Papers by Urszula Boryczka
Entropy
The aim of a recommender system is to suggest to the user certain products or services that most ... more The aim of a recommender system is to suggest to the user certain products or services that most likely will interest them. Within the context of personalized recommender systems, a number of algorithms have been suggested to generate a ranking of items tailored to individual user preferences. However, these algorithms do not generate identical recommendations, and for this reason it has been suggested in the literature that the results of these algorithms can be combined using aggregation techniques, hoping that this will translate into an improvement in the quality of the final recommendation. In order to see which of these techniques increase the quality of recommendations to the greatest extent, the authors of this publication conducted experiments in which they considered five recommendation algorithms and 20 aggregation methods. The research was carried out on the popular and publicly available MovieLens 100k and MovieLens 1M datasets, and the results were confirmed by statist...
Entropy
This article presents the author’s own metaheuristic cryptanalytic attack based on the use of dif... more This article presents the author’s own metaheuristic cryptanalytic attack based on the use of differential cryptanalysis (DC) methods and memetic algorithms (MA) that improve the local search process through simulated annealing (SA). The suggested attack will be verified on a set of ciphertexts generated with the well-known DES (data encryption standard) reduced to six rounds. The aim of the attack is to guess the last encryption subkey, for each of the two characteristics Ω. Knowing the last subkey, it is possible to recreate the complete encryption key and thus decrypt the cryptogram. The suggested approach makes it possible to automatically reject solutions (keys) that represent the worst fitness function, owing to which we are able to significantly reduce the attack search space. The memetic algorithm (MASA) created in such a way will be compared with other metaheuristic techniques suggested in literature, in particular, with the genetic algorithm (NGA) and the classical differe...
Journal of Applied Computer Science, Dec 31, 2018
The development of computer graphics forces new requirements on the developers, which will make t... more The development of computer graphics forces new requirements on the developers, which will make the virtual world more similar to the real world. One of these elements is the simulation of fog. Common fog algorithms mix the color of the scene with the color of the fog over a certain distance. However, one feature of the naturally foggy scenery is ignored. With the distance and density of the fog, the observed scenery or individual objects become more blurred. In this paper we will present our implementation of the distance fog in the Unreal Engine 4, including the effect of blurring the foggy areas, simulating of light scattering and variations in fog density using noise.
LogForum, 2019
Background: Transportation Problem (TP) is a special case of integer programming, characterised b... more Background: Transportation Problem (TP) is a special case of integer programming, characterised by indisputable practical significance (in particular in the area of logistics). For this reason, many techniques have been proposed to solve the problem both in optimum and approximate manner. The problem of selecting an effective technique for determining a suboptimal solution for TP was addressed by many researchers, however the implementation of only certain heuristics, 'test bed' applied, as well as non-performance of statistical tests make it impossible to clearly identify the recommended approach to application of heuristics in TP, leaving a research gap which determined the writing of this article. The additional purpose of this paper is to provide a summary of selected approximate methods, taking into consideration the number of iterations necessary to design the optimal solution by means of Modified Distribution (MODI) method and to demonstrate potential correlations between the parameters describing a problem instance and the efficiency of the methods. Methods: This paper presents a comparative study of four classic techniques (NWC, LCM, VAM and RAM). The tests were performed on three sets of 2,500 pseudo-randomly generated tasks and the observations were also checked by means of the Wilcoxon Signed-Rank Test and Pearson correlation coefficient. Results: The results confirms that VAM is characterised by a significant quality of the determined results, whereas NWC develops solutions of low efficiency. However, contrary to the observations made for small TP instances, RAM was characterised by a higher error value than LCM for huge set, demonstrating the impossibility to generalise results obtained for small problems (presented e.g. in literature), in order to determine their efficiency for higher instances. Conclusions: It is recommended to apply VAM both for the determination of initial solution in MODI method and for performing allocation of resources, using only heuristics. However, taking into consideration the utilitarian approach and possible occurrence of the necessity to solve TP instances without using the appropriate software, it is recommended to use LCM for solving large instances of TP. The presence of strong correlation between the number of nodes describing the TP instance and the number of iterations necessary to determine the optimal solution by MODI method has been identified.
Intelligent Information and Database Systems, 2019
This article proposes three approaches to designing the Parallel Harmony Search Algorithm, adjust... more This article proposes three approaches to designing the Parallel Harmony Search Algorithm, adjusted to effectively solve the Asymmetric Traveling Salesman Problem. The paper contains a comparative study of the developed models, based on 19 varied instances of the problem, as well as recommendations concerning their appropriate configuration. The quality of developed models was assessed through determination of the percentage difference between the surplus of the values of objective function of solutions and the values describing the sequential algorithm variant. Each of the examined Parallel Harmony Search variants created better results than the sequential Harmony Search algorithm.
2017 IEEE International Conference on INnovations in Intelligent SysTems and Applications (INISTA), 2017
The number of highly sophisticated software-based attacks on ATMs is growing these days. New type... more The number of highly sophisticated software-based attacks on ATMs is growing these days. New types of threats require new ways of system protection. Most secure solutions are based on whitelists and sandboxes, which unlike antivirus solutions are able to protect the system against any new threat. Sadly, they are hard to configure therefore require an expert-level knowledge of operating system mechanisms and software security techniques. The main purpose of this article is to present the possibilities of using clustering algorithms in a configuration process of a sandbox-based security solution. Results of the experimental studies show that even basic clustering algorithms can be used as a part of the configuration process. Achieved results were deemed promising by the control algorithm and thus can be used as a base for future research.
Intelligent Information and Database Systems, 2019
The present article summarizes two techniques allowing to avoid premature convergence in Harmony ... more The present article summarizes two techniques allowing to avoid premature convergence in Harmony Search algorithm, which was adapted for solving the instances of the Asymmetric Traveling Salesman Problem (ATSP). The efficiency of both approaches was demonstrated on the basis of the results of statistical test and ‘test bed’ consisting of nineteen instances of ATSP. The conclusion was that the best results were obtained in case of applying mechanisms which enable to reset the components of harmony memory at the moment of reaching stagnation. This process is controlled by parameters which are depended on the problem size.
The development of computer graphics forces new requirements on the developers, which will make t... more The development of computer graphics forces new requirements on the developers, which will make the virtual world more similar to the real world. One of these elements is the simulation of fog. Common fog algorithms mix the color of the scene with the color of the fog over a certain distance. However, one feature of the naturally foggy scenery is ignored. With the distance and density of the fog, the observed scenery or individual objects become more blurred. In this paper we will present our implementation of the distance fog in the Unreal Engine 4, including the effect of blurring the foggy areas, simulating of light scattering and variations in fog density using noise.
Electronics, 2022
Recommendation systems are a powerful tool that is an integral part of a great many websites. Mos... more Recommendation systems are a powerful tool that is an integral part of a great many websites. Most often, recommendations are presented in the form of a list that is generated by using various recommendation methods. Typically, however, these methods do not generate identical recommendations, and their effectiveness varies between users. In order to solve this problem, the application of aggregation techniques was suggested, the aim of which is to combine several lists into one, which, in theory, should improve the overall quality of the generated recommendations. For this reason, we suggest using the Differential Evolution algorithm, the aim of which will be to aggregate individual lists generated by the recommendation algorithms and to create a single list that will be fine-tuned to the user’s preferences. Additionally, based on our previous research, we present suggestions to speed up this process.
Klucze dzikich ptaków czy ławice ryb w harmonijny i skoordynowany sposób podążają w danym kierunk... more Klucze dzikich ptaków czy ławice ryb w harmonijny i skoordynowany sposób podążają w danym kierunku w symetrycznym układzie. Gdy nagle zmieniają kierunek nie wpadają na siebie i nie łamią szyku. To prawdziwy przyrodniczy fenomen. I to właśnie takim fenomenom zaczęli przyglądać się badacze niekoniecznie związani z naukami przyrodniczymi, którzy na podstawie modeli matematycznych zaobserwowanych w świecie przyrody stworzyli algorytmy sztucznej inteligencji przydatne w rozwiązaniu konkretnych problemów optymalizacyjnych. Kolektywny wzorzec zachowań indywiduów (czyli poszczególnych osobników) w samoorganizujących się systemach, takich jak klucz ptaków, ławica ryb czy kolonia owadów, posłużył do stworzenia techniki sztucznej inteligencji zwanej inteligencją stadną (ang. Swarm Intelligence, SI).
This article presents the application of the differential evolution algorithm (DE), the aim of wh... more This article presents the application of the differential evolution algorithm (DE), the aim of which is to adjust weights to particular features of an active users profile, in order to render his preferences in the best possible way. In our system we applied a popular technique of collaborative filtering, which is used to generate recommendations. The users profile is a vector, which consists of values, which characterize a given user. Using a hybrid feature made it possible to use simple weighted Euclidean distance, which significantly decreased the amount of necessary computations and made it possible to compare particular profiles in a system faster. The results of the conducted experiments were compared with a modified weighted Euclidean distance and Pearsons correlation.
This paper describes an improvement to the Harmony Search algorithm, which has been adjusted to e... more This paper describes an improvement to the Harmony Search algorithm, which has been adjusted to effectively solve a problem with indisputable practical significance, i.e. Asymmetric Traveling Salesman Problem (ATSP). We modify the technique structure, enabling the value of PAR parameter to be changed dynamically, which has an impact on the frequency of greedy movements during the construction of another harmony. The article demonstrates the effectiveness of the described approach and presents a comparative study of three sets of characteristic PAR values used during the method execution. The research was conducted on a ‘test bed’ consisting of nineteen instances of the ATSP.
Procedia Computer Science, 2021
Recommendation systems can suggest users list of items they have not yet seen but might be intere... more Recommendation systems can suggest users list of items they have not yet seen but might be interested in. To improve the quality of the generated recommendations, different techniques are often used which try to personalize recommendations. Usually user preferences are stored in the form of a vector in which individual values describe to what extent a given feature is desired by the user. To find this vector, metaheuristic algorithms can be used, however their main drawback is their computational complexity. Therefore, in this paper, a modification of the Differential Evolution algorithm is proposed to enable faster computation of the ranking score for each item in the system, which is used to create a recommendation list. Experiments have been performed on the current MovieLens 25m database and they show that our modification can significantly speed up the process of finding a preference vector, without losing their quality for the top-N recommendation task. We will also address the vulnerability of recommendation systems to profile injection attacks, as a result of which an attacker can influence the generated recommendations.
Engineering Optimization, 2019
This article is part of a study aiming to adapt the Harmony Search (HS) algorithm to the asymmetr... more This article is part of a study aiming to adapt the Harmony Search (HS) algorithm to the asymmetric travelling salesman problem. The populationbased approach of the HS algorithm and its weak exploitation mechanism constituted a good reason for starting the research on improving the method, by applying the approach found in other population-based metaheuristics characterized by the same imperfection. The proposed HS is supplemented with the mechanism allowing the efficiency of the exploitation process to be increased by combining the algorithm with hill climbing. The proposed approach assumes modifying the solutions at three different places, affecting the HS exploration and exploitation mechanism in different ways. As a result, the appropriate approach to the effective design of hybrid harmony search is recommended. In addition, by measuring the actual time of the method's execution, the recommended ratio of time devoted to local search and the creation of harmonies was determined.
Expert Systems with Applications, 2018
This article constitutes the continuation of the work on adapting the Harmony Search algorithm to... more This article constitutes the continuation of the work on adapting the Harmony Search algorithm to effectively solve the Asymmetric Traveling Salesman Problem (ATSP) instances. The author's modification suggested in this work enables the improvement of harmony memory during the time of running the technique, making it possible to increase the effectiveness of the technique by almost 59% (the summary average error was reduced from 13.42 to 5.54%) and to eliminate the imperfectness revealed in the previous research. The article includes a description of the approach and a comparative study spanning various variants of implementing the improvement, conducted on 19 instances of ATSP, described by means of the occurrence of 17 to 443 nodes. The achieved results were also compared with the results found in reference articles, showing a significant effectiveness of the modification.
Smart Innovation, Systems and Technologies, 2016
This paper presents a new approach to an automatic categorization of email messages into mailbox ... more This paper presents a new approach to an automatic categorization of email messages into mailbox folders. The aim of this paper is to create an algorithm that would allow one to improve the classification of emails into folders by using solutions that have been applied in Ant Colony Decision Tree (ACDT). Additionally, elements of Social Network Analysis (SNA) were included in this algorithm. The new algorithm that is proposed here was tested on the publicly available Enron E-mail data set and all experiments were conducted on uncleaned data. For the purpose of comparing the results, additional tests were carried out by using selected classifiers which were generally available. The obtained results confirm that the proposed approach allows one to improve the accuracy with which new emails are assigned to particular folders based on an analysis of previous correspondence, even when uncleaned data sets are used.
Computational Collective Intelligence, 2015
This paper presents a new approach to an automatic categorization of email messages into folders.... more This paper presents a new approach to an automatic categorization of email messages into folders. The aim of this paper is to create a new algorithm that will allow one to improve the accuracy with which emails are assigned to folders (Email Foldering Problem) by using solutions that have been applied in Ant Colony Optimization algorithms (ACO) and Social Networks Analysis (SNA). The new algorithm that is proposed here has been tested on the publicly available Enron email data set. The obtained results confirm that this approach allows one to better organize new emails into folders based on an analysis of previous correspondence.
Procedia Computer Science, 2020
The recommendation systems are used to suggest new, still not discovered items to users. At the m... more The recommendation systems are used to suggest new, still not discovered items to users. At the moment, in order to achieve the best quality of the generated recommendations, users and their choices in the system must be analyzed to create a certain profile of preferences for a given user in order to adjust the generated recommendation to his personal taste. This article will present a recommendation system, which based on the Differential Evolution (DE) algorithm will learn the ranking function while directly optimizing the average precision (AP) for the selected user in the system. To achieve that, items are represented through a feature vectors generated using user-item matrix factorization. The experiments have been conducted on a popular and widely available public dataset MovieLens, and show that our approach in certain situations can significantly improve the quality of the generated recommendations. Results of experiments are compared with other techniques.
Applied Soft Computing, 2009
Intelligent Information and Database Systems, 2015
In this article an approach to the automatic classification of email messages in mailboxes has be... more In this article an approach to the automatic classification of email messages in mailboxes has been proposed. The aim of this paper is to devise methods to build decision tables from the collection of email messages on which it is possible to build Ant Colony Optimization-based ensemble classifiers, whose application allows to use the collection of emails without cleaning, at the same time improving the accuracy of the email folders classification. The proposed method has been tested by the selected algorithms on the Enron Email Dataset. The results confirm that the proposed solutions allows to improve the accuracy of classification of new emails to folders.
Uploads
Papers by Urszula Boryczka