Kulakov, Andrea

Preferred name

Kulakov, Andrea

Official Name

Kulakov, Andrea

Main Affiliation

andrea.kulakov@finki.ukim.mk

Now showing 1 - 10 of 72

Some of the metrics are blocked by your
consent settings
Classifying Protein Structures by Using Protein Ray Based Descriptor, KNN and FuzzyKNN Classification Methods
(IEEE, 2020-09-28)
;
;
Some of the metrics are blocked by your
consent settings
Automating Feature Extraction from Entity-Relation Models: Experimental Evaluation of Machine Learning Methods for Relational Learning
(MDPI AG, 2024-04-01)
Stanoev, Boris
;
Mitrov, Goran
;
;
;
<jats:p>With the exponential growth of data, extracting actionable insights becomes resource-intensive. In many organizations, normalized relational databases store a significant portion of this data, where tables are interconnected through some relations. This paper explores relational learning, which involves joining and merging database tables, often normalized in the third normal form. The subsequent processing includes extracting features and utilizing them in machine learning (ML) models. In this paper, we experiment with the propositionalization algorithm (i.e., Wordification) for feature engineering. Next, we compare the algorithms PropDRM and PropStar, which are designed explicitly for multi-relational data mining, to traditional machine learning algorithms. Based on the performed experiments, we concluded that Gradient Boost, compared to PropDRM, achieves similar performance (F1 score, accuracy, and AUC) on multiple datasets. PropStar consistently underperformed on some datasets while being comparable to the other algorithms on others. In summary, the propositionalization algorithm for feature extraction makes it feasible to apply traditional ML algorithms for relational learning directly. In contrast, approaches tailored specifically for relational learning still face challenges in scalability, interpretability, and efficiency. These findings have a practical impact that can help speed up the adoption of machine learning in business contexts where data is stored in relational format without requiring domain-specific feature extraction.</jats:p>
Some of the metrics are blocked by your
consent settings
Classification of Protein Structures by Making Fuzzy-Rough Feature Selection
(IEEE, 2020-10-22)
;
;
Some of the metrics are blocked by your
consent settings
Automating feature extraction from entity-relation models: Experimental evaluation of machine learning methods for relational learning
(MDPI, 2024-04-01)
Stanoev, Boris
;
Mitrov, Goran
;
;
Mirceva, Georgina
;
With the exponential growth of data, extracting actionable insights becomes resource-intensive. In many organizations, normalized relational databases store a significant portion of this data, where tables are interconnected through some relations. This paper explores relational learning, which involves joining and merging database tables, often normalized in the third normal form. The subsequent processing includes extracting features and utilizing them in machine learning (ML) models. In this paper, we experiment with the propositionalization algorithm (i.e., Wordification) for feature engineering. Next, we compare the algorithms PropDRM and PropStar, which are designed explicitly for multi-relational data mining, to traditional machine learning algorithms. Based on the performed experiments, we concluded that Gradient Boost, compared to PropDRM, achieves similar performance (F1 score, accuracy, and AUC) on multiple datasets. PropStar consistently underperformed on some datasets while being comparable to the other algorithms on others. In summary, the propositionalization algorithm for feature extraction makes it feasible to apply traditional ML algorithms for relational learning directly. In contrast, approaches tailored specifically for relational learning still face challenges in scalability, interpretability, and efficiency. These findings have a practical impact that can help speed up the adoption of machine learning in business contexts where data is stored in relational format without requiring domain-specific feature extraction.
Some of the metrics are blocked by your
consent settings
Classification of Protein Structures Using Deep Learning Models
(IEEE, 2022-05-23)
Mirceva, Georgina
;
;
Some of the metrics are blocked by your
consent settings
Protein classification by using four approaches for extraction of the protein ray-based descriptor
(Ss. Cyril and Methodius University in Skopje, Faculty of Computer Science and Engineering, Republic of North Macedonia, 2020-05-08)
Mirceva, Georgina
;
The knowledge about the protein molecules, and how they influence the processes in the humans is very worth, because it is really needed in order to develop new drugs for diseases. In proteomics, one of the most important tasks is solving the problem of classification of protein molecules. The literature provides plethora of methods that could be used for this task. However, it is still an open issue where still there is a need for fast computational methods that would provide accurate classification of proteins. In this paper, we focus on solving this task. For that purpose, first, we extract feature vectors that hold information about the main features of the proteins. The feature vectors that are used in this study are obtained by following the procedure for extraction of our protein ray-based descriptor that we have introduced in our former studies. For that purpose, the skeleton of the protein is interpolated with predefined number of interpolation points, and then the elements of the feature vector are extracted as Euclidean distances between the interpolation points and center of mass. Besides this approach, in this study we also use three additional approaches for extraction of the feature vectors, where we focus on the change of the Euclidean distance to the center of mass between two consecutive interpolation points. After extracting feature vectors, next we apply several wellknown classification methods in order to generate classification model. We present the results obtained with these four approaches used for extraction of the feature vectors.
Some of the metrics are blocked by your
consent settings
Classification of Protein Structures Using Deep Learning Models
(IEEE, 2022-05-23)
Mirceva, Georgina
;
;
Some of the metrics are blocked by your
consent settings
Machine learning approach for emotion recognition in speech
(2014)
Gjoreski, Martin
;
Gjoreski, Hristijan
;
This paper presents a machine learning approach to automatic recognition of human emotions from speech. The approach consists of three steps. First, numerical features are extracted from the sound database by using audio feature extractor. Then, feature selection method is used to select the most relevant features. Finally, a machine learning model is trained to recognize seven universal emotions: anger, fear, sadness, happiness, boredom, disgust and neutral. A thorough ML experimental analysis is performed for each step. The results showed that 300 (out of 1582) features, as ranked by the gain ratio, are sufficient for achieving 86% accuracy when evaluated with 10 fold cross-validation. SVM achieved the highest accuracy when compared to KNN and Naive Bayes. We additionally compared the accuracy of the standard SVM (with default parameters) and the one enhanced by Auto-WEKA (optimized algorithm parameters) using the leave-one-speaker-out technique. The results showed that the SVM enhanced with Auto-WEKA achieved significantly better accuracy than the standard SVM, i.e., 73% and 77% respectively. Finally, the results achieved with the 10 fold cross-validation are comparable and similar to the ones achieved by a human, i.e., 86% accuracy in both cases. Even more, low energy emotions (boredom, sadness and disgust) are better recognized by our machine learning approach compared to the human.
Some of the metrics are blocked by your
consent settings
Architecture for wireless sensor and actor networks control and data acquisition
(IEEE, 2011-06-27)
;
;
;
Wireless Sensor and Actor Networks (WSANs) have received increased attention from the research community. This is mainly because as an extension to Wireless Sensor Networks(WSN), they have the ability to actively participate in the environment trough the actors. This however introduces new challenges as to how to transfer commands between nodes, actors and central station who may be from different manufacturers and use different communication protocols. Another important aspect is the ability of the WSAN to present the data to the interested party or to receive the command from the operator, and do this with in the simplest and most user friendly way as possible. In this paper we propose architecture for interconnection between different layers of WSANs and the central stations that would allow building a simple interface that would ease the operation with WSANs in view of Control and Data Acquisition.
Some of the metrics are blocked by your
consent settings
Interactivism in artificial intelligence (AI) and intelligent robotics
(Pergamon, 2006-08-01)
Stojanov, Georgi
;
Trajkovski, Goran
;
This paper overviews the interactivist model of representation and its applications in artificial intelligence (AI) and intelligent robotics. Selected examples from approaches in AI and robotics are contrasted with the generic interactivist architecture in order to illustrate specific features of it. Petitage´, an artificial agent that instantiates our interactivist-expectative theory of agency and learning (IETAL), is discussed in detail from the interactivist perspective.

Kulakov, Andrea

Filters

Settings

Sort By

Results per page