Mishkovski, Igor

Preferred name

Mishkovski, Igor

Official Name

Mishkovski, Igor

Main Affiliation

igor.mishkovski@finki.ukim.mk

Now showing 1 - 10 of 63

Some of the metrics are blocked by your
consent settings
Performance Evaluation of Word and Sentence Embeddings for Finance Headlines Sentiment Analysis
(Springer International Publishing, 2019)
;
Gjorgjevikj, Ana
;
;
;
Vodenska, Irena
Some of the metrics are blocked by your
consent settings
Classification of Companies using Graph Neural Networks
(IEEE, 2024-05-20)
Manchev, Jovan
;
;
Classification of companies into GICS categories can be addressed using Graph Neural Networks (GNN), by utilizing the different types of relationship between companies such as customer, supplier, partner, competitor, and investor. We use the Relato business graph data and compare the performances of several GNNs and a large language model like BERT that is trained only on the descriptions of the companies. Our goal is company classification into its corresponding category within the four tiers of the GICS hierarchy. Several architectures of GNNs are explored such as GCN, GraphSAGE and GAT, but also RGCN and RGAT that consider the edge type, or relationship between the companies. The main purpose is to reveal what kind of relationship between the companies is most valuable when determining the category of the company. The findings indicate that Graph Neural Networks (GNNs) enhance both classification performance and the understanding of collaboration patterns among companies, providing valuable insights for determining the industry in which these companies operate. This contrasts with the classification based solely on company descriptions using BERT.
Some of the metrics are blocked by your
consent settings
Comparative Study of Random Walks with One-Step Memory on Complex Networks
(Springer, Cham, 2023-03-30)
;
;
We investigate searching efficiency of different kinds of random walk on complex networks which rely on local information and one-step memory. For the studied navigation strategies we obtained theoretical and numerical values for the graph mean first passage times as an indicator for the searching efficiency. The experiments with generated and real networks show that biasing based on inverse degree, persistence and local two-hop paths can lead to smaller searching times. Moreover, these biasing approaches can be combined to achieve a more robust random search strategy. Our findings can be applied in the modeling and solution of various real-world problems.
Some of the metrics are blocked by your
consent settings
Comparative analysis of NLP-based models for company classification
(MDPI, 2024-01-31)
Rizinski, Maryan
;
Jankov, Andrej
;
Sankaradas, Vignesh
;
Pinsky, Eugene
;
The task of company classification is traditionally performed using established standards, such as the Global Industry Classification Standard (GICS). However, these approaches heavily rely on laborious manual efforts by domain experts, resulting in slow, costly, and vendor-specific assignments. Therefore, we investigate recent natural language processing (NLP) advancements to automate the company classification process. In particular, we employ and evaluate various NLP-based models, including zero-shot learning, One-vs-Rest classification, multi-class classifiers, and ChatGPT-aided classification. We conduct a comprehensive comparison among these models to assess their effectiveness in the company classification task. The evaluation uses the Wharton Research Data Services (WRDS) dataset, consisting of textual descriptions of publicly traded companies. Our findings reveal that the RoBERTa and One-vs-Rest classifiers surpass the other methods, achieving F1 scores of 0.81 and 0.80 on the WRDS dataset, respectively. These results demonstrate that deep learning algorithms offer the potential to automate, standardize, and continuously update classification systems in an efficient and cost-effective way. In addition, we introduce several improvements to the multi-class classification techniques: (1) in the zero-shot methodology, we use TF-IDF to enhance sector representation, yielding improved accuracy in comparison to standard zero-shot classifiers; (2) next, we use ChatGPT for dataset generation, revealing potential in scenarios where datasets of company descriptions are lacking; and (3) we also employ K-Fold to reduce noise in the WRDS dataset, followed by conducting experiments to assess the impact of noise reduction on the company classification results.
Some of the metrics are blocked by your
consent settings
Company classification using zero-shot learning
(Ss Cyril and Methodius University in Skopje, Faculty of Computer Science and Engineering, Republic of North Macedonia, 2023-07)
Rizinski, Maryan
;
Jankov, Andrej
;
Sankaradas, Vignesh
;
Pinsky, Eugene
;
In recent years, natural language processing (NLP) has become increasingly important in a variety of business applications, including sentiment analysis, text classification, and named entity recognition. In this paper, we propose an approach for company classification using NLP and zero-shot learning. Our method utilizes pre-trained transformer models to extract features from company descriptions, and then applies zero-shot learning to classify companies into relevant categories without the need for specific training data for each category. We evaluate our approach on publicly available datasets of textual descriptions of companies, and demonstrate that it can streamline the process of company classification, thereby reducing the time and resources required in traditional approaches such as the Global Industry Classification Standard (GICS). The results show that this method has potential for automation of company classification, making it a promising avenue for future research in this area.
Some of the metrics are blocked by your
consent settings
MEMRISTIVE NETWORKS OF CHUA’S CIRCUITS
(2014)
;
;
Kocarev, Ljupco
Although envisioned in 1971 by L. Chua, memristors have attracted the attention of the research community recently by the promotion of their feasibility and a vast number of possible applications in non-volatile computer memory, pattern recognition and modelling neural networks. Synchronization is widely studied as a phenomenon in neural networks. This work provides synchronization analyses of two kinds of memristive networks of oscillators. First, we numerically examine networks of Chua’s circuits coupled by memristors that adapt according to the local state disagreements. As second, we employ the Master stability function (MSF) approach to study synchronization in networks of memristive Chua’s circuits coupled through simple resistors.
Some of the metrics are blocked by your
consent settings
Organizations analysis with complex network theory
(Springer, Berlin, Heidelberg, 2010-09-12)
Banova, Todorka
;
;
;
Kocarev, Ljupcho
In this paper, we propose network measures and analytical procedures for modeling the structure and the behavior of the basic types of organizations, such as: line, functional, line-and-staff, project and matrix organization. In order to obtain some tangible information about the connectivity between employees and structural properties of organizations, we develop network generators for all five types of organizations. We review various roles and groups of employees within the organizational network, and we assess social position and impact of a particular employee. Except, assessed locations of actors within an organizational network, we analyze the structure of network to find specific employees who have similar roles in the organization and have a tendency to be equivalent in terms of their potential to act in the organization. We estimate what is the confidentiality of the organizational network depending on the removal of a certain communication between employees and what is the percentage of communications that disconnect the organization in unconnected parts.
Some of the metrics are blocked by your
consent settings
Comparison of DDOS detection methods in real world scenario
(Ss Cyril and Methodius University in Skopje, Faculty of Computer Science and Engineering, Republic of North Macedonia, 2023-07)
Bidikov, Vladislav
;
In this paper we will see the current and new developed methods for distributed denial of service attack (DDOS) detection. We will also see some of the possibility for mitigation of attacks in scenarios where they are detected sooner. We will use data from the DDOS attack in June/August/September 2022 on the Faculty and learn valuable lessons from there.
Some of the metrics are blocked by your
consent settings
Last Mile Delivery with Autonomous Vehicles: Fiction or Reality?
(Ss. Cyril and Methodius University in Skopje, Faculty of Computer Science and Engineering, Republic of North Macedonia, 2019-05)
;
;
;
Autonomous vehicles (AVs) are a disruptive technology of the 21-st century that are beginning the next revolution of the transportation of people and goods. Their presence has a particular impact on the future directions of development of E-commerce. The number of online orders is in a steep incline, and so is the necessity to deliver goods to the customer in an efficient and environmental friendly way. Using autonomous drones, pods and vans for delivery of goods has already become reality. But, what is the state of the art of the companies offering these services and how do people feel about it? The aim of this paper is to make an overview of the business models of the companies developing AVs for Last Mile Delivery (LMD) of goods and to find out what is the attitudes of the online customers towards using AVs for delivery of their goods.
Some of the metrics are blocked by your
consent settings
Demographic analysis of music preferences in streaming service networks
(Springer, Cham, 2020-02-22)
Jovanovska, Lidija
;
Evkoski, Bojan
;
;
As Daniel J. Levitin noted, music is a cross-cultural phenomenon, a ubiquitous activity found in every known human culture. It is indeed, a living matter that flows through cultures, which makes it a complex system potentially holding valuable information. Therefore, we model country-to-country interactions to reveal macro-level music trends. The purpose of this paper is twofold. Firstly, we explore the way specific demographic characteristics, such as language and geographic location affect the global community structure in streaming service networks. Secondly, we examine whether a clear flow of musical trends exists in the world by identifying countries who are prominent leaders on the music streaming charts. The community analysis shows that there is strong support for the first claim. Next, we find that the flow of musical trends is not strongly directional globally, although we were still able to identify prominent leaders and followers within the communities. The obtained results can further lead to the development of more sophisticated music recommendation systems, kindle new cultural studies and bring discoveries in the field of musicology.

Mishkovski, Igor

Filters

Settings

Sort By

Results per page