Rozdziały

Stały URI dla kolekcjihttp://hdl.handle.net/11652/4775

Przeglądaj

collection.search.results.head

Teraz wyświetlane 1 - 10 z 13
  • Pozycja
    Lung Xray Images Analysis for COVID-19 Diagnosis
    (Wydawnictwo Politechniki Łódzkiej, 2023) Kloska, Anna; Tarczewska, Martyna; Giełczyk, Agata; Marciniak, Beata
    Background: The SARS-CoV-2 pandemic began in early 2020. It paralyzed human life all over the world and threatened our security. Thus, proposing some novel and effective approaches to diagnosing COVID-19 infections became paramount. Methods: This article proposes a method for the classification of chest X-ray images based on the transfer learning. We examined also different scenarios of dataset augmentation. Results: The paper reports accuracy=98%, precision=97%, recall=100% and F1-score=98% in the most promising approach. Conclusion: Our research proofs that machine learning can be used in order to support medics in chest X-ray classification and implementing augmentation can lead to improvements in accuracy, precision, recall, and F1-scores.
  • Pozycja
    AMUseBot: Towards Making the Most out of a Task-oriented Dialogue System
    (Wydawnictwo Politechniki Łódzkiej, 2023) Christop, Iwona; Dudzic, Kacper; Krzymiński, Mikołaj
    This paper presents AMUseBot, a task-oriented dialogue system designed to assist the user in completing multi-step tasks. Taking into consideration that the fundamental issues with such systems are poor user ratings and high rates of uncompleted tasks, the main goal of the project is to keep the user focused and provide engaging conversations. We approach these problems by the introduction of dynamic multimodal communication and graph-based task management.
  • Pozycja
    AloneKnight – Enabling Affective Interaction within Mobile Video Games
    (Wydawnictwo Politechniki Łódzkiej, 2023) Jemioło, Paweł; Świder, Krzysztof; Storman, Dawid; Adrian, Weronika T.
    Artificial intelligence is used in various contexts, including video games, where it can enhance the game design and adapt content to players’ emotional states through affective computing. In this paper, we present an example of an affective mobile game and compare participants’ opinions after playing two versions of the game, with and without an affective loop. The game was developed using Unity. In the affective version, physiological data is recorded and analysed to detect emotions based on facial expressions and electrodermal activity, which then affects the game. The study with 11 participants showed positive feedback for the game with affective loop.
  • Pozycja
    Predictive User Interface for Emerging Experiences
    (Wydawnictwo Politechniki Łódzkiej, 2023) Kapusta, Paweł; Duch, Piotr
    This research paper focuses on the use of predictive techniques to improve interaction with user interfaces in emerging experiences such as Virtual Reality, Augmented Reality, Metaverse, and touchless kiosks and dashboards. We propose the concept of intelligent snapping, which uses gaze tracking, head-pose tracking, hand tracking, as well as gesture recognition and hand posture recognition to catch the intent of the person rather than the actual input.
  • Pozycja
    Are Quantified Boolean Formulas Hard for Reason-Able Embeddings?
    (Wydawnictwo Politechniki Łódzkiej, 2023) Potoniec, Jędrzej
    We aim to establish theoretical boundaries for the applicability of reason-able embeddings, a recently proposed method employing a transferable neural reasoner to shape a latent space of knowledge graph embeddings. Since reason-able embeddings rely on the ALC description logic, we construct a dataset of the hardest concepts in ALC by translating quantified boolean formulas (QBF) from QBFLIB, a benchmark for QBF solvers. We experimentally show the dataset is hard for a symbolic reasoner FaCT++, and analyze the results of reasoning with reason-able embeddings, concluding that the dataset is too hard for them.
  • Pozycja
    Increasing Skin Lesions Classification Rates using Convolutional Neural Networks with Invariant Dataset Augmentation and the Three-Point Checklist of Dermoscopy
    (Wydawnictwo Politechniki Łódzkiej, 2023) Milczarski, Piotr; Borowski, Norbert; Beczkowski, Michał
    In the paper, we show how to tackle the problem of lack of the rotation invariance in CNN networks using the authors’ Invariant Dataset Augmentation (IDA) method. The IDA method allows to increase the classification rates taking into account as an example the classification of the skin lesions using a small image set. In order to solve the problem of the lack of rotation invariance, IDA method was used and the dataset was increased in an eightfold and invariant way. In the research, we applied the IDA methods and compared the results of VGG19, XN and Inception-ResNetv2 CNN networks in three skin lesions features classification defined by wellknown dermoscopic criterions e.g. the Three-Point Checklist of Dermoscopy or the Seven-Point Checklist. Due to Invariant Dataset Augmentation, the classification rate parameters like true positive rate by almost 20%, false positive rate as well as the F1 score and Matthews correlation coefficient have been significantly increased opposite to type II error that has significantly decreased. In the paper, the confusion matrix parameters result in: 98-100% accuracy, 98-100% true positive rate, 0.0-2.3% false positive rate, tests F1=0.95 and MCC=0.95. That general approach can provide higher results while using CNN networks in other applications.
  • Pozycja
    Grounded HyperSymbolic Representations Learned through Gradient-Based Optimization
    (Wydawnictwo Politechniki Łódzkiej, 2023) Łuczak, Piotr; Ślot, Krzysztof; Kucharski, Jacek
    Hyperdimensional computing is a novel paradigm, capable of processing complex data structures with simple operations. Its main limitations lie in the conversion of real world data onto hyperdimensional space, which due to lack of a universal translation scheme, oftentimes requires application-specific methods. This work presents a novel method for unsupervised hyperdimensional conversion of arbitrary image data. Additionally, this method is augmented by the ability of creating HyperSymbols, or class prototypes, provided that such class labels are available. The proposed method achieves promising performance on MNIST dataset, both in translating individual samples as well as producing HyperSymbols for downstream classification task.
  • Pozycja
    A Convolutional and Recurrent Neural Network-based Approach for Speech Emotion Recognition
    (Wydawnictwo Politechniki Łódzkiej, 2023) Duch, Piotr; Wiatrowska, Izabela; Kapusta, Paweł
    Speech emotion recognition (SER) is a crucial aspect of humancomputer interaction. In this article, we propose a deep learning approach, using CNN and RNN architectures, for SER using both convolutional and recurrent neural networks. We evaluated the approach on four audio datasets, including CREMA-D, RAVDESS, TESS, and EMOVO. Our experiments tested various feature sets and extraction settings to determine optimal features for SER. Our results demonstrate that the proposed approach achieves high accuracy rates and outperforms state-of-the-art algorithms.
  • Pozycja
    Performance Analysis of Machine Learning Platforms Using Cloud Native Technology on Edge Devices
    (Wydawnictwo Politechniki Łódzkiej, 2023) Cłapa, Konrad; Grudzień, Krzysztof; Sierszeń, Artur
    This article presents the results of an experiment performed on a machine learning edge computing platform composed of a virtualized environment with a K3s cluster and Kubeflow software. The study aimed to analyze the effectiveness of executing Kubeflow pipelines for simulated parallel executions. A benchmarking environment was developed for the experiment to allow system performance measurements based on parameters, including the number of pipelines and nodes. The results demonstrate the impact of the number of cluster nodes on computational time, revealing insights that could inform future decisions regarding increasing the effectiveness of running machine learning pipelines on edge devices.
  • Pozycja
    Integrating Anomaly Detection for Enhanced Data Protection in Cloud-Based Applications
    (Wydawnictwo Politechniki Łódzkiej, 2023) Czerkas, Konrad; Drozd, Michał; Duraj, Agnieszka; Lichy, Krzysztof; Lipiński, Piotr; Morawski, Michał; Napieralski, Piotr; Puchała, Dariusz; Kwapisz, Marcin; Warcholiński, Adrian; Karbowańczyk, Michał; Wosiak, Piotr
    In this research, anomaly detection techniques and artificial neural networks were employed to address the issue of attacks on cluster computing systems. The study investigated the detection of Distributed Denial of Service (DDoS) and Partition attacks by monitoring metrics such as network latency, data transfer rate, and number of connections. Additionally, outlier detection algorithms, namely Local Outlier Factor (LOF) and COF, as well as ARIMA and SHESD models were tested for anomaly detection. Two types of neural network architectures, multi-layer perceptron (MLP) and recursive LSTM networks, were used to detect attacks by classifying events as “attack” or “no attack”. The study underscores the importance of implementing proactive security measures to protect cluster computing systems from cyber threats.