Music classification is an essential application of Music Information Retrieval (MIR) in organizing extensive collections of music. The tasks to classify different music with reliable accuracy observed to be challenging. Most of these tasks employ handcrafted feature engineering to build a classifier, yet unable to identify the original characteristics of music. Several combinations of neural networks using convolutional neural networks (CNNs) and recurrent neural networks (RNNs) have been in consideration of many researchers. However, it has been noticed that the joint architecture of CNN and RNN suffers some problems due to batch normalization, which causes low accuracy and more training time. To handle these issues, the Global Layer Regularization (GLR) technique is proposed on the hybrid model of CNN and RNN using Mel-spectrograms for the evaluation of training and accuracy. Our experiments, with few hyper-parameters, improve performance on GTZAN and Free Music Achieve (FMA) datasets by achieving modest accuracy of 87.79% and 68.87% respectively. Empirically, our proposed model takes the advantages of spatiotemporal domain features and the global layer regularization technique to accomplish reliable accuracy as compared to the other state of art works.

Breast cancer affects a large number of women around the world who are more likely to die as a result of this condition. To seek out the main cause of breast cancer, samples were collected by employing a variety of cutting-edge procedures. The most modern techniques used in this regard are logistic regression, discriminant analysis and principal component analysis (PCA), all of which are useful in determining the causes of breast cancer. The Breast Cancer Wisconsin Diagnostic Dataset collects information about breast cancer via the machine learning repository approach. As a result of the data correlation matrix, we were able to root our job positively. PCA, discriminant analysis, and logistic regression were utilized to extract the dataset features. Models such as decision tree, naive Bayes, logistic regression, support vector machine (SVM), and artificial neural networks were utilized, and their performances were rigorously examined. The results suggested that the proposed strategy works effectively and reduces the training time. These new methods will help doctors to understand the origins of breast cancer and to distinguish between tumor kinds. Data mining techniques are used extensively, especially for feature selection. Finally, it was concluded that among all models, the hybrid discriminant-logistic (DA-LR) feature selection model outperforms SVM and naive Bayes.

Music classification using deep neural networks has gained a lot of attention in recent years. This is due to the difficult task of capturing every essential aspect of music in features and interpretability of classifiers. There is limited research on the integration of VGG16 and RNNs, but the researchers found that few classifiers accurately capture intrinsic musical characteristics. Previous work in this field has primarily focused on spectral features, which has constrained overall performance. To address this issue, we proposed a novel hybrid neural architecture based on Visual Geometry Group 16 (VGG16), which is highly effective in extracting important features from musical variations. We combined VGG16 with several recurrent neural network (RNN) variants, including Gated Recurrent Unit (GRU), Bidirectional GRU (BiGRU), Long Short-Term Memory (LSTM), and Bidirectional LSTM (BiLSTM). Additionally, we compared their performance for the GTZAN dataset using both Mel-Spectrogram and Mel-Frequency Cepstral Coefficients (MFCC) features. Our results indicate that the VGG16+GRU model achieved the highest accuracy of 89. 60% with Mel spectrograms and 82. 70% with MFCC features. These findings demonstrate the effectiveness of combining advanced feature extraction techniques with deep learning models for music genre classification.

The rapid adoption of online examinations has increased the demand for secure, intelligent, and reliable assessment systems. However, existing solutions often lack adaptive evaluation capabilities and formal verification of examination workflows. This paper presents a Context-Aware Evaluation Framework that addresses these limitations through an integrated architecture. The framework combines secure examination management, LLM-based assessment, and formally verified workflow execution. A key contribution is a dynamic rubric ingestion pipeline that transforms evaluation rubrics into structured contextual inputs for the grading model. The proposed rubric representation (R={C,S,W,X}), captures evaluation criteria, scoring scales, contextual weights, and anchor exemplars. This design enables context-aware and rubric-conditioned grading of diverse student responses. Grading consistency is evaluated using the Quadratic Weighted Kappa (QWK) metric. Another contribution is a Petri net-based workflow model that formally represents concurrent and state-dependent examination processes. The model incorporates synchronization mechanisms for complete rubric evaluation and bounded correction loops for reliable LLM output re-parsing. Reachability graph analysis confirms liveness, boundedness, and deadlock-freedom, ensuring mathematically verified workflow correctness. The framework also strengthens examination security through facial recognition, encryption, cryptographic hashing, ActiveX-based anti-cheating mechanisms, and natural language processing-based spoken-text relevance analysis. By integrating intelligent assessment, formal verification, and multi-layer security, the proposed framework provides a reliable foundation for next-generation online examination systems. Future work will investigate real-time instructor intervention and adaptive prompt refinement to further improve grading accuracy and workflow resilience.