Search

Brain tumor segmentation, among various tasks in medical image analysis, has garnered significant attention in the research community. Despite continuous efforts by researchers, accurate brain tumor segmentation remains a key challenge. This challenge arises due to various factors, including location uncertainty, morphological uncertainty, low contrast imaging, annotation bias, and data imbalance. Magnetic resonance imaging (MRI) plays a vital role in providing detailed images of the brain, enabling the extraction of crucial information about the tumor’s shape, size, and location. In literature, deep learning algorithms have shown their efficiency in dealing with semantic segmentation, particularly the U-Net architecture. The latter has demonstrated impressive performance in Medical image segmentation. In this paper, a U-Net-based architecture for brain tumor segmentation is proposed. To further enhance the segmentation performance of our model, a novel ensemble learning method, EnsUNet, is introduced by integrating four pre-trained networks namely MobileNet, DeepLabV3+, ResNet, and DenseNet as the encoder within the U-Net architecture. The conducted experimental evaluation demonstrates promising results, achieving an Intersection over Union (IoU) score of 0.86, a Dice Coefficient (DC) of 0.92, and an accuracy of approximately 0.99. These findings underscore the effectiveness of our proposed EnsUNet for accurately segmenting brain tumors. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

Story generation requires AI systems to comprehend, create, and convey coherent and engaging narratives through text generation generation techniques. Studying story generation in natural language processing (NLP) is crucial in recent days as it enables AI systems to generate texts that resonate with human experiences, fostering more engaging interactions and facilitating applications ranging from virtual assistants to personalized content creation. To enhance the quality of generated stories, recently researchers have been exploring the use of higher-level elements such as plots and common sense knowledge to make the stories more than just grammarly correct texts. Leveraging zero-shot learning with large language models (LLMs), for example ChatGPT by OpenAI, has demonstrated remarkable performance on a variety of natural language processing tasks. This thesis conducts a thorough examination, utilizing both automated and human evaluation methods, to compare the storytelling capabilities of large language models (GPT-3.5-turbo, Gemma-2b), state-of-the-art storytelling models (SimCTG) and classic pre-trained models (BART, GPT-2). For two large language models (LLMs), zero-shot learning is used without any fine-tuning, while the other three models are fine-tuned. The models are compared across two story datasets, which feature variations in genre and story length. The findings reveal that large language models (LLMs) can produce stories with significantly superior quality compared to other types of models even without specific fine-tuning., Berättelsegenerering kräver AI-system för att förstå, skapa och förmedla sammanhängande och engagerande berättelser genom tekniker för generering av text. Att studera berättelsegenerering i naturlig språkbehandling (NLP) är avgörande de senaste dagarna eftersom det gör det möjligt för AI-system att generera texter som resonerar med mänskliga erfarenheter, främjar mer engagerande interaktioner och underlättar applikationer från virtuella assistenter till personligt skapande av innehåll. För att förbättra kvaliteten på genererade berättelser har forskare nyligen undersökt användningen av element på högre nivå, såsom intrig och sunt förnuft, för att göra berättelserna till mer än bara grammatiskt korrekta texter. Att utnyttja noll-shot-inlärning med stora språkmodeller (LLM), till exempel ChatGPT av OpenAI, har visat en anmärkningsvärd prestation på en mängd olika naturliga språkbearbetningsuppgifter. Det här examensarbetet genomför en grundlig granskning, med användning av både automatiserade och mänskliga utvärderingsmetoder, för att jämföra berättarförmågan hos stora språkmodeller (GPT-3.5-turbo, Gemma-2b), toppmoderna berättarmodeller (SimCTG) och klassiska förtränade modeller (BART, GPT-2). För två stora språkmodeller (LLM) används zero-shot learning utan någon finjustering, medan de andra tre modellerna är finjusterade. Modellerna jämförs över två berättelsedataset, som har variationer i genre och berättelselängd. Resultaten visar att stora språkmodeller (LLM) kan producera berättelser med avsevärt överlägsen kvalitet jämfört med andra typer av modeller även utan specifik finjustering.

Story generation requires AI systems to comprehend, create, and convey coherent and engaging narratives through text generation generation techniques. Studying story generation in natural language processing (NLP) is crucial in recent days as it enables AI systems to generate texts that resonate with human experiences, fostering more engaging interactions and facilitating applications ranging from virtual assistants to personalized content creation. To enhance the quality of generated stories, recently researchers have been exploring the use of higher-level elements such as plots and common sense knowledge to make the stories more than just grammarly correct texts. Leveraging zero-shot learning with large language models (LLMs), for example ChatGPT by OpenAI, has demonstrated remarkable performance on a variety of natural language processing tasks. This thesis conducts a thorough examination, utilizing both automated and human evaluation methods, to compare the storytelling capabilities of large language models (GPT-3.5-turbo, Gemma-2b), state-of-the-art storytelling models (SimCTG) and classic pre-trained models (BART, GPT-2). For two large language models (LLMs), zero-shot learning is used without any fine-tuning, while the other three models are fine-tuned. The models are compared across two story datasets, which feature variations in genre and story length. The findings reveal that large language models (LLMs) can produce stories with significantly superior quality compared to other types of models even without specific fine-tuning., Berättelsegenerering kräver AI-system för att förstå, skapa och förmedla sammanhängande och engagerande berättelser genom tekniker för generering av text. Att studera berättelsegenerering i naturlig språkbehandling (NLP) är avgörande de senaste dagarna eftersom det gör det möjligt för AI-system att generera texter som resonerar med mänskliga erfarenheter, främjar mer engagerande interaktioner och underlättar applikationer från virtuella assistenter till personligt skapande av innehåll. För att förbättra kvaliteten på genererade berättelser har forskare nyligen undersökt användningen av element på högre nivå, såsom intrig och sunt förnuft, för att göra berättelserna till mer än bara grammatiskt korrekta texter. Att utnyttja noll-shot-inlärning med stora språkmodeller (LLM), till exempel ChatGPT av OpenAI, har visat en anmärkningsvärd prestation på en mängd olika naturliga språkbearbetningsuppgifter. Det här examensarbetet genomför en grundlig granskning, med användning av både automatiserade och mänskliga utvärderingsmetoder, för att jämföra berättarförmågan hos stora språkmodeller (GPT-3.5-turbo, Gemma-2b), toppmoderna berättarmodeller (SimCTG) och klassiska förtränade modeller (BART, GPT-2). För två stora språkmodeller (LLM) används zero-shot learning utan någon finjustering, medan de andra tre modellerna är finjusterade. Modellerna jämförs över två berättelsedataset, som har variationer i genre och berättelselängd. Resultaten visar att stora språkmodeller (LLM) kan producera berättelser med avsevärt överlägsen kvalitet jämfört med andra typer av modeller även utan specifik finjustering.

Transformer-based models attain excellent results and generalize well when trained on sufficient amounts of data. However, constrained by the limited data available in the audio domain, most transformer-based models for audio tasks are finetuned from pre-trained models in other domains (e.g. image), which has a notable gap with the audio domain. Other methods explore the self-supervised learning approaches directly in the audio domain but currently do not perform well in the downstream tasks. In this paper, we present a novel self-supervised learning method for transformer-based audio models, called masked spectrogram prediction (MaskSpec), to learn powerful audio representations from unlabeled audio data (AudioSet used in this paper). Our method masks random patches of the input spectrogram and reconstructs the masked regions with an encoder-decoder architecture. Experimental results demonstrate MaskSpec reaches the performance of 0.471 (mAP) on AudioSet, 0.854 (mAP) on Open-MIC2018, 0.982 (accuracy) on ESC-50, 0.976 (accuracy) on SCV2, and 0.823 (accuracy) on DCASE2019 Task1A. The source code and pre-trained models have been released.1 © 2023 IEEE.

The incremental learning paradigm in machine learning has consistently been a focus of academic research. It is similar to the way in which biological systems learn, and reduces energy consumption by avoiding excessive retraining. Existing studies utilize the powerful feature extraction capabilities of pre-trained models to address incremental learning, but there remains a problem of insufficient utilization of neural network feature knowledge. To address this issue, this paper proposes a novel method called Pre-trained Model Knowledge Distillation (PMKD) which combines knowledge distillation of neural network representations and replay. This paper designs a loss function based on centered kernel alignment to transfer neural network representations knowledge from the pre-trained model to the incremental model layer-by-layer. Additionally, the use of memory buffer for Dark Experience Replay helps the model retain past knowledge better. Experiments show that PMKD achieved superior performance on various datasets and different buffer sizes. Compared to other methods, our class incremental learning accuracy reached the best performance. The open-source code is published athttps://github.com/TianSongS/PMKD-IL. © 2023 The Author(s), Funding: National Natural Science Foundation of China (No. 61901436)

Research in source code summarization, that is the description of the functionality of a program with short sentences expressed in natural language, is a topic of great interest in the software engineering community, since it can help in automatically generating software documentation, and in general can ease the effort of the developers in understanding the code they are working on. In this work, which is conceived as a negative results paper, we study the existing neural models designed for this purpose, pointing out their high sensitivity to the natural elements present in the source code (i.e. comments and identifiers) and the related drop in performance when such elements are ablated or masked. We then propose a novel source code summarization approach based on the aid of an intermediate pseudo-language, through which we are able to fine-tune the BRIO model for natural language on source code summarization, and to achieve results comparable to that obtained by the state-of-the-art source code competitors (e.g. PLBART and CodeBERT). We finally discuss about the limitations of these NLP-based approaches when transferred in the domain of source code processing, and we provide some insights for further research directions.

The main purpose of our project is to implement the device in Rural areas which enables the village people to use the schemes offered by the competent authorities. Here, we are training the device in the local language (Kannada) using AI and ML, so that the rural people who are more comfortable with their local language (Kannada) will be able to get the information about the government schemes. This device is trained to tell the usage of particular tablets. It helps the farmer in the detection of crop disease (leaf) and gives suitable remedies for the particular disease.

The main purpose of our project is to implement the device in Rural areas which enables the village people to use the schemes offered by the competent authorities. Here, we are training the device in the local language (Kannada) using AI and ML, so that the rural people who are more comfortable with their local language (Kannada) will be able to get the information about the government schemes. This device is trained to tell the usage of particular tablets. It helps the farmer in the detection of crop disease (leaf) and gives suitable remedies for the particular disease.

The main purpose of our project is to implement the device in Rural areas which enables the village people to use the schemes offered by the competent authorities. Here, we are training the device in the local language (Kannada) using AI and ML, so that the rural people who are more comfortable with their local language (Kannada) will be able to get the information about the government schemes. This device is trained to tell the usage of particular tablets. It helps the farmer in the detection of crop disease (leaf) and gives suitable remedies for the particular disease.

The main purpose of our project is to implement the device in Rural areas which enables the village people to use the schemes offered by the competent authorities. Here, we are training the device in the local language (Kannada) using AI and ML, so that the rural people who are more comfortable with their local language (Kannada) will be able to get the information about the government schemes. This device is trained to tell the usage of particular tablets. It helps the farmer in the detection of crop disease (leaf) and gives suitable remedies for the particular disease.

With the rapid growth of Internet users around the world, languages have become crucial to our ability to access information, education, and communication on the Internet. In the case of people such as us who are not native English speakers, the language barrier makes it difficult for us to access the information. There is no doubt that Thai and Myanmar languages are included in the low-resource language group. The majority of natural language processing (NLP) research has been conducted on high-resource languages, whereas research on low-resource languages has been scarce. Based on this research gap, the authors decided to concentrate on the development of NLP research for our low-resource languages. Besides the low-resource data problem, finding an appropriate MT method due to the difference in nature of the languages is a challenging task for our thesis. When considering theappropriate MT approach for Thai and Myanmar languages, reverse grammatical structures as well as syllable-based written forms are key factors to consider. To solve these research problems, our thesis proposes three methods for improving machine translation (MT) performance for low-resource English-Thai-Myanmar language pairs, including 1) EDITOR and Levenshtein transformer (LevT) models, 2) SwitchOut data augmentation algorithm, and 3) mBART-50 pre-trained model. As part of our contribution to low-resource MT, we built our own corpus. Our proposed methods were tested on the newly constructed En-My-Th medical corpus and the existing ASEAN-MT corpus. The results of our experiment indicate that contributions from linguistics, such as building a good-quality corpus and using data augmentation, can substantially improve the translation score for low-resource language pairs. Particularly, applying the SwitchOut data augmentation algorithm significantly increases the scores for English ⇔ Thai translation, which is over 30 BLEU. For English-Myanmar with syllable segmentation in the Myanmar side achie

With the rapid growth of Internet users around the world, languages have become crucial to our ability to access information, education, and communication on the Internet. In the case of people such as us who are not native English speakers, the language barrier makes it difficult for us to access the information. There is no doubt that Thai and Myanmar languages are included in the low-resource language group. The majority of natural language processing (NLP) research has been conducted on high-resource languages, whereas research on low-resource languages has been scarce. Based on this research gap, the authors decided to concentrate on the development of NLP research for our low-resource languages. Besides the low-resource data problem, finding an appropriate MT method due to the difference in nature of the languages is a challenging task for our thesis. When considering theappropriate MT approach for Thai and Myanmar languages, reverse grammatical structures as well as syllable-based written forms are key factors to consider. To solve these research problems, our thesis proposes three methods for improving machine translation (MT) performance for low-resource English-Thai-Myanmar language pairs, including 1) EDITOR and Levenshtein transformer (LevT) models, 2) SwitchOut data augmentation algorithm, and 3) mBART-50 pre-trained model. As part of our contribution to low-resource MT, we built our own corpus. Our proposed methods were tested on the newly constructed En-My-Th medical corpus and the existing ASEAN-MT corpus. The results of our experiment indicate that contributions from linguistics, such as building a good-quality corpus and using data augmentation, can substantially improve the translation score for low-resource language pairs. Particularly, applying the SwitchOut data augmentation algorithm significantly increases the scores for English ⇔ Thai translation, which is over 30 BLEU. For English-Myanmar with syllable segmentation in the Myanmar side achie

Named-entity recognition (NER) is one of the primary components in various natural language processing tasks such as relation extraction, information retrieval, question answering, etc. The majority of the research work deals with flat entities. However, it was observed that the entities were often embedded within other entities. Most of the current state-of-the-art models deal with the problem of embedded/nested entity recognition with very complex neural network architectures. In this research work, we proposed to solve the problem of nested named-entity recognition using the transfer-learning approach. For this purpose, different variants of fine-tuned, pretrained, BERT-based language models were used for the problem using the joint-labeling modeling technique. Two nested named-entity-recognition datasets, i.e., GENIA and GermEval 2014, were used for the experiment, with four and two levels of annotation, respectively. Also, the experiments were performed on the JNLPBA dataset, which has flat annotation. The performance of the above models was measured using F1-score metrics, commonly used as the standard metrics to evaluate the performance of named-entity-recognition models. In addition, the performance of the proposed approach was compared with the conditional random field and the Bi-LSTM-CRF model. It was found that the fine-tuned, pretrained, BERT-based models outperformed the other models significantly without requiring any external resources or feature extraction. The results of the proposed models were compared with various other existing approaches. The best-performing BERT-based model achieved F1-scores of 74.38, 85.29, and 80.68 for the GENIA, GermEval 2014, and JNLPBA datasets, respectively. It was found that the transfer learning (i.e., pretrained BERT models after fine-tuning) based approach for the nested named-entity-recognition task could perform well and is a more generalized approach in comparison to many of the existing approaches.

The main purpose of our project is to implement the device in Rural areas which enables the village people to use the schemes offered by the competent authorities. Here, we are training the device in the local language (Kannada) using AI and ML, so that the rural people who are more comfortable with their local language (Kannada) will be able to get the information about the government schemes. This device is trained to tell the usage of particular tablets. It helps the farmer in the detection of crop disease (leaf) and gives suitable remedies for the particular disease.

With the rapid growth of Internet users around the world, languages have become crucial to our ability to access information, education, and communication on the Internet. In the case of people such as us who are not native English speakers, the language barrier makes it difficult for us to access the information. There is no doubt that Thai and Myanmar languages are included in the low-resource language group. The majority of natural language processing (NLP) research has been conducted on high-resource languages, whereas research on low-resource languages has been scarce. Based on this research gap, the authors decided to concentrate on the development of NLP research for our low-resource languages. Besides the low-resource data problem, finding an appropriate MT method due to the difference in nature of the languages is a challenging task for our thesis. When considering theappropriate MT approach for Thai and Myanmar languages, reverse grammatical structures as well as syllable-based written forms are key factors to consider. To solve these research problems, our thesis proposes three methods for improving machine translation (MT) performance for low-resource English-Thai-Myanmar language pairs, including 1) EDITOR and Levenshtein transformer (LevT) models, 2) SwitchOut data augmentation algorithm, and 3) mBART-50 pre-trained model. As part of our contribution to low-resource MT, we built our own corpus. Our proposed methods were tested on the newly constructed En-My-Th medical corpus and the existing ASEAN-MT corpus. The results of our experiment indicate that contributions from linguistics, such as building a good-quality corpus and using data augmentation, can substantially improve the translation score for low-resource language pairs. Particularly, applying the SwitchOut data augmentation algorithm significantly increases the scores for English ⇔ Thai translation, which is over 30 BLEU. For English-Myanmar with syllable segmentation in the Myanmar side achie

Emojis are like small icons or images used to express our sentiments or feelings via text messages. They are extensively used in different social media platforms like Facebook, Twitter, Instagram etc. We considered hand-drawn emojis to classify them into 8 classes in this research paper. Hand-drawn emojis are the emojis drawn in any digital platform or in just a paper with a pen. This paper will enable the users to classify the hand-drawn emojis so that they could use them in any social media without any confusion. We made a local dataset of 500 images for each class summing a total of 4000 images of hand-drawn emojis. We presented a system which could recognise and classify the emojis into 8 classes with a convolutional neural network model. The model could favorably recognise as well as classify the hand-drawn emojis with an accuracy of 97%. Some pre-trained CNN models like VGG16, VGG19, ResNet50, MobileNetV2, InceptionV3 and Xception are also trained on the dataset to compare the accuracy and check whether they are better than the proposed one. On the other hand, machine learning models like SVM, Random Forest, Adaboost, Decision Tree and XGboost are also implemented on the dataset., ISBN för värdpublikation: 978-1-6654-1917-8, 978-0-7381-3144-3

Energy efficiency of district heating systems is of great interest to energy stakeholders. However, it is not uncommon that district heating systems fail to achieve the expected performance due to inappropriate operations. Night setback is one control strategy, which has been proved to be not a suitable setting for well-insulated modern buildings in terms of both economic and energy efficiency. Therefore, identification of a night setback control is vital to district heating companies to smoothly manage their heat energy distribution to their customers. This study is motivated to automate this identification process. The method used in this thesis is a Convolutional Neural Network(CNN) approach using the concept of transfer learning. 133 substations in Oslo are used in this case study to design a machine learning model that can identify a substation as night setback or non-night setback series. The results show that the proposed method can classify the substations with approximately 97% accuracy and 91% F1-score. This shows that the proposed method has a high potential to be deployed and used in practice to identify a night setback control in district heating substations.

Energy efficiency of district heating systems is of great interest to energy stakeholders. However, it is not uncommon that district heating systems fail to achieve the expected performance due to inappropriate operations. Night setback is one control strategy, which has been proved to be not a suitable setting for well-insulated modern buildings in terms of both economic and energy efficiency. Therefore, identification of a night setback control is vital to district heating companies to smoothly manage their heat energy distribution to their customers. This study is motivated to automate this identification process. The method used in this thesis is a Convolutional Neural Network(CNN) approach using the concept of transfer learning. 133 substations in Oslo are used in this case study to design a machine learning model that can identify a substation as night setback or non-night setback series. The results show that the proposed method can classify the substations with approximately 97% accuracy and 91% F1-score. This shows that the proposed method has a high potential to be deployed and used in practice to identify a night setback control in district heating substations.

Emojis are like small icons or images used to express our sentiments or feelings via text messages. They are extensively used in different social media platforms like Facebook, Twitter, Instagram etc. We considered hand-drawn emojis to classify them into 8 classes in this research paper. Hand-drawn emojis are the emojis drawn in any digital platform or in just a paper with a pen. This paper will enable the users to classify the hand-drawn emojis so that they could use them in any social media without any confusion. We made a local dataset of 500 images for each class summing a total of 4000 images of hand-drawn emojis. We presented a system which could recognise and classify the emojis into 8 classes with a convolutional neural network model. The model could favorably recognise as well as classify the hand-drawn emojis with an accuracy of 97%. Some pre-trained CNN models like VGG16, VGG19, ResNet50, MobileNetV2, InceptionV3 and Xception are also trained on the dataset to compare the accuracy and check whether they are better than the proposed one. On the other hand, machine learning models like SVM, Random Forest, Adaboost, Decision Tree and XGboost are also implemented on the dataset., ISBN för värdpublikation: 978-1-6654-1917-8, 978-0-7381-3144-3

Diplomová práca sa zaoberá problematikou efektívneho spracovávania logových záznamov a ich následnou analýzou pomocou korelačných pravidiel. Cieľom práce bolo implementovať spracovávanie logových záznamov do štruktúrovanej podoby, extrahovať jednotlivé polia záznamu pomocou modelu pre spracovanie prirodzeného jazyka riešením úlohy zodpovedania otázok, a vyvinúť korelačné pravidlá pre detekciu škodlivého správania. Počas riešenia zadania boli vyhotovené dve dátové sády, jedna so záznamami zo zariadení Windows, druhá obsahuje záznamy z firewallu Fortigate. Vytvorené modely na báze predtrénovaných modelov s architektúrou BERT a XLNet, ktoré boli doučené na riešenie problému parsovania logov pomocou vyhotovených datasetov a ich výsledky boli analyzované a porovnané. Druhá čásť diplomovej práce bola venovaná vývoju korelačných pravidiel, kde bol skúmaný koncept obecného zápisu Sigma. Bolo vytvorených a úspešne otestovaných šesť pravidiel, ktoré boli nasadené vo vlastnom experimentálnom pracovisku v systéme Elastic Stack, pričom každé pravidlo je popísané taktikami, technikami a subtechnikami frameworku MITRE ATT&CK., The diploma thesis deals with the problem of efficient processing of log records and their subsequent analysis using correlation rules. The goal of the thesis was to implement log processing in a structured form, extract individual log fields using a natural language processing model by solving a question answering problem, and develop correlation rules for detecting malicious behavior. Two datasets were produced during the task solution, one with records from Windows devices, and the other containing records from the Fortigate firewall. Pre-trained models based on the BERT and XLNet architecture were created and trained to solve the log parsing problem using the produced datasets, and the results were analyzed and compared. The second part of the thesis was devoted to the development of correlation rules, where the concept of a generic Sigma notation was investigated. It was developed, successfully tested and deployed six correlation rules into own experimental environment in Elastic Stack system. Each rule is also described by tactics, techniques and sub-techniques of the MITRE ATT&CK framework.

Diplomová práca sa zaoberá problematikou efektívneho spracovávania logových záznamov a ich následnou analýzou pomocou korelačných pravidiel. Cieľom práce bolo implementovať spracovávanie logových záznamov do štruktúrovanej podoby, extrahovať jednotlivé polia záznamu pomocou modelu pre spracovanie prirodzeného jazyka riešením úlohy zodpovedania otázok, a vyvinúť korelačné pravidlá pre detekciu škodlivého správania. Počas riešenia zadania boli vyhotovené dve dátové sády, jedna so záznamami zo zariadení Windows, druhá obsahuje záznamy z firewallu Fortigate. Vytvorené modely na báze predtrénovaných modelov s architektúrou BERT a XLNet, ktoré boli doučené na riešenie problému parsovania logov pomocou vyhotovených datasetov a ich výsledky boli analyzované a porovnané. Druhá čásť diplomovej práce bola venovaná vývoju korelačných pravidiel, kde bol skúmaný koncept obecného zápisu Sigma. Bolo vytvorených a úspešne otestovaných šesť pravidiel, ktoré boli nasadené vo vlastnom experimentálnom pracovisku v systéme Elastic Stack, pričom každé pravidlo je popísané taktikami, technikami a subtechnikami frameworku MITRE ATT&CK., The diploma thesis deals with the problem of efficient processing of log records and their subsequent analysis using correlation rules. The goal of the thesis was to implement log processing in a structured form, extract individual log fields using a natural language processing model by solving a question answering problem, and develop correlation rules for detecting malicious behavior. Two datasets were produced during the task solution, one with records from Windows devices, and the other containing records from the Fortigate firewall. Pre-trained models based on the BERT and XLNet architecture were created and trained to solve the log parsing problem using the produced datasets, and the results were analyzed and compared. The second part of the thesis was devoted to the development of correlation rules, where the concept of a generic Sigma notation was investigated. It was developed, successfully tested and deployed six correlation rules into own experimental environment in Elastic Stack system. Each rule is also described by tactics, techniques and sub-techniques of the MITRE ATT&CK framework.

Diplomová práca sa zaoberá problematikou efektívneho spracovávania logových záznamov a ich následnou analýzou pomocou korelačných pravidiel. Cieľom práce bolo implementovať spracovávanie logových záznamov do štruktúrovanej podoby, extrahovať jednotlivé polia záznamu pomocou modelu pre spracovanie prirodzeného jazyka riešením úlohy zodpovedania otázok, a vyvinúť korelačné pravidlá pre detekciu škodlivého správania. Počas riešenia zadania boli vyhotovené dve dátové sády, jedna so záznamami zo zariadení Windows, druhá obsahuje záznamy z firewallu Fortigate. Vytvorené modely na báze predtrénovaných modelov s architektúrou BERT a XLNet, ktoré boli doučené na riešenie problému parsovania logov pomocou vyhotovených datasetov a ich výsledky boli analyzované a porovnané. Druhá čásť diplomovej práce bola venovaná vývoju korelačných pravidiel, kde bol skúmaný koncept obecného zápisu Sigma. Bolo vytvorených a úspešne otestovaných šesť pravidiel, ktoré boli nasadené vo vlastnom experimentálnom pracovisku v systéme Elastic Stack, pričom každé pravidlo je popísané taktikami, technikami a subtechnikami frameworku MITRE ATT&CK., The diploma thesis deals with the problem of efficient processing of log records and their subsequent analysis using correlation rules. The goal of the thesis was to implement log processing in a structured form, extract individual log fields using a natural language processing model by solving a question answering problem, and develop correlation rules for detecting malicious behavior. Two datasets were produced during the task solution, one with records from Windows devices, and the other containing records from the Fortigate firewall. Pre-trained models based on the BERT and XLNet architecture were created and trained to solve the log parsing problem using the produced datasets, and the results were analyzed and compared. The second part of the thesis was devoted to the development of correlation rules, where the concept of a generic Sigma notation was investigated. It was developed, successfully tested and deployed six correlation rules into own experimental environment in Elastic Stack system. Each rule is also described by tactics, techniques and sub-techniques of the MITRE ATT&CK framework.

Diplomová práca sa zaoberá problematikou efektívneho spracovávania logových záznamov a ich následnou analýzou pomocou korelačných pravidiel. Cieľom práce bolo implementovať spracovávanie logových záznamov do štruktúrovanej podoby, extrahovať jednotlivé polia záznamu pomocou modelu pre spracovanie prirodzeného jazyka riešením úlohy zodpovedania otázok, a vyvinúť korelačné pravidlá pre detekciu škodlivého správania. Počas riešenia zadania boli vyhotovené dve dátové sády, jedna so záznamami zo zariadení Windows, druhá obsahuje záznamy z firewallu Fortigate. Vytvorené modely na báze predtrénovaných modelov s architektúrou BERT a XLNet, ktoré boli doučené na riešenie problému parsovania logov pomocou vyhotovených datasetov a ich výsledky boli analyzované a porovnané. Druhá čásť diplomovej práce bola venovaná vývoju korelačných pravidiel, kde bol skúmaný koncept obecného zápisu Sigma. Bolo vytvorených a úspešne otestovaných šesť pravidiel, ktoré boli nasadené vo vlastnom experimentálnom pracovisku v systéme Elastic Stack, pričom každé pravidlo je popísané taktikami, technikami a subtechnikami frameworku MITRE ATT&CK., The diploma thesis deals with the problem of efficient processing of log records and their subsequent analysis using correlation rules. The goal of the thesis was to implement log processing in a structured form, extract individual log fields using a natural language processing model by solving a question answering problem, and develop correlation rules for detecting malicious behavior. Two datasets were produced during the task solution, one with records from Windows devices, and the other containing records from the Fortigate firewall. Pre-trained models based on the BERT and XLNet architecture were created and trained to solve the log parsing problem using the produced datasets, and the results were analyzed and compared. The second part of the thesis was devoted to the development of correlation rules, where the concept of a generic Sigma notation was investigated. It was developed, successfully tested and deployed six correlation rules into own experimental environment in Elastic Stack system. Each rule is also described by tactics, techniques and sub-techniques of the MITRE ATT&CK framework.

Diplomová práca sa zaoberá problematikou efektívneho spracovávania logových záznamov a ich následnou analýzou pomocou korelačných pravidiel. Cieľom práce bolo implementovať spracovávanie logových záznamov do štruktúrovanej podoby, extrahovať jednotlivé polia záznamu pomocou modelu pre spracovanie prirodzeného jazyka riešením úlohy zodpovedania otázok, a vyvinúť korelačné pravidlá pre detekciu škodlivého správania. Počas riešenia zadania boli vyhotovené dve dátové sády, jedna so záznamami zo zariadení Windows, druhá obsahuje záznamy z firewallu Fortigate. Vytvorené modely na báze predtrénovaných modelov s architektúrou BERT a XLNet, ktoré boli doučené na riešenie problému parsovania logov pomocou vyhotovených datasetov a ich výsledky boli analyzované a porovnané. Druhá čásť diplomovej práce bola venovaná vývoju korelačných pravidiel, kde bol skúmaný koncept obecného zápisu Sigma. Bolo vytvorených a úspešne otestovaných šesť pravidiel, ktoré boli nasadené vo vlastnom experimentálnom pracovisku v systéme Elastic Stack, pričom každé pravidlo je popísané taktikami, technikami a subtechnikami frameworku MITRE ATT&CK., The diploma thesis deals with the problem of efficient processing of log records and their subsequent analysis using correlation rules. The goal of the thesis was to implement log processing in a structured form, extract individual log fields using a natural language processing model by solving a question answering problem, and develop correlation rules for detecting malicious behavior. Two datasets were produced during the task solution, one with records from Windows devices, and the other containing records from the Fortigate firewall. Pre-trained models based on the BERT and XLNet architecture were created and trained to solve the log parsing problem using the produced datasets, and the results were analyzed and compared. The second part of the thesis was devoted to the development of correlation rules, where the concept of a generic Sigma notation was investigated. It was developed, successfully tested and deployed six correlation rules into own experimental environment in Elastic Stack system. Each rule is also described by tactics, techniques and sub-techniques of the MITRE ATT&CK framework.

Diplomová práca sa zaoberá problematikou efektívneho spracovávania logových záznamov a ich následnou analýzou pomocou korelačných pravidiel. Cieľom práce bolo implementovať spracovávanie logových záznamov do štruktúrovanej podoby, extrahovať jednotlivé polia záznamu pomocou modelu pre spracovanie prirodzeného jazyka riešením úlohy zodpovedania otázok, a vyvinúť korelačné pravidlá pre detekciu škodlivého správania. Počas riešenia zadania boli vyhotovené dve dátové sády, jedna so záznamami zo zariadení Windows, druhá obsahuje záznamy z firewallu Fortigate. Vytvorené modely na báze predtrénovaných modelov s architektúrou BERT a XLNet, ktoré boli doučené na riešenie problému parsovania logov pomocou vyhotovených datasetov a ich výsledky boli analyzované a porovnané. Druhá čásť diplomovej práce bola venovaná vývoju korelačných pravidiel, kde bol skúmaný koncept obecného zápisu Sigma. Bolo vytvorených a úspešne otestovaných šesť pravidiel, ktoré boli nasadené vo vlastnom experimentálnom pracovisku v systéme Elastic Stack, pričom každé pravidlo je popísané taktikami, technikami a subtechnikami frameworku MITRE ATT&CK., The diploma thesis deals with the problem of efficient processing of log records and their subsequent analysis using correlation rules. The goal of the thesis was to implement log processing in a structured form, extract individual log fields using a natural language processing model by solving a question answering problem, and develop correlation rules for detecting malicious behavior. Two datasets were produced during the task solution, one with records from Windows devices, and the other containing records from the Fortigate firewall. Pre-trained models based on the BERT and XLNet architecture were created and trained to solve the log parsing problem using the produced datasets, and the results were analyzed and compared. The second part of the thesis was devoted to the development of correlation rules, where the concept of a generic Sigma notation was investigated. It was developed, successfully tested and deployed six correlation rules into own experimental environment in Elastic Stack system. Each rule is also described by tactics, techniques and sub-techniques of the MITRE ATT&CK framework.

Accepted conference paper for IEEE GCAIoT. Citation information and links to final, published version coming soon.