Transformer-LSTM-KOA: A novel approach for prediction of chlorophyll in the South China Sea

A three-dimensional model of the ankle system was constructed based on CT scan image data with MIMICS software. This model was refined through surface smoothing and fitting operations in Geomagic software. Finite element method (FEM) was applied to simulate the forces acting on the ankle system during standing in order to generate plantar pressure nephogram. FEM results were compared with the experimental results, which illustrated a strong correlation regarding pressure peaks, thus indicating the effectiveness of the proposed model. Moreover, by changing the interaction force between the ankle system and the ground at different angles during the gait cycle, the FEM was utilized to obtain the curve of peak plantar pressure throughout the gait cycle in order to provide data for the training and testing of the AI algorithm model. The Long Short-Term Memory (LSTM) and Recurrent Neural Network (RNN) models were developed to predict peak plantar pressure during the gait cycle. After training, the mean squared error (MSE) of the LSTM model converged to [Formula: see text], and the coefficient of determination ([Formula: see text] on the test set reached 99.47%. Meanwhile, the MSE of the RNN model converged to [Formula: see text], and the [Formula: see text] on the test set reached 99.20%. Additionally, the prediction results were compared with experimental data to evaluate the validity of both models. The prediction accuracy of the two algorithms was evaluated based on the [Formula: see text]value. The results show that both models’ predictions effectively reflect variations in peak plantar pressure, with the LSTM model proving to be more accurate for predicting peak plantar pressure throughout the gait cycle. Moreover, when taking the overall computation time into account, the RNN model demonstrates significantly higher efficiency in general.

Personalized recommendation systems have transformed numerous industries, including healthcare, by enabling tailored solutions to user needs. This paper presents an innovative recommendation system designed to assist users in identifying appropriate healthcare centers through the integration of clustering and association mining techniques. By combining these methodologies, the system enhances the accuracy and relevance of recommendations while addressing the inherent challenges of healthcare center selection. Specifically, clustering is employed to group healthcare centers based on shared attributes, and association mining uncovers hidden patterns and relationships that improve recommendation quality. The proposed system's effectiveness is evaluated through empirical testing and performance benchmarking against conventional recommendation models. A significant focus is placed on overcoming challenges such as the Cold Start problem and data sparsity, which often limit the efficiency of recommendation systems. To address these issues, a hybrid approach is introduced, integrating data clustering with the Eclat Algorithm to produce more effective association rules. The process begins with clustering the rating matrix based on user similarities, followed by converting the clustered data into Boolean form for application of the Eclat Algorithm, resulting in a refined and optimized recommendation system.

Liver disease, a major global health issue causing approximately 2 million deaths annually, requires accurate predictive models for early detection. This study proposes GANN, a novel framework combining Generative Adversarial Networks (GANs) for synthetic data generation and Artificial Neural Networks (ANNs) for liver disease classification using the Indian Liver Patient Dataset (ILPD). The ILPD, with 583 samples and 10 features, faces challenges like missing values (1.7% in Albumin_and_Globulin_Ratio), class imbalance (416 liver disease vs. 167 non-liver disease cases), and outliers. We address these through preprocessing techniques such as MICE imputation, log transformation, and Proximity Weighted Synthetic Oversampling (PROW). Five GAN variants?CTGAN, Vanilla GAN, Copula GAN, Gaussian Copula GAN, and TVAE?generate 2,000 synthetic samples, validated by Kolmogorov-Smirnov (KS) tests (mean correlation 0.92 with real data). Visualizations, including histograms and correlation matrices, reveal data distributions and relationships. The GANN model achieves 95% accuracy with combined real and synthetic data, compared to 90% with real data alone, outperforming state-of-the-art methods (82?91.2% accuracy). These results suggest GANN?s potential as a robust tool for liver disease prediction, pending further validation.

This study proposes a novel approach for predicting variations in water quality at wastewater treatment plants (WWTPs), which is crucial for optimizing process management and pollution control. The model combines convolutional bi-directional gated recursive units (CBGRUs) with adaptive bandwidth kernel function density estimation (ABKDE) to address the challenge of multivariate time series interval prediction of WWTP water quality. Initially, wavelet transform (WT) was employed to smooth the water quality data, reducing noise and fluctuations. Linear correlation coefficient (CC) and non-linear mutual information (MI) techniques were then utilized to select input variables. The CBGRU model was applied to capture temporal correlations in the time series, integrating the Multiple Heads of Attention (MHA) mechanism to enhance the model's ability to comprehend complex relationships within the data. ABKDE was employed, supplemented by bootstrap to establish upper and lower bounds of the prediction intervals. Ablation experiments and comparative analyses with benchmark models confirmed the superior performance of the model in point prediction, interval prediction, the analysis of forecast period, and fluctuation detection for water quality data. Also, this study verifies the model's broad applicability and robustness to anomalous data. This study contributes significantly to improved effluent treatment efficiency and water quality control in WWTPs.

Early hospital readmission refers to unplanned emergency admission of patients within 30 days of discharge. Predicting early readmission risk before discharge can help to reduce the cost of readmissions for hospitals and decrease the death rate for Intensive Care Unit patients. In this paper, we propose a novel approach for prediction of unplanned hospital readmissions using discharge notes from the MIMIC-III database. This approach is based on first extracting relevant information from clinical reports using a pretrained Named Entity Recognition model called BioMedical-NER, which is built on Bidirectional Encoder Representations from Transformers architecture, with the extracted features then used to train machine learning models to predict unplanned readmissions. Our proposed approach achieves better results on clinical reports compared to the state-of-the-art methods, with an average precision of 88.4% achieved by the Gradient Boosting algorithm. In addition, explainable Artificial Intelligence techniques are applied to provide deeper comprehension of the predictive results.

INTRODUCTION: Heart disease remains one of the leading causes of mortality worldwide, necessitating the development of accurate and efficient prediction models OBJECTIVES: To research new models for heart disease prediction METHODS: This paper presents a novel approach for predicting heart disease using advanced artificial intelligence (AI) techniques, including machine learning (ML) and deep learning (DL) algorithms RESULTS By leveraging patient data and integrating various AI models, this approach aims to enhance prediction accuracy and support early diagnosis and intervention CONCLUSION: This study presents a novel AI-based approach for heart disease prediction, demonstrating the efficacy of ML and DL models in improving diagnostic accuracy

ABSTRACT This study explores the potential of the GPBoost approach for groundwater quality assessment in comparison to three other gradient boosting-based algorithms. Three methods, random search, grid search, and Bayesian optimization were used to find the optimal values of various hyperparameters with all four-gradient boosting-based algorithms. One hundred and two samples of Entropy weighted water quality index with 14 input parameters are used for assessing groundwater quality. The calculated EWQI values for drinking range between 80.4 and 394.96 in pre-monsoon and 39.6 to 338.79 during the post-monsoon period. Moreover, spatial distribution maps displayed that the central portions of the study area fall under medium water quality. The performances of models were compared based on multiple statistical criteria, including Correlation Coefficient (CC), root mean square error (RMSE), and mean absolute error (MAE). The results reveal that the CC value by all modeling approaches is more than 0.93, suggesting a comparable performance by all methods. Results in terms of RMSE values in predicting the EWQI values suggest GPBoost (random search) model performed better than the other three models, thus suggesting a competitive performance by GPBoost in comparison to other gradient boosting-based approaches. Relative importance analysis provided by random and grid search methods highlights the significance of NO3 −, Mg2+, TDS, EC, and TH as important input parameters for predicting EWQI.

A novel approach for prediction of eye microcirculatory disorder using enhanced random forest algorithm and accuracy comparison with Naive Bayes algorithm

Covid19 corona virus has caused widespread disruption across the world, in terms of the health, economy, and society problems. X-ray images of the chest can be helpful in making an accurate diagnosis because the corona virus typically first manifests its symptoms in patients' lungs. In this study, a classification method based on deep learning is proposed as a means of identifying lung disease from chest X-ray images. In the proposed study, the detection of covid19 corona virus disease from chest X-ray images was made with MobileNet and Densenet models, which are deep learning methods. Several different use cases can be built with the help of MobileNet model and case modelling approach is utilized to achieve 96% accuracy and an Area Under Curve (AUC) value of 94%. According to the result, the proposed method may be able to more accurately identify the signs of an impurity from dataset of chest X-ray images. This research also compares various performance parameters such as precision, recall and F1-Score.

Generative Adversarial Network (GAN) is a revolution in modern artificial systems. Deep learning-based Generative adversarial networks generate realistic synthetic tabular data. Synthetic data are used to enhance the size of a relatively small training dataset while ensuring the confidentiality of the original data. In this context, we implemented the GAN framework for generating diabetes data to help the health care professional in more clinical applications. GAN is used to validate the Pima Indian Diabetes (PID) Dataset. Various preprocessing techniques, such as handling missing values, outliers and data imbalance problems, enhance data quality. Some exploratory data analyses, such as heat maps, bar graphs and histograms, are used for data visualisation. We employed hypothesis testing to examine the resemblance between real data and GAN-generated synthetic data. In this study, we proposed a GAN-Long Short-Term Memory (GLSTM) system, in which GAN is used for data augmentation, and LSTM is used for diabetes classification. Additionally, various GAN models such as CTGAN, Vanilla GAN, Coupula GAN, Gaussian Coupula GAN, and TVAE GAN are used to generate the synthetic dataset. Experiments were conducted on real data, synthetic data, and by combining real and synthetic data. The model that used both real and synthetic data obtained a substantially better accuracy of 97% compared to 92% when only real data was used. We also observed that synthetic data could be used in place of real data, as the mean correlation between synthetic and real data is 0.93. Our study's findings outperformed when compared to state-of-the-art methodologies.

In fiber-reinforced composites, fibers are randomly positioned during the manufacturing process. Most of the theoretical studies assume regular arrangement of fibers in matrix that permits the use of representative volume element (RVE) for prediction of mechanical properties. Theoretical methods established in the literature for random fiber composites are complex and approximate. A simple approach is established in this work to accurately predict the transverse thermal conductivity of unidirectional random fiber-reinforced composites using the principle of electrical analogy (EA). In this study, circular fiber is transformed into an equivalent square fiber that facilitates the application of EA, and predicted the transverse thermal conductivity of random fiber composites. The conductivity of equivalent square fiber is evaluated by back-track approach using an equivalent thermal system. Python code is developed to generate the random location of fibers in the composite. The results obtained in this way are verified and found to be in good agreement with benchmark results over a wide range of fiber composites with varying conductivity ratios, fiber volume fraction, and randomness.

Gestational diabetes mellitus occurs due to high glucose levels in the blood. Pregnant women are affected by this type of diabetes. A blood test is to be performed to identify diabetes. The Oral Glucose Tolerance Test (OGTT) is a blood test performed between the 24th and 28th week of pregnancy that is necessary to identify and overcome the side effects of GDM. The main objective of this work is to train a model by utilizing the training data, evaluate the trained model using the test data, and compare existing machine learning algorithms with a Gradient boosting machine (GBM) to achieve a better model for the effective prediction of gestational diabetes. In this work, the analysis was done with a few existing algorithms and the Extreme learning machine and Gradient boosting techniques. The k-fold cross-validation technique is applied with values of k as 3, 5, and 10 to obtain better performance. The existing algorithms implemented are the Naive Bayes classifier, Support Vector Machine, K-Nearest Neighbour, ID3, CART and J48. The proposed algorithms are Gradient boosting and ELM. These algorithms are implemented in R programming. The metrics like accuracy, kappa statistic, sensitivity/Recall, specificity, precision, f-measure and AUC are used to compare all the algorithms. GBM has obtained better performance than existing algorithms. Then finally, GBM is compared with the other proposed robust Machine Learning algorithm, namely the Extreme learning machine, and the GBM performed better. So, It is recommended to use a gradient-boosting algorithm to predict gestational diabetes effectively.

The primary objective is to construct a sustainable machine-learning model that utilizes multiple variables to forecast the success of a startup enterprise. It incorporates a Flask application for creating a user-friendly interface, where users can input specific parameters related to a startup, such as financial metrics, industry sector, and location. These inputs are then passed through a sustainable machine learning prediction model, which has been trained on a comprehensive dataset of startup information. The model employs sustainable advanced algorithms to evaluate their startup ventures' potential success. Through the development and deployment of the Flask application and the integration of sustainable machine learning prediction model, this model contributes to the field of startup analysis and decision-making. It offers a sustainable and efficient solution for predicting startup success, empowering users to make data-backed decisions and optimize their resource allocation.

A novel approach for prediction of exothermic decomposition temperature of energetic complexes through additive and non-additive descriptors

A novel approach for prediction of surface roughness in turning of EN353 steel by RVR-PSO using selected features of VMD along with cutting parameters

The accurate methods for the forecasting of hydrological characteristics are significantly important for water resource management and environmental aspects. In this study, a novel approach for daily streamflow discharge data forecasting is proposed. Streamflow discharge, temperature, and precipitation data were used for feature extraction which were systematically employed for forecasting studies. While the correlation-based feature selection (CFS) was used for feature selection, Random Forest (RF) model is employed for forecasting of following 7 days. Moreover, an accuracy comparison between the RF model and CFS-RF model is drawn by using streamflow discharge data. Acquired results confirmed the accuracy of CFS-RF model for both, middle and extended forecasting times compared to RF model which had similar accuracy values for the closer forecasting times. Moreover, the CFS-RF model proved to be much robust for extended forecasting durations.

The aim of this paper is to improve Accuracy in Disease prediction using symptoms by a novel multilayer perceptron classifier in comparison with the naive Bayes algorithm . Materials and Methods : Novel Multilayer perceptron classifier and naive bayes algorithm sample size (N=10) to predict the accuracy percentage of predicted disease. G-power is calculated for two different groups, alpha (0.05), power (80%). Results: Based on the measurement of data, Statistical Analysis and independent sample T-test, it shows that there is a statistically insignificant difference between the two study groups with value p=0.212 (p> 0.05). It was observed that the novel multilayer perceptron algorithm obtains an accuracy of 95%. It appears to have better accuracy than the naive Bayes algorithm (92%). Conclusion: The results prove that novel multilayer perceptron algorithm approaches with varied seed value have significant improvement in disease prediction using symptoms.

Aim : The aim of this paper is to improve Accuracy in Disease prediction using symptoms by a novel multilayer perceptron classifier in comparison with the K nearest neighbor algorithm .Materials and Methods : Multilayer perceptron classifier and naive bayes algorithm sample size (N=10) to predict the accuracy percentage of predicted disease.G-power is calculated for two different groups, alpha (0.05), power (80%).Results: Based on the measurement of data, statistical analysis, and independent sample T-test,there is a statistically insignificant difference between the two study groups with value p=0.768 (p>0.05) .It was observed that the novel Multilayer perceptron algorithm obtains the accuracy as 95%.It appears to have better accuracy than the K nearest neighbor (81%).Conclusion: The results prove that the novel Multilayer perceptron algorithm approaches with varying seed value have significant improvement in disease prediction using symptoms.

SummaryOnline social network is a platform that plays an essential role in identifying the emotional values of user‐generated content such as blogs, posts, and comments along with their influential factors. Especially on Twitter, network users are growing worldwide day by day and creating a massive amount of data that is not analyzed effectively in a quick way. Identifying the most influential persons on the social network is also a challenging task over the wide range of real‐time applications like recommendation systems. Now, to handle these situations, this article proposes a novel approach for prediction of information diffusion that includes emotion recognition with sarcasm detection based influence spreader identification (PID‐ERSDISI). The proposed method uses the user‐generated posts for emotion recognition in tandem with sarcasm detection both implicitly and explicitly. This approach helps to gauge the leverage that influences spreaders and also enhances the prediction accuracy of information diffusion in a better way. The implementation of the proposed work executed their task one after another in the following way, namely, sarcasm detection, emotional‐level computation, breakpoint computation, breakpoint validation, influential user generation, and information diffusion. After the successful implementation of this proposed PID‐ERSDIS, it produced prominent results against other state‐of‐art methods.