Authentication, an identity verification and confirmation method, is a defense mechanism that reduces the risk of adversarial attacks, specifically to identity theft and impersonation in computer systems. Existing authentication methods exhibit vulnerabilities, such as password dictionary attack, credential stuffing, and identity spoofing. In this study, we examine the possibility of using a class of biometric data, namely step counts, to investigate their potential in person identification and verification. For this purpose, we collected step-count data from research volunteers over a period of 33 days or over 560 hours. Subsequently, we used these data to establish an appropriate threshold and tested their accuracy using a confusion matrix. Our evaluations showed that a suitable threshold range for step-count authentication is x ̅-1S.D.≤Range ≤ x ̅+1S.D., where S.D. represents standard deviation and x ̅ is the mean value of step counts of an individual. Moreover, we constructed a receiver operating characteristic curve and calculated the area under the curve, which showed that step counts have the potential to be used in behavioral biometric authentication methods. Thus, using the threshold range method, step counts can potentially become another behavioral biometric factor for authentication systems.

Time series forecasting in big data analytics is crucial for making decisions in a variety of fields. but faces challenges due to high dimensionality, non-stationarity, and dynamic patterns. Conventional approaches frequently produce inaccurate results because they are unable to capture sudden variations and intricate temporal connections. This study proposes a Multi-scale Dynamic Time Warping-based Hierarchical Clustering (MDTWbH) approach to improve forecasting accuracy and scalability. Multi-scale Dynamic Time Warping (MDTW) transforms time series data into multi-scale representations, preserving local and global patterns, while Hierarchical Clustering groups similar sequences for enhanced predictive performance. The proposed framework integrates data preprocessing, outlier detection, and missing value interpolation to refine input data. It employs Apache Hadoop and Spark for efficient big data processing. Long Short Term Memory (LSTM) is applied within each cluster for accurate forecasting, and accuracy, precision, recall, F1-score, MAE, and RMSE are used to assess the performance of the model. Experimental results on electricity demand, wind speed, and taxi demand datasets demonstrate superior performance compared to existing techniques. MDTWbH provides a scalable and interpretable solution for large-scale time series forecasting by efficiently capturing evolving temporal patterns.

The development of technology around us is going through a rapid and significant state that is almost causing a technological revolution, so one of the most important problems facing us in the current technological era is the management of warehouse data and the growth occurring in the volume of data that is dealt with on a daily basis, whether in terms of its storage or security, especially if the data is huge and large. Therefore, we developed a proposed model in our study that provides security in addition to storage/warehouse management. In our proposed model, the El-Gamal and GLUON functions address the security problem. In addition to supporting other security methods, such as GLUON, which is secure and fast, for encryption. Hybrid Blockchain technology is used in our proposed model to deal with the storage of this type of huge data, and also for the purpose of organizing warehouse storage. Data is exposed to intrusion or loss when using any traditional, centralized technology or when storing it in databases, so we chose the hybrid Blockchain to be an integrated fit with our proposed model, and also because it allows the distribution of data across public and private domains. Our proposed model, upon examination, shows that it effectively dealt with defending against attacks such as NotPetya, GoldenEye, WannaCry, Emotet, Trickbot, Conti, and DarkSide. In addition, the results of lightweight GLUON and El-Gamal showed that the performance analysis of our model was very successful, where the time it takes to create a block was between 0.01 ns and not more than 0.09 ns which is considered too fast for such a system that deals with a big data. As a result, we were able to gain an effective model for data repository control, security, performance, and management. 

This paper presents a novel cryptanalysis method of DES (2-DES and 3-DES) using nature-inspired algorithms; namely Cuckoo Search Algorithm and Grey Wolf Optimization Algorithm. We have shown the loophole of 2-DES and 3-DES encryption systems and discovered the vulnerabilities by some simple mathematical calculations. The Meet-In-The-Middle approach can be executed on 2-DES along with Known Plaintext Attack, Chosen Plaintext Attack, and Chosen Ciphertext Attack. The valid key pairs along with the original key pairs can successfully be recovered by this attack algorithm. But in the Ciphertext Only Attack, the Meet-In-The-Middle approach fails to recover the plaintext as well as the valid key pairs both for 2-DES and 3-DES. To overcome this problem, we have proposed a novel cryptanalysis method of 3-DES with Ciphertext Only Attack using Cuckoo Search Algorithm and Grey Wolf Optimization Algorithm (GWO). We have developed a suitable fitness function, accelerating the algorithm toward the optimal solution. This paper shows how CSA and GWO can break a 3-DES cryptosystem using a Ciphertext Only Attack. This proposed cryptanalysis method can also be applied to any round of DES.

Binary code similarity detection (BCSD) is a method for identifying similarities between two or more slices of binary code (machine code or assembly code) without access to their original source code. BCSD is often used in many areas, such as vulnerability detection, plagiarism detection, malware analysis, copyright infringement and software patching. Numerous approaches have been developed in these areas via graph matching and deep learning algorithms. Existing solutions have low detection accuracy and lack cross-architecture analysis. This work introduces a cross-platform graph deep learning-based approach, i.e., GraphConvDeep, which uses graph convolution networks to compute the embedding. The proposed GraphConvDeep approach relies on the control flow graph (CFG) of individual binary functions. By evaluating the distance between two embeddings of functions, the similarity is detected. The experimental results show that GraphConvDeep is better than other cutting-edge methods at accurately detecting similarities, achieving an average accuracy of 95% across different platforms. The analysis shows that the proposed approach achieves better performance with an area under the curve (AUC) value of 96%, particularly in identifying real-world vulnerabilities.

The paper offers a novel digital signature scheme that integrates ElGamal cryptographic principles with the Short Integer Solution (SIS) problem, specifically designed to ensure post-quantum security. As quantum computers advance and present significant risks to traditional cryptographic systems, this scheme offers an interesting alternative for securing digital signatures against potential quantum threats. The scheme uses only basic secure principles. The offered approach offers key generation, where parameters and random matrices are selected, and signature generation, which involves creating signatures based on hashed messages and matrix computations. Verification ensures the authenticity and integrity of signatures. We provide experimental evaluations detailing key generation, signature creation, and verification times across different matrix dimensions and message sizes. Key generation takes between 2.5–10.2 seconds, while signature generation ranges from 0.20 to 9.30 milliseconds and verification from 0.18 to 8.90 milliseconds, depending on message size and matrix dimension. The scheme maintains a consistent signature size of 1.7 KB, independent of message length due to a hash-and-sign strategy. These results demonstrate that the scheme balances post-quantum security with practical performance, especially in high-security contexts. A comparison with traditional ElGamal encryption reveals the trade-offs between security and efficiency. While the SIS-based scheme delivers enhanced protection against quantum threats, it also entails increased computational complexity and larger signature sizes compared to conventional schemes.
Overall, our proposed digital signature scheme stands as an excellent option for safe communications in a post-quantum world, representing a crucial step in protecting the authenticity and integrity of digital exchanges against changing technological risks. We believe that as quantum computing continues to develop, research into robust cryptographic alternatives will become increasingly important for safeguarding sensitive information across various sectors.

Voice User Interfaces (VUIs) focus on their application in IT and linguistics. Our research examines the capabilities and limitations of small and multilingual BERT models in the context of speech recognition and command conversion. We evaluate the performance of these models through a series of experiments, including the application of confusion matrices to assess their effectiveness. The findings reveal that larger models like multilingual BERT theoretically offer advanced capabilities but often demand more substantial resources and well-balanced datasets. Conversely, smaller models, though less resource-intensive, may sometimes provide more practical solutions. Our study underscores the importance of dataset quality, model fine-tuning, and efficient resource management in optimising VUIS. Insights gained from this research highlight the potential of neural networks to enhance and improve user interaction. Despite challenges in achieving a fully functional interface, the study provides valuable contributions to the VUIs development and sets the stage for future advancements in integrating AI with linguistic technologies. The article describes the development of a voice user interface (VUI) capable of recognising, analysing, and interpreting the Ukrainian language. For this purpose, several neural network architectures were used, including the Squeezeformer-CTC model, as well as a modified w2v-bert-2.0-uk model, which was used to decode speech commands into text. The multilingual BERT model (mBERT) for the classification of intentions was also tested. The developed system showed the prospects of using BERT models in combination with lightweight ASR architectures to create an effective voice interface in Ukrainian. Accuracy indicators (F1 = 91.5%, WER = 12.7%) indicate high-quality recognition, which is provided even in models with low memory capacity. The system is adaptable to conditions with limited resources, particularly for educational and living environments with a Ukrainian-speaking audience.

In today’s internet age, cyberbullying is a serious threat to people. Bullies hurt their victims by using online social media sites like Facebook, Twitter, and so on. Since then, cyberbullying or online bullying has become an increasingly significant societal problem due to its psychological and emotional impact on the victim. To overcome this issue, the proposed method is used for the identification of social network cyberbullying using many modalities. The first step is data collection, the data are collected through text, image, audio and video. Firstly, the text data is pre-processed using tokenization, stemming, lemmatization, spell correction, and PoS tagging methods. After pre-processing, feature extraction is done using Log Term Frequency-based Modified Inverse Class Frequency (LTF-MICF) and glove modelling, and the feature is extracted using the convolutional dense capsule network (Conv_DCapNet) model. Second, image data is pre-processed using Gabor filtering and image resizing. Once the pre-processing is done, the feature is extracted using image modality generation through Self-attention based bidirectional gated recurrent auto-encoder network (SA_BiGR_AENet). Third, audio (acoustic) data is extracted using the Librosa library and Attentional convolutional BiLSTM. The next step is video extraction, in this video data are extracted using keyframe extraction and Residual 101 network. Finally, in the fusion method, all four extracted features (text, image, audio and video) are fused by the deep tensor fusion framework. After that, the sigmoid layer classifies whether the data is cyberbullying or not. Then, the notification of highly detected severe is sent to the user. Thus, the proposed method detects cyberbullying. Python tool is used for implementation, and the performance metrics of recall is 96%, F1-score for bullying class 92% and weighted F1-score 91% compared with existing methods.