Barbara Probierz

Dr. Barbara Probierz

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
1 December 2021 in Entropy
Verified
Journal Article
Open access
PDF file
Przemysław JuszczukPrzemysław Juszczuk
Przemysław Juszczuk
Jan KozakJan Kozak
Jan Kozak
GDGrzegorz Dziczkowski
Grzegorz Dziczkowski
Szymon GłowaniaSzymon Głowania
Szymon Głowania
TJTomasz Jach
Tomasz Jach
Barbara ProbierzBarbara Probierz
Barbara Probierz
In the era of the Internet of Things and big data, we are faced with the management of a flood of information. The complexity and amount of data presented to the decision-maker are enormous, and existing methods often fail to derive nonredundant information quickly. Thus, the selection of the most satisfactory set of solutions is often a struggle. This article investigates the possibilities of using the entropy measure as an indicator of data difficulty. To do so, we focus on real-world data covering various fields related to markets (the real estate market and financial markets), sports data, fake news data, and more. The problem is twofold: First, since we deal with unprocessed, inconsistent data, it is necessary to perform additional preprocessing. Therefore, the second step of our research is using the entropy-based measure to capture the nonredundant, noncorrelated core information from the data. Research is conducted using well-known algorithms from the classification domain to investigate the quality of solutions derived based on initial preprocessing and the information indicated by the entropy measure. Eventually, the best 25% (in the sense of entropy measure) attributes are selected to perform the whole classification procedure once again, and the results are compared.

10 April 2022 in Entropy
Verified
Journal Article
Open access
PDF file
Jan KozakJan Kozak
Jan Kozak
Barbara ProbierzBarbara Probierz
Barbara Probierz
KKKrzysztof Kania
Krzysztof Kania
Przemysław JuszczukPrzemysław Juszczuk
Przemysław Juszczuk
Classification is one of the main problems of machine learning, and assessing the quality of classification is one of the most topical tasks, all the more difficult as it depends on many factors. Many different measures have been proposed to assess the quality of the classification, often depending on the application of a specific classifier. However, in most cases, these measures are focused on binary classification, and for the problem of many decision classes, they are significantly simplified. Due to the increasing scope of classification applications, there is a growing need to select a classifier appropriate to the situation, including more complex data sets with multiple decision classes. This paper aims to propose a new measure of classifier quality assessment (called the preference-driven measure, abbreviated p-d), regardless of the number of classes, with the possibility of establishing the relative importance of each class. Furthermore, we propose a solution in which the classifier’s assessment can be adapted to the analyzed problem using a vector of preferences. To visualize the operation of the proposed measure, we present it first on an example involving two decision classes and then test its operation on real, multi-class data sets. Additionally, in this case, we demonstrate how to adjust the assessment to the user’s preferences. The results obtained allow us to confirm that the use of a preference-driven measure indicates that other classifiers are better to use according to preferences, particularly as opposed to the classical measures of classification quality assessment.

16 February 2025 in Applied Sciences
Verified
Journal Article
Open access
PDF file
BBBartłomiej Brzęk
Bartłomiej Brzęk
Barbara ProbierzBarbara Probierz
Barbara Probierz
Jan KozakJan Kozak
Jan Kozak
This paper investigates the application of genetic algorithms (GAs) for hyperparameter optimisation in deep reinforcement learning (RL), focusing on the Deep Q-Learning (DQN) algorithm. This study aims to identify approaches that enhance RL model performance through the effective exploration of the configuration space. By comparing different GA methods for selection, crossover, and mutation, this study focuses on deep RL models. The results indicate that GA techniques emphasising the exploration of the configuration space yield significant improvements in optimisation efficiency, reducing training time and enhancing convergence. The most effective GA improved the fitness function value from 68.26 (initial best chromosome) to 979.16 after 200 iterations, demonstrating the efficacy of the proposed approach. Furthermore, variations in specific hyperparameters, such as learning rate, gamma, and update frequency, were shown to substantially affect the DQN model’s learning ability. These findings suggest that exploration-driven GA strategies outperform GA approaches with limited exploration, underscoring the critical role of selection and crossover methods in enhancing DQN model efficiency and performance. Moreover, a mini case study on the CartPole environment revealed that even a 5% sensor dropout impaired the performance of a GA-optimised RL agent, while a 20% dropout almost entirely halted improvements.

26 October 2024 in Applied Sciences
Verified
Journal Article
Open access
PDF file
NKNatalia Anna Krawczyk
Natalia Anna Krawczyk
Barbara ProbierzBarbara Probierz
Barbara Probierz
Jan KozakJan Kozak
Jan Kozak
The detection of essays written by AI compared to those authored by students is increasingly becoming a significant issue in educational settings. This research examines various numerical text representation techniques to improve the classification of these essays. Utilizing a diverse dataset, we undertook several preprocessing steps, including data cleaning, tokenization, and lemmatization. Our system analyzes different text representation methods such as Bag of Words, TF-IDF, and fastText embeddings in conjunction with multiple classifiers. Our experiments showed that TF-IDF weights paired with logistic regression reached the highest accuracy of 99.82%. Methods like Bag of Words, TF-IDF, and fastText embeddings achieved accuracies exceeding 96.50% across all tested classifiers. Sentence embeddings, including MiniLM and distilBERT, yielded accuracies from 93.78% to 96.63%, indicating room for further refinement. Conversely, pre-trained fastText embeddings showed reduced performance, with a lowest accuracy of 89.88% in logistic regression. Remarkably, the XGBoost classifier delivered the highest minimum accuracy of 96.24%. Specificity and precision were above 99% for most methods, showcasing high capability in differentiating between student-created and AI-generated texts. This study underscores the vital role of choosing dataset-specific text representations to boost classification accuracy.

27 June 2023 in Electronics
Verified
Journal Article
Open access
PDF file
Barbara ProbierzBarbara Probierz
Barbara Probierz
AHAnita Hrabia
Anita Hrabia
Jan KozakJan Kozak
Jan Kozak
Natural language processing is still an emerging field in machine learning. Access to more and more data sets in textual form, new applications for artificial intelligence and the need for simple communication with operating systems all simultaneously affect the importance of natural language processing in evolving artificial intelligence. Traditional methods of textual representation, such as Bag-of-Words, have some limitations that result from the lack of consideration of semantics and dependencies between words. Therefore, we propose a new approach based on graph representations, which takes into account both local context and global relationships between words, allowing for a more expressive textual representation. The aim of the paper is to examine the possibility of using graph representations in natural language processing and to demonstrate their use in text classification. An innovative element of the proposed approach is the use of common cliques in graphs representing documents to create a feature vector. Experiments confirm that the proposed approach can improve classification efficiency. The use of a new text representation method to predict book categories based on the analysis of its content resulted in accuracy, precision, recall and an F1-score of over 90%. Moving from traditional approaches to a graph-based approach could make a big difference in natural language processing and text analysis and could open up new opportunities in the field.

9 January 2020 in Neural Computing and Applications
Verified
Journal Article
Closed access
PDF file
Jan KozakJan Kozak
Jan Kozak
Przemysław JuszczukPrzemysław Juszczuk
Przemysław Juszczuk
Barbara ProbierzBarbara Probierz
Barbara Probierz