Optical Memory and Neural Networks最新文献

Scrutiny of consumer tweets posted on social media is found to be indispensable for numerous business applications. In this manner, the model of big data analytics is applied in processing data and analyzes it to predict consumer behavioral patterns on social media. Different machine learning algorithms have gathered consumer data to analysis consumer behavior. Conventional methods are unable to discover extreme hidden patterns and require to be enhanced to produce more accurate behavioral patterns. In this work a hybrid method called, proposed Bouldin Correlation Clustering and Gradient Extreme Learning Machine (BCC-GELM) method to perform the consumer behavior analysis in social network with big data. The BCC-GELM method in hybrid model split into two modules. At first, Davis-Bouldin Index-based Correlation Clustering selects clusters with most edges within clusters as positive (i.e., similar information) while most edges between clusters as negative (i.e., dissimilar information), therefore minimizing the error rate. Consumer previous behavioral characteristics and twitter messages are analyzed by means of focal points (i.e., cluster center) via Davis-Bouldin Index. Subsequently, Stochastic Gradient Descent Extreme Learning Machine yields good results by considering distribution of tweets, therefore paving way for predicting consumer behavioral patterns in an optimal manner. The performance of BCC-GELM method is evaluated using experimental analysis and comparison is also made with traditional consumer behavioral pattern methods. The findings demonstrate that BCC-GELM method performs well than the traditional consumer behavioral pattern methods in terms of 9% of clustering accuracy, 45 and 54% of clustering time using without and with preprocessing (percent), 23% of clustering overhead and 46% of error rate.

This work presents results from using a Monte Carlo model to describe photon interactions with a scattering and absorbing medium, exemplified by the eye cornea in polarization-sensitive optical coherence tomography (PS-OCT) approaches. The interaction of an incident photon packet with a weakly scattering birefringent object was analyzed using the meridian plane Monte Carlo approach, which made it possible to take into account the depolarization of radiation during interaction with the scattering centers of the eye corneal epithelium and to increase the signal-to-noise ratio of object information. The dynamic and geometric phase reconstruction in a modified Mach-Zehnder interferometer scheme allows to obtain data of collagen fibers orientation non-invasive, to restore lost information of the birefringent object structure. The result of this reconstruction is a complete picture of the stromal structure with an accuracy that surpasses current levels achieved with existing PS-OCT systems.

The properties of an antiferromagnetic substance are investigated in the presence of a magnetic field. Analytical expressions are obtained in terms of the mean-field approximation. An external magnetic field is shown to be non-destructive to the phase transition in the antiferromagnetic substance. It only changes critical exponents and shifts the critical point. This allows us to control the critical properties of the system. The number of critical points can vary from one (the second-order phase transition) to four (two first-order phase transitions and two second-order phase transitions). It is shown that variations in the magnetic field magnitude can raise the critical temperature by three-odd times in materials with strong antiferromagnetic interactions. A Monte Carlo simulation carried out for a three-dimensional lattice with a finite interaction radius substantiates that the action of an external field brings about a shift in the temperature of the transition. The simulation results agree well with the analytical expressions of the mean field theory.

Development of 5G internet in today’s trend leads to the evaluation of many IOT devices. The information is transmitted by a network in IOT to store the data in the cloud. Due to the wide usage of IOT devices by people, congestion may occurs in IOT networks, which delays the information or sometimes resulting in data loss despite the implementation of congestion control methods. So many machine learning and congestion control protocols are used to predict and avoid congestion in IOT network. But these existing systems consist of drawbacks such as accuracy drop for prediction, packet loss and time delay. Hence, the Bandwidth Aware Routing Strategy (BARS) protocol using Jordan Neural Network (JNN) was developed to predict and avoid congestion in the network. Initially, the IOT nodes are deployed and the data are collected and preprocessed using a sigmoidal function and Extreme Learning machine to improve the quality of the original data. Then extract the features from the pre-processed data using Locality Preserving Projection (LPP). After that, Jordan Neural Network is used for congestion prediction and pine cone optimization is used to tune the hyper parameters such as learning rate and batch size which is utilized to improve the classifier performance. Then, BARS protocol is used to avoid the congestion present in the IOT network. According to the experimental approach, the proposed techniques achieves 95.45% of Accuracy, 95.71% of Precision, 95.39% of F1-Scorce and 95.02 of specificity. Thus, the congestion and avoidance of Information in the IOT network is processed in high efficiency by using this proposed approach.

A person’s personality comprises a range of behaviours, attitudes, and emotional patterns that shift throughout time due to ecological and biological influences. Personality prediction from the MBTI dataset poses computational efficiency, memory utilisation, and class imbalance challenges. This study proposes a novel approach leveraging Knowledge Distillation-based BERT to address these challenges. The process involves three stages: pre-processing, feature extraction, and classification. Initially, data is cleaned by removing irrelevant characters and URLs, followed by tokenisation and conversion to lowercase for consistency. The padding ensures uniform input size for DistilBERT, with attention masks aiding focus on relevant tokens. DistilBERT extracts contextual embeddings, enhanced by segment and positional embeddings, capturing semantic meaning via multi-head self-attention. A fully connected layer with GELU activation and batch normalisation mitigates overfitting, followed by a classification layer with Sparsemax activation, addressing the class imbalance. Fine-tuning pre-trained DistilBERT maximises detection accuracy while excluding irrelevant learning objectives. Dynamic masking during inference replaces static masking, and the Radam optimiser optimises hyperparameters for improved convergence. Our approach offers a robust solution that achieves 93% accuracy and 95% F1-score for accurate personality prediction while mitigating computational complexities and class imbalance issues.

This work presents a technique developed by utilizing Generative Adversarial Networks (GANs) to generate Sign Language videos. Sign Language is the main mode of communication for people in the hearing impaired community. The process of teaching sign language is difficult as there are not a lot of tools available for this purpose. Generative artificial intelligence can be very helpful for this task as it is able to learn from the limited data and is able to generate various images and videos. In this work, Conditional GANs (cGANs) were employed to generate videos for Indian Sign Language (ISL) based on a text input. It is found that the results obtained from cGANs exhibit superior quality and control based on the performance metrics such as SSIM, FID and MSE values. The effectiveness of the cGANs in generating accurate and visually appealing sign language videos highlights their potential for teaching sign language and improving sign language communication systems.

Brain tumor identification using MRI images involves the detailed examination of brain tissues to detect and characterize tumors. Conventional ML and DL algorithms sometimes encounter difficulties due to a lack of labelled data, resulting in inferior performance and poor generalization. To address these issues, this study introduces an Advanced Attention-based Pre-trained Transfer Learning (TL) model that enhances accuracy and resilience in identifying and categorizing brain tumors using MRI images. The methodology starts with pre-processing, which includes image scaling and noise reduction with an adaptive median filter. After pre-processing, the images are fed into a CNN-based framework called Pre-trained Attention-fused Image SpectraNet. This framework comprises of five convolutional layers, after which Rectified Linear Unit (ReLU) activation and pooling layers are added to learn progressively more complex features. A novel self-attention layer is implemented to capture deep features that reveal aberrant tissue patterns, hence increasing model interpretability and accuracy. A globally average pooling layer is employed to reduce computational complexity, and it is accompanied by a fully connected layer with batch normalization to assure stability and convergence during training. The last layer uses softmax to categorize normal, pituitary, glioma, and meningioma. Utilizing the Adam optimizer, the suggested approach enhances performance, yielding excellent metrics such as 98.33% accuracy, 98.35% precision, 98.28% recall, and a 98.31% F1-score. These measures show considerable increases over existing ML and DL methods, demonstrating the system’s ability to improve brain tumor detection accuracy. The advancement of these treatments has significant implications for medical professionals who specialize in the timely identification of brain tumors.

Anomaly detection is the identification of aberrations in the dataset using statistical methods or machine learning algorithms. It is widely performed using unsupervised learning algorithms because labelling the data manually can be expensive. While unsupervised anomaly detection is sufficient for data cleaning, this is not the case in real-world applications, where accuracy is of the utmost importance. For example, it would be unacceptable to misdiagnose someone as not having breast cancer and not provide them with treatment because our model failed to recognize it as an anomaly. In this paper, we propose an optimized model—IFDRF (Isolation Forest, DBSCAN, and Random Forest) that has incorporated feedback (corrections) into the unsupervised detection model. IFDRF is a novel hybrid model combining an unsupervised learning model at the first layer followed by a clustering model at the second layer and a supervised learning model at the end. The proposed model tunes the unsupervised learning model followed by a model fitting with the help of the feedback mechanism. It obviates the need to label the entire dataset and thus increases the scope of anomaly detection applications. We have compared our proposed model to the existing state-of-the-art anomaly detection baseline models to show its efficacy. The proposed model performed significantly ((P{text{-value}} < 2.2 times {{10}^{{ - 16}}})) better than the other algorithms, with an AUC score of 0.875.

This paper proposes an approach for tracking the behavior of people in a group on video by using convolutional neural networks. At the beginning, definitions of group movement of people are given, and features for accompaniment are defined that can be used to analyze people’s behavior. Next, an algorithm is proposed for calculating the distance between people in video, which includes three stages: detection and tracking of objects, coordinate transformation, calculation of the distance between people and detection of distance violations. The results of experimental studies and comparison with known algorithms are presented, which confirms the effectiveness of the algorithm.

Cardiac magnetic resonance imaging (MRI) commonly yields numerous images per scan, and manually delineating structures from these images is a laborious and time-intensive task. The automation of this process is highly desirable as it would enable the generation of crucial clinical measurements like ejection fraction and stroke volume. However, due to variations in scanning settings and patient characteristics, automated segmentation faces several challenges that lead to a high degree of variability in picture statistics and quality. Our study presents a neural network approach that utilizes the UNet and ResNet-50 architectures to efficiently partition the left and right ventricles' endocardial and epicardial boundaries. The Dice metric is used as the loss function in our strategy to maximize the trainable parameters in the network. Additionally, in the neural network’s predicted binary picture, we employed a preprocessing step to save just the segmentation labels' most connected component. Using datasets from the Multi-Vendor & Multi-Disease Cardiac Image Segmentation Challenge, the suggested method was learned. The test set of 160 that had been reserved for testing was used by the challenge organizers to evaluate the approach.