TY  - JOUR
AU  - Stojanović, Vladica
AU  - Ranđelović, Dragan
AU  - Kuk, Kristijan
PY  - 2018
UR  - http://jakov.kpu.edu.rs/handle/123456789/846
AB  - This paper presents a modification and, at the same time, a generalization of the linear first order nonnegative integer-valued autoregressive processes, well-known as INAR(1) processes. By using the so-called Noise-Indicator, a nonlinear model with the threshold regime and with more complex structure than the appropriate linear models was obtained. The new model, named NIINAR(1) process, has been investigated in terms of the most general, the power series distribution of its innovations. Basic stochastic properties of the NIINAR(1) model (e.g., correlation structure, over-dispersion conditions and distributional properties) are given. Also, besides of some standard parameters estimators, a novel estimation techniques, together with the asymptotic properties of the obtained estimates is described. At last, a Monte Carlo study of this process is also given, as well as its application in the analysis of dynamics of two empirical dataset.
PB  - Brazilian Statistical Association, Sao Paulo
T2  - Brazilian journal of probability and statistics
T1  - Noise-indicator nonnegative integer-valued autoregressive time series of the first order
VL  - 32
IS  - 1
SP  - 147
EP  - 171
DO  - 10.1214/16-BJPS335
ER  - 

@article{
author = "Stojanović, Vladica and Ranđelović, Dragan and Kuk, Kristijan",
year = "2018",
abstract = "This paper presents a modification and, at the same time, a generalization of the linear first order nonnegative integer-valued autoregressive processes, well-known as INAR(1) processes. By using the so-called Noise-Indicator, a nonlinear model with the threshold regime and with more complex structure than the appropriate linear models was obtained. The new model, named NIINAR(1) process, has been investigated in terms of the most general, the power series distribution of its innovations. Basic stochastic properties of the NIINAR(1) model (e.g., correlation structure, over-dispersion conditions and distributional properties) are given. Also, besides of some standard parameters estimators, a novel estimation techniques, together with the asymptotic properties of the obtained estimates is described. At last, a Monte Carlo study of this process is also given, as well as its application in the analysis of dynamics of two empirical dataset.",
publisher = "Brazilian Statistical Association, Sao Paulo",
journal = "Brazilian journal of probability and statistics",
title = "Noise-indicator nonnegative integer-valued autoregressive time series of the first order",
volume = "32",
number = "1",
pages = "147-171",
doi = "10.1214/16-BJPS335"
}

Stojanović, V., Ranđelović, D.,& Kuk, K.. (2018). Noise-indicator nonnegative integer-valued autoregressive time series of the first order. in Brazilian journal of probability and statistics
Brazilian Statistical Association, Sao Paulo., 32(1), 147-171.
https://doi.org/10.1214/16-BJPS335

Stojanović V, Ranđelović D, Kuk K. Noise-indicator nonnegative integer-valued autoregressive time series of the first order. in Brazilian journal of probability and statistics. 2018;32(1):147-171.
doi:10.1214/16-BJPS335 .

Stojanović, Vladica, Ranđelović, Dragan, Kuk, Kristijan, "Noise-indicator nonnegative integer-valued autoregressive time series of the first order" in Brazilian journal of probability and statistics, 32, no. 1 (2018):147-171,
https://doi.org/10.1214/16-BJPS335 . .

TY  - JOUR
AU  - Kuk, Kristijan
AU  - Milentijević, Ivan Z.
AU  - Ranđelović, Dragan
AU  - Popović, Brankica
AU  - Čisar, Petar
PY  - 2017
UR  - http://jakov.kpu.edu.rs/handle/123456789/756
AB  - A bot is one of the main elements of all computer video games, frequently used for the creation of various opponent characters within a game. Opponent modeling is the problem of predicting the agent actions in a gaming environment. This paper proposes and describes the implementation of a bot as a personal opponent in a small educational game. In order to increase the efficiency when using such a small educational application/module, artificial intelligence was added in the form of a bot competing with the students. Pedagogical elements of the intelligent learning system are introduced through the pedagogical model and the student model. This paper demonstrates the use of the student model to present the player model built by the experience of a human teacher, with true/false questions incorporated with the bot strategy into the opponent model. The authors use the Monte Carlo approach in this implementation, known as artificial intelligence technique and a best-first search method used in most video games, but to the best of their knowledge, it has not been used for prediction in educational games based on bot strategy. The results highlight that the Monte Carlo approach presented via the BFTree classifier provides the best classification accuracy compared with other predictive models based on data mining classifiers. It was shown that the training data from the human player can help in creating a bot strategy for a personalized game-based learning system. The Help option can be used for the assessment of the students' current knowledge by counting the number of Help option accesses, the player relies on Help as a 'source of knowledge' needed to complete the game task successfully. The obtained results show that the bot (personal opponent) stimulated players to replay the game multiple times, which may contribute to the increase of the students' knowledge.
PB  - Budapest Tech, Budapest
T2  - Acta polytechnica hungarica
T1  - The Design of the Personal Enemy - MIMLeBot as an Intelligent Agent in a Game-based Learning Environment
VL  - 14
IS  - 4
SP  - 121
EP  - 139
DO  - 10.12700/APH.14.4.2017.4.7
ER  - 

@article{
author = "Kuk, Kristijan and Milentijević, Ivan Z. and Ranđelović, Dragan and Popović, Brankica and Čisar, Petar",
year = "2017",
abstract = "A bot is one of the main elements of all computer video games, frequently used for the creation of various opponent characters within a game. Opponent modeling is the problem of predicting the agent actions in a gaming environment. This paper proposes and describes the implementation of a bot as a personal opponent in a small educational game. In order to increase the efficiency when using such a small educational application/module, artificial intelligence was added in the form of a bot competing with the students. Pedagogical elements of the intelligent learning system are introduced through the pedagogical model and the student model. This paper demonstrates the use of the student model to present the player model built by the experience of a human teacher, with true/false questions incorporated with the bot strategy into the opponent model. The authors use the Monte Carlo approach in this implementation, known as artificial intelligence technique and a best-first search method used in most video games, but to the best of their knowledge, it has not been used for prediction in educational games based on bot strategy. The results highlight that the Monte Carlo approach presented via the BFTree classifier provides the best classification accuracy compared with other predictive models based on data mining classifiers. It was shown that the training data from the human player can help in creating a bot strategy for a personalized game-based learning system. The Help option can be used for the assessment of the students' current knowledge by counting the number of Help option accesses, the player relies on Help as a 'source of knowledge' needed to complete the game task successfully. The obtained results show that the bot (personal opponent) stimulated players to replay the game multiple times, which may contribute to the increase of the students' knowledge.",
publisher = "Budapest Tech, Budapest",
journal = "Acta polytechnica hungarica",
title = "The Design of the Personal Enemy - MIMLeBot as an Intelligent Agent in a Game-based Learning Environment",
volume = "14",
number = "4",
pages = "121-139",
doi = "10.12700/APH.14.4.2017.4.7"
}

Kuk, K., Milentijević, I. Z., Ranđelović, D., Popović, B.,& Čisar, P.. (2017). The Design of the Personal Enemy - MIMLeBot as an Intelligent Agent in a Game-based Learning Environment. in Acta polytechnica hungarica
Budapest Tech, Budapest., 14(4), 121-139.
https://doi.org/10.12700/APH.14.4.2017.4.7

Kuk K, Milentijević IZ, Ranđelović D, Popović B, Čisar P. The Design of the Personal Enemy - MIMLeBot as an Intelligent Agent in a Game-based Learning Environment. in Acta polytechnica hungarica. 2017;14(4):121-139.
doi:10.12700/APH.14.4.2017.4.7 .

Kuk, Kristijan, Milentijević, Ivan Z., Ranđelović, Dragan, Popović, Brankica, Čisar, Petar, "The Design of the Personal Enemy - MIMLeBot as an Intelligent Agent in a Game-based Learning Environment" in Acta polytechnica hungarica, 14, no. 4 (2017):121-139,
https://doi.org/10.12700/APH.14.4.2017.4.7 . .

TY  - JOUR
AU  - Nikolić, Vojkan
AU  - Markoski, Branko
AU  - Kuk, Kristijan
AU  - Ranđelović, Dragan
AU  - Čisar, Petar
PY  - 2017
UR  - http://jakov.kpu.edu.rs/handle/123456789/750
AB  - Today, e-Governments services operate quick answer systems in many different domains in order to help citizens receive answers to their questions at any time and within a very short time. The automatic mapping of relevant documents stands out as an important application for automatic questions-documents classification strategies. This paper presents a contribution to the identification concepts in text comprehension in unstructured documents as an important step towards clarifying the role of explicit concepts in information retrieval in general. The most common representational scheme in text categorization is the Bag of Words approach when the dictionary, as incorporating background knowledge, is large. The authors introduce a new approach to create conceptbased text representations, and apply it to a text categorization collection to create predefined classes in the case of short text document analysis problem. Further in this paper, a classification-based algorithm for questions matching topic modelling is presented. The paper also describes the weighting of concepts that present terms with high frequency of occurrence in questions is based on their similarity relevance in the predefined classes of documents. The results of the experiment within the criminal law domain in the present case study show that the performance of concept-based text representations has proven to be satisfactory in the case when there is no dictionary for this domain.
PB  - Budapest Tech, Budapest
T2  - Acta polytechnica hungarica
T1  - Modelling the System of Receiving Quick Answers for e-Government Services: Study for the Crime Domain in the Republic of Serbia
VL  - 14
IS  - 8
SP  - 143
EP  - 163
DO  - 10.12700/APH.14.8.2017.8.8
ER  - 

@article{
author = "Nikolić, Vojkan and Markoski, Branko and Kuk, Kristijan and Ranđelović, Dragan and Čisar, Petar",
year = "2017",
abstract = "Today, e-Governments services operate quick answer systems in many different domains in order to help citizens receive answers to their questions at any time and within a very short time. The automatic mapping of relevant documents stands out as an important application for automatic questions-documents classification strategies. This paper presents a contribution to the identification concepts in text comprehension in unstructured documents as an important step towards clarifying the role of explicit concepts in information retrieval in general. The most common representational scheme in text categorization is the Bag of Words approach when the dictionary, as incorporating background knowledge, is large. The authors introduce a new approach to create conceptbased text representations, and apply it to a text categorization collection to create predefined classes in the case of short text document analysis problem. Further in this paper, a classification-based algorithm for questions matching topic modelling is presented. The paper also describes the weighting of concepts that present terms with high frequency of occurrence in questions is based on their similarity relevance in the predefined classes of documents. The results of the experiment within the criminal law domain in the present case study show that the performance of concept-based text representations has proven to be satisfactory in the case when there is no dictionary for this domain.",
publisher = "Budapest Tech, Budapest",
journal = "Acta polytechnica hungarica",
title = "Modelling the System of Receiving Quick Answers for e-Government Services: Study for the Crime Domain in the Republic of Serbia",
volume = "14",
number = "8",
pages = "143-163",
doi = "10.12700/APH.14.8.2017.8.8"
}

Nikolić, V., Markoski, B., Kuk, K., Ranđelović, D.,& Čisar, P.. (2017). Modelling the System of Receiving Quick Answers for e-Government Services: Study for the Crime Domain in the Republic of Serbia. in Acta polytechnica hungarica
Budapest Tech, Budapest., 14(8), 143-163.
https://doi.org/10.12700/APH.14.8.2017.8.8

Nikolić V, Markoski B, Kuk K, Ranđelović D, Čisar P. Modelling the System of Receiving Quick Answers for e-Government Services: Study for the Crime Domain in the Republic of Serbia. in Acta polytechnica hungarica. 2017;14(8):143-163.
doi:10.12700/APH.14.8.2017.8.8 .

Nikolić, Vojkan, Markoski, Branko, Kuk, Kristijan, Ranđelović, Dragan, Čisar, Petar, "Modelling the System of Receiving Quick Answers for e-Government Services: Study for the Crime Domain in the Republic of Serbia" in Acta polytechnica hungarica, 14, no. 8 (2017):143-163,
https://doi.org/10.12700/APH.14.8.2017.8.8 . .