Subject matter: Speech emotion recognition (SER) is an ongoing interesting research topic. Its purpose is to establish interactions between humans and computers through speech and emotion. To recognize speech emotions, five deep learning models: Convolution Neural Network, Long-Short Term Memory, Artificial Neural Network, Multi-Layer Perceptron, Merged CNN, and LSTM Network (CNN-LSTM) are used in this paper. The Toronto Emotional Speech Set (TESS), Surrey Audio-Visual Expressed Emotion (SAVEE) and Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS) datasets were used for this system. They were trained by merging 3 ways TESS+SAVEE, TESS+RAVDESS, and TESS+SAVEE+RAVDESS. These datasets are numerous audios spoken by both male and female speakers of the English language. This paper classifies seven emotions (sadness, happiness, anger, fear, disgust, neutral, and surprise) that is a challenge to identify seven emotions for both male and female data. Whereas most have worked with male-only or female-only speech and both male-female datasets have found low accuracy in emotion detection tasks. Features need to be extracted by a feature extraction technique to train a deep-learning model on audio data. Mel Frequency Cepstral Coefficients (MFCCs) extract all the necessary features from the audio data for speech emotion classification. After training five models with three datasets, the best accuracy of 84.35 % is achieved by CNN-LSTM with the TESS+SAVEE dataset.

Farming has evolved from the basic irrigation techniques used in ancient river valley civilizations to the sophisticated Precision Agriculture of today. It plays an important role in the advancement of human society. This paper explores the use of Machine Learning and Deep Learning algorithms in Precision Agriculture, an essential task in agriculture that helps ensure a stable food supply and improves the efficiency of food production. Despite advances in Precision Agriculture and the widespread adoption of Machine Learning and Deep Learning algorithms, a comprehensive review that systematically addresses the challenges of data quality, model interoperability, and multisource data integration in Precision Agriculture is still lacking. We aim to bridge this gap by analyzing more than 100 related studies. We focus on applying several Machine Learning and Deep Learning algorithms, such as Artificial Neural Networks, Support Vector Machines, Convolutional Neural Networks, Random Forests, etc. We use a comparative analysis methodology to identify key features influencing Precision Agriculture, such as temperature, rainfall, remote sensing data, soil types, etc. Our findings highlight continuous challenges in standardizing data protocols and developing Explainable AI models that can be generalized across diverse agricultural conditions. The key takeaway is that integrating IoT with real-time data processing can significantly improve agricultural resilience and efficiency. Future research should focus on refining robust models and expanding multisource data integration to address these challenges effectively.

Digital image watermarking is a crucial technique used to protect the integrity and ownership of digital images by embedding imperceptible watermarks into the image content. This review concentrates on the utilization of discrete cosine transform (DCT) and discrete wavelet transform (DWT) in digital image watermarking schemes. DCT, widely used in image compression like JPEG, is an attractive choice for watermarking, modifying DCT coefficients with minimal impact on image quality. On the other hand, DWT offers multiresolution representation, enabling better localization and robustness against attacks. DWT-based methods use wavelet coefficients to embed watermarks in specific frequency bands or image regions. The review examines the strengths and weaknesses of DCT and DWT in digital image watermarking, exploring algorithms and approaches proposed in the literature. It also addresses challenges like attacks, synchronization, and robustness to image processing. Additionally, a comparative analysis of DCT and DWT-based methods considers imperceptibility, robustness, capacity, and computational complexity. By offering valuable insights, this review aids researchers and practitioners in implementing secure and efficient digital image watermarking solutions.

The subject matter of this research revolves around addressing the escalating global health threat posed by cardiovascular diseases, which have become a leading cause of mortality in recent times. The goal of this study was to develop a comprehensive diet recommendation system tailored explicitly for cardiac patients. The primary task of this study is to assist both medical practitioners and patients in developing effective dietary strategies to counter heart-related ailments. To achieve this goal, this study leverages the capabilities of machine learning (ML) to extract valuable insights from extensive datasets. This approach involves creating a sophisticated diet recommendation framework using diverse ML techniques. These techniques are meticulously applied to analyze data and identify optimal dietary choices for individuals with cardiac concerns. In pursuit of actionable dietary recommendations, classification algorithms are employed instead of clustering. These algorithms categorize foods as "heart-healthy" or "not heart-healthy," aligned with cardiac patients’ specific needs. In addition, this study delves into the intricate dynamics between different food items, exploring interactions such as the effects of combining protein- and carbohydrate-rich diets. This exploration serves as a focal point for in-depth data mining, offering nuanced perspectives on dietary patterns and their impact on heart health. The method used central to the diet recommendation system is the implementation of the Neural Random Forest algorithm, which serves as the cornerstone for generating tailored dietary suggestions. To ensure the system’s robustness and accuracy, a comparative assessment involving other prominent ML algorithms—namely Random Forest, Naïve Bayes, Support Vector Machine, and Decision Tree, was conducted. The results of this analysis underscore the superiority of the proposed -based system, demonstrating higher overall accuracy in delivering precise dietary recommendations compared with its counterparts. In conclusion, this study introduces an advanced diet recommendation system using ML, with the potential to notably reduce cardiac disease risk. By providing evidence-based dietary guidance, the system benefits both healthcare professionals and patients, showcasing the transformative capacity of ML in healthcare. This study underscores the significance of meticulous data analysis in refining dietary decisions for individuals with cardiac conditions.