Enhancing Keystroke Biometric Authentication Using Deep Learning Techniques

Medvedev, Viktor; Budžys, Arnoldas; Kurasova, Olga

doi:10.23919/cisti58278.2023.10211344

Cited by 2 publications

(3 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These methods encode the time series data into polar coordinates, effectively capturing the temporal relationships within the time series data. The Markov transition field technique represents sequential data, such as time series, as images by transforming data points into a matrix where each element indicates transition probabilities, thus capturing sequence dynamics [29], while recurrence plots convert time-series data into images to display similarities between data points [27]. Each of these methods has its own unique advantages when encoding time series data for deep neural network training.…”

Section: Data Transformationmentioning

confidence: 99%

“…This general architecture allows the Siamese neural network to acquire informative representations of the input data, enabling it to make accurate comparisons and successfully perform decision-making tasks. More details on the architecture, hyperparameters, and their tuning can be found in our previous research [27,32,33].…”

Section: Deep Neural Network: Siamese Neural Network With Triplet Lossmentioning

confidence: 99%

“…, such as the Gramian Angular Summation Field (GASF), the Gramian Angular Difference Field (GADF), and recurrence plots[26,27]. • Deep Neural Network Training.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Exploring Multidimensional Embeddings for Decision Support Using Advanced Visualization Techniques

Kurasova,

Budžys,

Medvedev

2024

Informatics

Self Cite

View full text Add to dashboard Cite

As artificial intelligence has evolved, deep learning models have become important in extracting and interpreting complex patterns from raw multidimensional data. These models produce multidimensional embeddings that, while containing a lot of information, are often not directly understandable. Dimensionality reduction techniques play an important role in transforming multidimensional data into interpretable formats for decision support systems. To address this problem, the paper presents an analysis of dimensionality reduction and visualization techniques that embrace complex data representations and are useful inferences for decision systems. A novel framework is proposed, utilizing a Siamese neural network with a triplet loss function to analyze multidimensional data encoded into images, thus transforming these data into multidimensional embeddings. This approach uses dimensionality reduction techniques to transform these embeddings into a lower-dimensional space. This transformation not only improves interpretability but also maintains the integrity of the complex data structures. The efficacy of this approach is demonstrated using a keystroke dynamics dataset. The results support the integration of these visualization techniques into decision support systems. The visualization process not only simplifies the complexity of the data, but also reveals deep patterns and relationships hidden in the embeddings. Thus, a comprehensive framework for visualizing and interpreting complex keystroke dynamics is described, making a significant contribution to the field of user authentication.

show abstract

Section: Data Transformationmentioning

confidence: 99%

Section: Deep Neural Network: Siamese Neural Network With Triplet Lossmentioning

confidence: 99%