Milan Čabarkapa

Dr. Milan Čabarkapa

orcid

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
3 April 2025 in Scientific Reports
Verified
Journal Article
Open access
PDF file
LBLjubiša Bojić
Ljubiša Bojić
OZOlga Zagovora
Olga Zagovora
AZAsta Zelenkauskaite
Asta Zelenkauskaite
VVVuk Vuković
Vuk Vuković
Milan ČabarkapaMilan Čabarkapa
Milan Čabarkapa
AJAna Jovančević
Ana Jovančević
In the era of rapid digital communication, vast amounts of textual data are generated daily, demanding efficient methods for latent content analysis to extract meaningful insights. Large Language Models (LLMs) offer potential for automating this process, yet comprehensive assessments comparing their performance to human annotators across multiple dimensions are lacking. This study evaluates the inter-rater reliability, consistency, and quality of seven state-of-the-art LLMs. These include variants of OpenAI’s GPT-4, Gemini, Llama-3.1-70B, and Mixtral 8 × 7B. Their performance is compared to human annotators in analyzing sentiment, political leaning, emotional intensity, and sarcasm detection. The study involved 33 human annotators and eight LLM variants assessing 100 curated textual items. This resulted in 3,300 human and 19,200 LLM annotations. LLM performance was also evaluated across three-time points to measure temporal consistency. The results reveal that both humans and most LLMs exhibit high inter-rater reliability in sentiment analysis and political leaning assessments, with LLMs demonstrating higher reliability than humans. In emotional intensity, LLMs displayed higher reliability compared to humans, though humans rated emotional intensity significantly higher. Both groups struggled with sarcasm detection, evidenced by low reliability. Most LLMs showed excellent temporal consistency across all dimensions, indicating stable performance over time. This research concludes that LLMs, especially GPT-4, can effectively replicate human analysis in sentiment and political leaning, although human expertise remains essential for emotional intensity interpretation. The findings demonstrate the potential of LLMs for consistent and high-quality performance in certain areas of latent content analysis.

10 June 2025 in Humanities and Social Sciences Communications
Verified
Journal Article
Open access
PDF file
LBLjubiša Bojić
Ljubiša Bojić
PKPredrag Kovačević
Predrag Kovačević
Milan ČabarkapaMilan Čabarkapa
Milan Čabarkapa
As Large Language Models (LLMs) become increasingly integrated into everyday life as general-purpose multimodal AI systems, their capabilities to simulate human understanding are under examination. This study investigates LLMs’ ability to interpret linguistic pragmatics, which involves context and implied meanings. Using Grice’s communication principles, we evaluated both LLMs (GPT-2, GPT-3, GPT-3.5, GPT-4, and Bard) and human subjects (N = 147) on dialogue-based tasks. Human participants included 71 primarily Serbian students and 76 native English speakers from the United States. Findings revealed that LLMs, particularly GPT-4, outperformed humans. GPT-4 achieved the highest score of 4.80, surpassing the best human score of 4.55. Other LLMs performed well: GPT-3.5 scored 4.10, Bard 3.75, and GPT-3 3.25; GPT-2 had the lowest score of 1.05. The average LLM score was 3.39, exceeding the human cohorts’ averages of 2.80 (Serbian students) and 2.34 (U.S. participants). In the ranking of all 155 subjects (including LLMs and humans), GPT-4 secured the top position, while the best human ranked second. These results highlight significant progress in LLMs’ ability to simulate understanding of linguistic pragmatics. Future studies should confirm these findings with more dialogue-based tasks and diverse participants. This research has important implications for advancing general-purpose AI models in various communication-centered tasks, including potential application in humanoid robots in the future.

18 September 2021 in Electronics
Verified
Journal Article
Open access
PDF file
RMRade Golub Matić
Rade Golub Matić
MKMilos Kabiljo
Milos Kabiljo
Miodrag ZivkovicMiodrag Zivkovic
Miodrag Zivkovic
Milan ČabarkapaMilan Čabarkapa
Milan Čabarkapa
In recent years, gradual improvements in communication and connectivity technologies have enabled new technical possibilities for the adoption of chatbots across diverse sectors such as customer services, trade, and marketing. The chatbot is a platform that uses natural language processing, a subset of artificial intelligence, to find the right answer to all users’ questions and solve their problems. Advanced chatbot architecture that is extensible, scalable, and supports different services for natural language understanding (NLU) and communication channels for interactions of users has been proposed. The paper describes overall chatbot architecture and provides corresponding metamodels as well as rules for mapping between the proposed and two commonly used NLU metamodels. The proposed architecture could be easily extended with new NLU services and communication channels. Finally, two implementations of the proposed chatbot architecture are briefly demonstrated in the case study of “ADA” and “COVID-19 Info Serbia”.

27 March 2021 in Symmetry
Verified
Journal Article
Open access
PDF file
IBIvan Babić
Ivan Babić
AMAleksandar Miljkovic
Aleksandar Miljkovic
Milan ČabarkapaMilan Čabarkapa
Milan Čabarkapa
VNVojkan Nikolic
Vojkan Nikolic
AĐAleksandar Đorđević
Aleksandar Đorđević
Dragan RandjelovicDragan Randjelovic
Dragan Randjelovic
This paper presents a novel approach for an Intrusion Detection System (IDS) based on one kind of asymmetric optimization which use any three already well-known IDS algorithms and Triple Modular Redundancy (TMR) algorithm together. Namely, a variable threshold which indicates an attack on an observed and protected network is determined by using all three values obtained with three known IDS algorithms i.e., on previously recorded data by making a decision by majority. For these algorithms authors used algorithm of k-nearest neighbors, cumulative sum algorithm, and algorithm of exponentially weighted moving average. Using a proposed method we can get a threshold that is more precisely determined than in the case of any method individual. Practically, using TMR we obtain a dynamically threshold adjustment of IDS software, which reduces the existence of false alarms and undetected attacks, so the efficiency of such IDS software is notably higher and can get better results. Today, Denial of Service attacks (DoS) are one of the most present type of attacks and the reason for the special attention paid to them in this paper. In addition, the authors of the proposed method for IDS software used a known CIC-DDoS2019 dataset, which contains various data recordings of such attacks. Obtained results with the proposed solution showed better characteristics than each individual used algorithm in this solution. IDS software with the proposed method worked precisely and timely, which means alarms were triggered properly and efficiently.

20 October 2022 in Electronics
Verified
Journal Article
Open access
PDF file
NMNikola Maksimović
Nikola Maksimović
Milan ČabarkapaMilan Čabarkapa
Milan Čabarkapa
MTMarko Tanasković
Marko Tanasković
Dragan RandjelovicDragan Randjelovic
Dragan Randjelovic
Concerning occupational safety, the aim of ergonomics as a scientific discipline is to study and adjust working conditions, worker equipment, and work processes from a psychological, physiological, and anatomical aspect instead of adapting the worker to the needs of the job. This paper will discuss and analyze the potential of the garment-embedded body posture tracking sensor and its usage as standard working equipment, which is meant to help correct improper and high-risk upper body positions during prolonged and static work activities. The analysis evaluation cross-reference is based on the Rapid Upper Limb Assessment ergonomics risk assessment tool. Signals generated by the wearable are meant to help the wearer and observer promptly-continuously detect and correct bad posture. The results show a positive progression of workers’ body posture to reduce the ergonomic risks this research covers. It can be concluded that wearable technology and sensors would significantly contribute to the observer as the evaluation tool and the wearer to spot the risk factors promptly and self-correct them independently. This feature would help workers learn and improve the correct habits of correcting ergonomically incorrect body postures when performing work tasks.

6 July 2025 in Information, Communication & Society
Verified
Journal Article
Open access
LBLjubiša Bojić
Ljubiša Bojić
Jörg MatthesJörg Matthes
Jörg Matthes
Agariadne Dwinggo SamalaAgariadne Dwinggo Samala
Agariadne Dwinggo Samala
Milan ČabarkapaMilan Čabarkapa
Milan Čabarkapa