
The proposed protein function prediction methods are mostly based on sequence or structure protein similarity and do not take into account the semantic similarity extracted from protein knowledge databases such as Gene Ontology. Many studies have shown that identification of protein complexes or functional modules can be effectively done by cluster...
In recent years, use of data mining and machine learning techniques in finance for such tasks as pattern recognition, classification, and time series forecasting have dramatically increased. However, the large numbers of parameters that must be selected to develop a good forecasting model have meant that the design process still involves much trial...
De novo assembly remains an ongoing challenge for the bio infor ma tics community. Many strategies have been developed; however
there is no perfect de novo assembler. This publication evaluates four assemblers run on simulated data, comparing their outputs and testing their performance
on two bacterial strains with different GC content (guanine – c...
The protein function is tightly related to classification of proteins in hierarchical levels where proteins share same or similar functions. One of the most relevant protein classification schemes is the structural classification of proteins (SCOP). The SCOP scheme has one negative drawback; due to its manual classification methods, the dynamic of...
Social media resurgence of antisocial behavior has exerted a downward spiral on stereotypical beliefs, and hateful comments towards individuals and social groups, as well as false or distorted news. The advances in graph neural networks employed on massive quantities of graph-structured data raise high hopes for the future of mediating communicatio...
There are strong indications that structural and functional magnetic resonance imaging (MRI) may help identify biologically relevant phenotypes of neurodevelopmental disorders such as Autism spectrum disorder (ASD). Extracting patterns from MRI data is challenging due to the high dimensionality, limited cardinality and data heterogeneity. In this p...
The issue of air pollution is increasingly prominent and represents a significant environmental challenge, particularly in urban areas affected by rising migration rates. Air pollution forecasting is crucial for understanding the mechanisms underlying pollution in a specific region, but analyzing high-dimensional data with spatial and temporal depe...
Recent studies have highlighted that gut microbiota can alter colorectal cancer susceptibility and progression due to its impact on colorectal carcinogenesis. This work represents a comprehensive technical approach in modeling and interpreting the drug-resistance mechanisms from clinical data for patients diagnosed with colorectal cancer. To accomp...
Network-based representations have introduced a revolution in neuroscience, expanding the understanding of the brain from the activity of individual regions to the interactions between them. This augmented network view comes at the cost of high dimensionality, which hinders both our capacity of deciphering the main mechanisms behind pathologies, an...
Data science and machine-learning techniques help banks to optimize enterprise operations, enhance risk analyses and gain competitive advantage. There is a vast amount of research in credit risk, but to our knowledge, none of them uses credit registry as a data source to model the probability of default for individual clients. The goal of this pape...
Air pollution is becoming a rising and serious environmental problem, especially in urban areas affected by an increasing migration rate. The large availability of sensor data enables the adoption of analytical tools to provide decision support capabilities. Employing sensors facilitates air pollution monitoring, but the lack of predictive capabili...
Applied machine learning in bioinformatics is growing as computer science slowly invades all research spheres. With the arrival of modern next-generation DNA sequencing algorithms, metagenomics is becoming an increasingly interesting research field as it finds countless practical applications exploiting the vast amounts of generated data. This stud...
Cancer is one of the most widespread diseases that we come across. The complexity of this disease makes it difficult to analyze and detect biomarkers with the purpose to ease the targeted treatments. This study presents a methodology based on gene expression data that provides promising results in terms of revealing potential biomarkers associated...
Data mining together with learning analytics are emerging topics because of the huge amount of educational data coming from learning management systems. This paper presents a case study about students’ grade prediction by using data mining methods. Data obtained from Moodle log files are explored to understand the trends and effects of students’ ac...
Interpreting what a deep learning model has learned is a challenging task. In this paper, we present a deep learning architecture relying upon an attention mechanism. The main focus is put on the exploratory evaluation of attention-based deep learning models on lexicons of affective words, and examination whether the word valence is the most signif...
This paper addresses the routes planning problem in a scenario where an UAV (Unmanned Aerial Vehicle) and a land-based transportation vehicle are used to deliver parcels to customer locations. We developed and implemented a solution based on the well-known Bellman-Held-Karp dynamic programming algorithm for the Travelling Salesman Problem that find...
One of the major challenges in managing security in broadband and high-speed networks is the detection of suspicious anomalies in network traffic. In recent years a lot of effort is focused on developing automatic detection of cyber-attacks using data mining techniques on the data generated from network traffic. In this paper a methodology for auto...
Melanoma is the most dangerous form of skin cancer, and its detection at an early stage can allow timely treatment and prevention of fatal consequences. In this paper we present a case study of computer-aided diagnostics of melanoma using images of patients moles. The initial study was performed on two datasets: a benchmark dataset which is publicl...
As result of the progressive technological trend, huge amounts of da-ta from different subjects and areas are continuously generated and classified on a daily basis. In the past, the main problems were the preservation and publish-ing of data sets, but today, one of the main challenges is the presentation for better understanding of the data. The a...
This book constitutes the refereed proceedings of the 10th International ICT Innovations Conference, ICT Innovations 2018, held in Ohrid, Macedonia, in September 2018.
The 21 full papers presented were carefully reviewed and selected from 81 submissions. They cover the following topics:sensor applications and deployments, embedded and cyber-physica...
One of the essential challenges in proteomics is the computational function prediction. In Protein Interaction Networks (PINs) this problem is one of proper labeling of corresponding nodes. In this paper a novel three-step approach for supervised protein function learning in PINs is proposed. The first step derives continuous vector representation...
In this paper we create and analyze a protein-protein interaction network (PPIN) of colorectal cancer (CRC). First we identify proteins that are related to the CRC (set of seed proteins). Using this set we generate the CRC PPIN with the help of Cytoscape. We analyze this PPIN in a twofold manner. We first extract important topological features for...
Activity detection is becoming an integral part of many mobile applications. Therefore, the algorithms for this purpose should be lightweight to operate on mobile or other wearable device, but accurate at the same time. In this paper, we develop a new lightweight algorithm for activity detection based on Long Short Term Memory networks, which is ab...
The purpose of this paper is to explore the linkage between recipe’s ingredients and identification of a cuisine. This has been tackled as a problem of cuisine classification. We will examine various approaches (different machine learning algorithms) for recipes classification based on the recipe’s ingredients. The output will be the recommendation...
Recommending API related functions is a problem of finding related functions for a particular function. Two functions are related if they are used together in specific user scenarios, appear in cross-references in "see also" sections in the API documentation, are called by the same functions etc. For the purpose of this paper, two functions are con...
This paper provides a comprehensive definition of "Internet of Toys" as a new application domain of "Internet of Things" including both sociological and technological aspects. We survey available Internet of Toys architectures from the literature and propose new architecture/framework that will enable interconnection of smart toys into complex ente...
The aim of the paper is to present image retrieval for Alzheimer’s Disease (AD) based on brain atrophy pattern captured by the SPARE-AD (Spatial Pattern of Abnormality for Recognition of Early Alzheimer’s Disease) index. SPARE-AD provides individualized scores of diagnostic and predictive value found to be far beyond standard structural measures. T...
Nowadays, finding the Unmanned Aerial Vehicle (UAV) position in the absence of GPS is attractive and challenging problem in the research community. In this paper, we present a novel algorithm for mini UAV indoor localization based on distance measurements between the UAV and the existing infrastructure consisting of WiFi Access Points. Our algorith...
Customer churn is one of the main problems in the telecommunications industry. Several studies have shown that attracting new customers is much more expensive than retaining existing ones. Therefore, companies are focusing on developing accurate and reliable predictive models to identify potential customers that will churn in the near future. The a...
Culinary data that are available online can be analysed from many different aspects. In this paper we provide methods for portraying the Macedonian cuisine, as a representative of the South-European cuisine, but highly influenced from the Middle-Eastern and Eastern- European cuisine. By performing different analyses on the Macedonian recipe dataset...
One third of the world’s population suffers from some kind of neurological disorder. The development of technology allows us to analyze, model and visualize these disorders in order to help MDs in further treatments. Resting state fMRI is one of the most common ways for investigating the functional connectivity of the brain, which produces time ser...
Proteins are the main building blocks of life. They are involved in virtually all cell functions. Each protein within the living organism has a specific function, but in most of the cases the protein functions can't be known a priori. In bioinformatics, a lot of effort is made to unravel the functions of an unknown protein. Most of the computationa...
Machine learning has received increased interest by both the scientific community and the industry. Most of the machine learning algorithms rely on certain distance metrics that can only be applied to numeric data. This becomes a problem in complex datasets that contain heterogeneous data consisted of numeric and nominal (i.e. categorical) features...
The increased availability of large-scale protein-protein interaction (PPI) data has made it possible to have a network level understanding of the basic components and organization of the cell machinery. A significant number of proteins in protein interaction networks (PIN) remain uncharacterized and predicting their function remains a major challe...
Background / Purpose:
We aimed to combine structural and functional connectomes in bipolar patients on same brain parcellation; apply graph theory algorithms on connectivity matrices and discuss the results.
Main conclusion:
Clustering coefficient may be a potential biomarker for bipolar disorder.
In machine learning, the data available for analysis is becoming more complex both in terms of high-dimensionality and the way it is structured. This emphasises the need for developing machine learning algorithms that are able to tackle both the high-dimensionality and the complex structure of the data. Our work in this paper, focuses on extending...
In machine learning, the data available for analysis is becoming more complex both in terms of high-dimensionality and the way it is structured. This emphasises the need for developing machine learning algorithms that are able to tackle both the high-dimensionality and the complex structure of the data. Our work in this paper, focuses on extending...
Effective team building is an important issue of human resource management (HRM). In order to keep up with technological improvements and changes, selecting the right person for the right job position is very important. This paper describes a research and development methodology for establishing a more sophisticated approach for composing effective...
Proteins are the most important parts of all living organism. They can be found in all living systems, starting from bacteria and viruses, to the human. They are responsible for the biological processes that happen in the cell. Therefore, the knowledge on protein function is of great significance. Mainly protein function is determined in laboratory...
The recent advent of high throughput methods has generated large amounts of protein-protein interaction network (PPIN) data. When studying the workings of a biological cell, it is useful to be able to detect known and predict still undiscovered protein complexes within the cell's PPINs. Such predictions may be used as an inexpensive tool to direct...
Proteins are the main, building cell blocks, responsible for the existing cell biological processes. Therefore, precise knowledge of protein function is of great significance. There are a lot of methods which are used for protein comparison and for determining protein function. Some of them use structure alignment, others use sequence alignment, wh...
Artificial neural networks (NN) can be used to model complex relations between inputs and outputs or to find patterns in data. When dealing with time series the process of prediction with NN has to be adopted to take into account the temporal characteristics of the data. A variety of different aspects of designing NN based forecasting models were i...
This paper presents a new method for dynamic calculation of weights that can be used in the process of aggregation of classifications by weighted majority vote. The proposed method can be used for all binary classification problems for classifiers that produce probabilistic classifications. Most aggregation functions produce an output which only re...
The increasing availability of large-scale protein-protein interaction (PPI) data has made it possible to understand the basic components and organization of cell machinery from the network level. Many studies have shown that clustering protein interaction network (PIN) is an effective approach for identifying protein complexes or functional module...
The advent of high-throughput sequencing platforms brought bioinformatics to a new level. This, so called 'next-generation' sequencing technology opened the researching doors of every laboratory allowing accomplishment of previously unimaginable scale and expensive experiments. As a result, novel research areas have emerged providing huge amounts o...
Proteins are the most important cell parts, therefore, knowing their exact function is of a great significance. However, the function of large amount of proteins is still unknown. In addition, today, biologists persist on hierarchical organization the living world, and thus in protein databases also. There are many protein classification algorithms...
To understand the structure-to-function relationship, life sciences researchers and biologists need to retrieve similar structures
and classify them into the same protein fold. In this paper, we propose a 3D structure-based approach for efficient classification
of protein molecules. Classification is performed in three phases. In the first phase, w...
Gene finding is crucial in understanding the genome of a species. The long genomic sequence is not very useful, unless its biologically functional subsequences (genes) are identified. Along with the ongoing revolution in sequencing technology, the number of sequenced genomes has increased drastically. Therefore, the development of reliable automate...
The recent advent of high throughput methods has generated large amounts of protein interaction network (PIN) data. A significant number of proteins in such networks remain uncharacterized and predicting their function remains a major challenge. A number of existing techniques assume that proteins with similar functions are topologically close in t...
In this paper, comparative analysis is presented of our three 3D structure-based approaches for the efficient retrieval of protein molecules. All approaches rely on the 3D structure of the proteins. In the first approach, the Spherical Trace Transform is applied to protein 3D structures in order to produce geometry based descriptors. Additionally,...
The classification of protein structures is essential for their function determination in bioinformatics. In this work, we
use PDB files to extract descriptors based on structural characteristics of the protein, enriched with the biological features
of the primary and secondary structure elements. Then we apply C4.5 algorithm and 10-fold-crossvalid...
A mass quantity of biological data have been produced by the development of the gene sequencing projects. Till May 2008 there are around 6.000.000 identified protein sequences and around 50.000 determined protein structures. For every newely discovered protein it is very important to determine its function, which is tightly related to the protein s...
Protein secondary structure prediction remains an open and important problem in life sciences as a first step towards the crucial tertiary structure prediction. In [3], a protein secondary structure prediction algorithm called PSIPRED presents an innovative approach - feeding the neural network (NN) with a position specific scoring matrix as input...
Traditional teaching, usually based on lectures and tutorials fosters the idea of instruction-driven learning model where students are passive listeners. Besides this approach, Project Based Learning (PBL) as a different learning paradigm is standing behind constructivism learning theory, where learning from real-world situations is put on the firs...
In this paper, a 3D structure-based approach is presented for the efficient classification of protein molecules. The method relies on the geometric 3D structure of the proteins. After proper positioning of the 3D structures, the spherical trace transform is applied to them to produce geometry - based descriptors, which are completely rotation invar...
Matching 3D objects by their similarity is a fundamental problem in computer vision, multimedia databases, molecular biology, computer graphics and a variety of other fields. A challenging aspect of this problem is to find a suitable shape signature/descriptor that can be constructed and compared quickly, while still discriminating between similar...
This paper introduces our approach in building multimedia learning environment based on digital library and mobile student services. The learning material can be accessed from standard desktop PCs, but also from wireless PDA or mobile phones. The last two approaches require development of a mobile distance educational system, especially a mobile ac...
Matching 3D objects by their similarity is a fundamental problem in computer vision, multimedia databases, molecular biology, computer graphics and a variety of other fields. A challenging aspect of this problem is to find a suitable shape signature/descriptor that can be constructed and compared quickly, while still discriminating between similar...
Matching 3D objects by their similarity is a fundamental problem in computer vision, multimedia databases, molecular biology, computer graphics and a variety of other fields. A challenging aspect of this problem is to find a suitable shape signature/descriptor that can be constructed and compared quickly, while still discriminating between similar...
This paper presents a novel implementation of a 3D rendering engine able to display 3D graphics MPEG-4 objects. By using the MPEG-4 SDK (Software Developer Kit), the 3D objects are first decoded and the MPEG-4 scene graph structure is filed. We introduce a scene manager able to address in an optimized manner the rendering requirements. It is develo...
This paper presents a novel implementation of a 3D rendering engine able to display 3D graphics MPEG-4 objects. By using the MPEG-4 SDK (Software Developer Kit), the 3D objects are first decoded and the MPEG-4 scene graph structure is filed. We introduce a scene manager able to address in an optimized manner the rendering requirements. It is develo...
The main idea of this paper is to introduce the support mechanism for the augmented reality system interface for dance analysis and presentation of single dancer. In our previous work [1] we described the basic characteristics of our system for dance analysis and presentation including the high interactivity between the user and the 3D dancer. In t...
We emphasize the generation of an augmented reality environment of a single dancer based on analysis of dance annotations. We introduce a new Web3D-based interactive technique for dance animation needed for educational purposes. This approach offers new possibilities for interactive dance step observation, slow movements of fast steps, different an...
In this paper, we focus on micro level spatio-temporal characteristics of dance incorporated in Labanotation representation of dance, which will serve us as an efficient index for retrieval of video dance data. On micro level, that is essential for dance analysis and learning, body parts of the dancer take different positions in space during perfor...
In this paper, the micro-level spatio temporal characteristics of dance incorporated in Labanotation representation of the dance were used as a starting point for modeling a 3D virtual dancer. Object-Oriented modeling techniques like Unified Modeling Language (UML) were used in the 3D animation process. This paper focuses on a subsystem for 3D visu...
The comparison of protein structures is one of the most popular topics in bioinformatics community nowadays, which is used for determining the protein function. The rapid development in technology results in a huge number of data that have to be retrieved. Therefore, the necessity of robust and efficient algorithms for protein retrieval arises. In...