Introducing an innovative approach to stress detection through multimodal data fusion, this study addresses the critical need for accurate stress level monitoring, essential for mental health assessments. Leveraging diverse data sources—including audio, biological sensors, social media, and facial expressions—the methodology integrates advanced algorithms such as XG-Boost, GBM, Naïve Bayes, and BERT. Through separate preprocessing of each dataset and subsequent feature fusion, the model achieves a comprehensive understanding of stress levels. The novelty of this study lies in its comprehensive fusion of multiple data modalities and the specific preprocessing techniques used, which improves the accuracy and depth of stress detection compared to traditional single-modal methods. The results demonstrate the efficacy of this approach, providing a nuanced perspective on stress that can significantly benefit healthcare, wellness, and HR sectors. The model's strong performance in accuracy and robustness positions it as a valuable asset for early stress detection and intervention. XG-Boost achieved an accuracy rate of 95%, GBM reached 97%, Naive Bayes achieved 90%, and BERT attained 93% accuracy, demonstrating the effectiveness of each algorithm in stress detection. This innovative approach not only improves stress detection accuracy but also offers potential for use in other fields requiring analysis of multimodal data, such as affective computing and human-computer interaction. The model's scalability and adaptability make it well-suited for incorporation into existing systems, opening up opportunities for widespread adoption and impact across various industries.

The electrical load forecasting plays a vital role on the economy of a country in context of fuel saving, working hours of employee and depreciation cost of equipment of power generating station.  In this paper, we use several machine learning techniques relevant to fuzzy system to forecast the demand of electrical load on short-term basis. Here, we consider temperature, humidity, wind speed, types of day such as working day or holiday, barometric pressure as the parameters, which govern the demand of electrical load. To cope with the variables and the power demand, the previous data of Bangladesh Power Development Board (BPDB) and Bangladesh Space Research and Remote Sensing Organization (SPARRSO) were taken for training purpose and then data of current day was used as the test data. For each of the weather parameter several membership functions (MFs) were used as the fuzzy input and then Takagi-Sugeno, Mamdani rule, FCM + Mamdani and ANFIS were applied to acquire the output as the demand of load. The average percentage of error as the difference between forecasted demand and actual demand of test data was found 1.675% for Takagi-Sugeno, 1.91% for Mamdani (centroid), 2.56% for FCM + Mamdani and 3.62% for ANFIS, which were found superior to some previous research works.

Smart farming is undergoing a transformation with the integration of machine learning (ML) and artificial intelligence (AI) to improve crop recommendations. Despite the advancements, a critical gap exists in opaque ML models that need to explain their predictions, leading to a trust deficit among farmers. This research addresses the gap by implementing explainable AI (XAI) techniques, specifically focusing on the crop recommendation technique in smart farming.
An experiment was conducted using a Crop recommendation dataset, applying XAI algorithms such as Local Interpretable Model-agnostic Explanations (LIME), Differentiable InterCounterfactual Explanations (dice_ml), and SHapley Additive exPlanations (SHAP). These algorithms were used to generate local and counterfactual explanations, enhancing model transparency in compliance with the General Data Protection Regulation (GDPR), which mandates the right to explanation.
The results demonstrated the effectiveness of XAI in making ML models more interpretable and trustworthy. For instance, local explanations from LIME provided insights into individual predictions, while counterfactual scenarios from dice_ml offered alternative crop cultivation suggestions. Feature importance from SHAP gave a global perspective on the factors influencing the model's decisions. The study's statistical analysis revealed that the integration of XAI increased the farmers' understanding of the AI system's recommendations, potentially reducing food insufficiency by enabling the cultivation of alternative crops on the same land.

The article is devoted to the study of the regularities of the statistical organization of nucleotide sequences of single-stranded DNAs in genomes of higher and lower organisms, as well as their connections with cyclic Gray codes and the problem of holistic structures (gestalts) in physiology. The author presents stable statistical structures of an algebraic nature, which are found in many genomic DNAs and are called algebraic gestalt-archetypes of probabilities in genomic DNAs. They are discussed as a possible basis for several genetically inherited physiological and psychophysical properties. The numerical rules of these genomic archetypes realized in nature for a representative class of genomic DNAs, whose initial data were taken by the author from the publicly available genomic data bank “GenBank”, are formulated. The analysis of single-stranded genomic DNAs was carried out using the author's method of "hierarchies of multilayer statistics", representing the nucleotide sequence of DNA as a multilayer text structure, in which each n-th layer is a sequence of n-plets (that is, of monoplets, or duplets, or triplets, etc.). In each such layer, the percentages of each of the types of its n-plets are calculated, the values of which are inserted into the so-called genetic (2n∙2n) Karnaugh matrices, whose columns and rows are numbered by n-bit Gray codes by analogy with Karnaugh maps from the Boolean algebra of logic. The data of the analysis of the nucleotide sequence of DNA of the human chromosome № 1, containing about 250 million nucleotides, are represented as an example. The obtained data are discussed in light of the problem of genetically inherited holistic structures in biology and the tasks of developing algebraic biology, genetic biomechanics, quantum bioinformatics, artificial intelligence, and genetic algorithms.

Alzheimer’s Disease (AD) is the neuro-degenerative dementia, where the precise and early recognition of AD is vital for timely treatment to reduce mortality rate. A new automated model is implemented in this work for early discovery of AD in the Magnetic Resonance Imaging (MRI) brain scans. Initially, the input brain scans are taken from the Alzheimer's disease Neuroimaging Initiative (ADNI) database. Further, the acquired raw brain scans are visually improved by employing the binary normalization technique. The denoised brain scans are fed to the pre-trained Convolutional Neural Network (CNN) named GoogleNet for feature extraction. Next, the extracted richer feature values are fed to the Long Short Term Memory (LSTM) network for classifying the brain scan as Normal Control (NC), Mild Cognitive Impairment (MCI) and AD. In this manuscript, a Honey Badger Optimization Algorithm (HBOA) technique is incorporated with the LSTM networks for hyper-parameters optimization, where this procedure helps in diminishing the LSTM network’s complexity and computational time. The experimental results conducted on the ADNI database underscore the HBOA-based LSTM network's effectiveness, showcasing a remarkable mean classification accuracy of 97.83% in multi-class classification. Moreover, the sensitivity of HBOA based LSTM for AD/NC is 96.73% which is high when compared to the existing methodologies such as SVM with radial basis kernel function and NCSINs. This performance surpasses that of other comparative models for AD detection, emphasizing the superior capabilities and potential of the proposed method in the early detection.

About one person dies every minute from cardiovascular disease; consequently, it has almost surpassed war as the largest cause of death in the twenty-first century. In cardiology, early and accurate diagnosis of heart illness is a cornerstone of effective healthcare. Predictive analytics, which involves machine-learning algorithms, can be a great option for contributing towards the early detection of cardiovascular disease. This study evaluates the data preprocessing techniques involved in building machine learning models to predict cardiovascular disease and identify the features contributing to the cardio attack. A novel data transformation technique named the superlative boundary binning method was proposed to enhance machine learning and ensemble learning classification models for predicting cardiac illness based on independent physiological feature parameters. The results revealed that the ensemble learning classifier AdaBoost using the superlative boundary binning method has performed well with a classification accuracy of 93% when compared with the other data transformation and machine learning classifier models.