Lung cancer remains one of the leading causes of cancer-related deaths globally, where early detection plays a crucial role in improving survival rates. The disease includes various types, such as adenocarcinoma, large cell carcinoma, and squamous cell carcinoma, which can be identified using lung CT imaging. However, accurately classifying these subtypes through imaging is challenging due to their similar visual features. Recent progress in artificial intelligence, particularly deep learning, has introduced effective methods to enhance diagnostic precision through automated image analysis. This research proposes a hybrid deep learning approach that combines ResNet50 with a Vision Transformer (ViT) for classifying lung cancer CT images. ResNet50 effectively captures hierarchical spatial features, while the ViT component leverages global attention to extract contextual information. The proposed model was trained for 100 epochs, achieving 99

In this research, using a dataset of 1000 patient records that we have taken from Kaggle dataset repository, on which we have performed a thorough analysis to forecast the risk levels of lung cancer, which were divided into three categories: low, medium and high. We attempted to train and test the data for optimal predictive accuracy using sophisticated machine learning models, specifically Naive Bayes (NB), Logistic Model Tree (LMT), Random Tree (RT) and Decision Stump (DS). The Logistic Model Tree (LMT) was found to be the best performer following a thorough evaluation based on key performance indicators like accuracy, Root Absolute Error (RAE), Mean Absolute Error (MAE), Root Relative Square Error (RRSE) and Root Mean Square Error (RMSE). Notably, the LMT model outperformed the other three models with an astounding accuracy rate of 98

In response to growing environmental concerns and the need for sustainable development, alternative energy systems have emerged as crucial solutions to the problems associated with fossil fuels. This research focuses on the significant role cloud computing plays in improving the efficiency and integration of various alternative energy systems such as solar, wind, hydropower, biomass, geothermal, and tidal/wave energy. The study explores how cloud computing technologies enable real-time data processing, storage, and analysis of the large datasets generated by these renewable energy sources, thus optimizing their performance and integration into the energy grid. Cloud computing provides advanced data management tools that are vital for managing the variable and intermittent nature of renewable energy production. By utilizing cloud-based platforms, energy producers can achieve improved forecasting, load balancing, and energy distribution, which are essential for enhancing efficiency and reliability. Additionally, this research highlights how cloud computing fosters collaboration and innovation by offering a scalable and flexible infrastructure. This infrastructure supports the development of new technologies and applications within the renewable energy sector, promoting continuous advancements and breakthroughs. This research underscores the crucial role of cloud computing in advancing a sustainable energy future.

Fire outbreak is a common issue which is occurring worldwide, causing significant damage to both nature and human life. Recently, vision-based fire detection systems have gained popularity over traditional sensor-based systems. However, the detection process using image processing techniques can be tedious. In the current study, we propose a technique for fire detection that utilizes optimal convolution neural networks (OPCNN) to achieve highly accurate detection of fire images in forest. The result of proposed model is compared with two other models: CNN and J48. The proposed model performs better than these models. The proposed algorithm was trained using a dataset consisting of 755 images of fire and 244 images of non-fire, for a total of 999 images. These images were obtained from Kaggle data set. We resized and reshaped 1380 of these images for use in training and 460 images for testing. The model was trained using convolution, activation functions, and max pooling operations with different batch sizes and epoch values. The resulting model achieved an accuracy of 95.11

Lung cancer remains an ongoing global health challenge, necessitating the development of innovative methods for early detection and risk assessment. This study presents a comprehensive approach for predicting lung cancer by utilizing a carefully curate dataset consisting of 1000 individuals from Kaggle dataset. The cutting-edge machine learning models, including Support Vector Machines (SVM), Naive Bayes Multinomial (NBM), Meta Bagging, PART and Random Forest (RF are used to enhance the accuracy of our predictions. The dataset we’ve compiled includes a wide array of patient characteristics, encompassing demographics, lifestyle factors, medical history, and health data gathered through IoT devices. By harnessing the capabilities of IoT technology, it is possible to enable real-time and continuous health monitoring, facilitating a dynamic assessment of lung cancer risk. The finding of this study reveal that the PART model achieves an impressive accuracy of 95%, surpassing other models like NBM (67%), RF (93%), SVM (92%), and Bagging (86%). This innovative approach shows promise in the early detection of lung cancer and the provision of personalized risk assessments, potentially leading to improved patient outcomes and reduced healthcare burdens.

Groundwater, a vital resource for various purposes, faces increasing threats to its quality due to factors like overuse, pollution and climate change. In the semi-arid climate of the state of Telangana, India, exacerbated by intensive agricultural practices, groundwater quality is a significant concern. This study presents a machine learning-based approach for predicting and managing groundwater quality in Telangana, utilizing a stacking ensemble method. Our proposed approach combines decision trees (DT) with meta-models such as logistic regression, K-nearest neighbors (K-NN), Bayesian ridge regression (BRR) and random forest (RF). Trained on a comprehensive dataset from the Telangana Open Data portal (2018-2021), the stacking ensemble demonstrates superior performance, achieving an accuracy of 96.07%, surpassing the best individual model (DT) at 92.68%. It not only enhances predictive capabilities but also provides insights into the spatial distribution of groundwater quality, identifying areas vulnerable to contamination. These findings offer valuable information for policymakers, environmental scientists and water resource managers, guiding informed decision-making for sustainable water resource management in Telangana. This research contributes to the ongoing discourse on environmental conservation and the imperative for effective groundwater management.

Digestive disorders have become very rampant these days and some of these disorders which include gastritis may cause rise in stomach polyps. It is therefore important to identify these precancerous polyps early in the disease process so that treatment can be taken to ensure they do not grow into cancer. Specifically, in the context of semantic segmentation, there has been growing concern on using deep learning methods. Among them, several segmentation models that have been introduced here have been used in various medical image segmentation tasks. With respect to our research, we investigate semantic segmentation using the U-Net3+ architecture on Kvasir-Sessile, containing 1,200 high resolution endoscopic images of sessile polyps; a diverse set of polyps that challenge segmentation due to their subtle and diverse appearance. We applied U-Net3+ to perform the segmentation of sessile polyps in the Kvasir-Sessile dataset. The increase in the ROC AUC to 0.987, MCC of 0.998, and F1-score to 0.977 from the experiments indicate that our proposed method reaches higher accurate in segmenting sessile polyps from complex endoscopic background. In addition, through qualitative analysis of the segmented images, histogram intersection, Hausdorff distance, and MSE values supported by the visual inspection confirmed precise and accurate segmentation of the various shapes and sizes of polyps.

In an increasingly interconnected world, language diversity should be celebrated rather than hindered. This paper delves into the significance of open-source technology as a means of breaking down language barriers and promoting linguistic inclusivity. Focusing on the development of an open-source language typing tool, this research addresses a critical need in digital communication. While several commercial software solutions exist for language input, their proprietary nature limits accessibility and customization. In contrast, the proposed tool, crafted in Python, embraces the open-source ethos, fostering collaboration and allowing for tailored solutions to accommodate the unique linguistic nuances of Indian Languages. The core aim of this study is twofold. First, it seeks to establish a robust and user-friendly open-source input tool for Odia and Hindi with both Rule Based and free typing. Second, it lays the foundation for broader Indian language integration, recognizing the rich tapestry of languages spoken throughout the country.

Real-time crowd monitoring plays a pivotal role in effectively managing public spaces and ensuring safety. This study investigates the fusion of IoT devices and the YOLO object detection model to accurately count crowds. IoT devices facilitate the instantaneous collection of data from cameras, while the YOLO model adeptly identifies individuals within recorded video frames. The study rigorously assesses the performance of three YOLO variants: YOLO V5, YOLO V8 and YOLO V8 NAS. Findings reveal that YOLO V8 NAS surpasses YOLO V5 and YOLO V8 in mean average precision (mAP), achieving an exceptional mAP of 95.1%. This heightened precision is attributed to the integration of Neural Architecture Search (NAS) into the YOLO V8 NAS model, fine-tuning its architecture specifically for crowd counting tasks. It analyzes various networking models proposed in earlier studies for analyzing crowded scenes in public spaces. It emphasizes the potential of a hybrid model involving an IP camera module and a Deep Neural Network for effective crowd sensing. In this setup, the IP camera captures video footage, while the DNN detects individuals and assesses crowd density based on the count of people recognized. This approach presents an encouraging solution for real-time crowd monitoring and management in public environments.

The field of healthcare has witnessed significant advancements in leveraging technology for in-depth analysis of fetal health data, offering profound insights into child development. We meticulously gathered an extensive dataset comprising 1500 instances of child fetal health data sourced from New Born and Pediatric Health Centre, Cuttack, Odisha. We proposed a robust analytical pipeline for effectively classifying and predicting fetal health outcomes. The process initiated with meticulous data preprocessing and refinement, ensuring data integrity and quality. We use clustering techniques unveiled inherent patterns and groupings within the dataset, setting the stage for in-depth analysis and feature extraction. Further enriching the analysis, Principal Component Analysis (PCA) is deployed to streamline data dimensionality while retaining essential information. For prediction we use ensemble learning techniques. To predict the condition, we used voting ensemble method encompassing Decision Trees (DT), Support Vector Machines (SVM) and Naïve Bayes (NB), after predictive models are successfully constructed, showcasing notable accuracies of 92.28%, 83.89% and 82.22%. These results underscore the effectiveness of these models in predicting fetal health outcomes based on the comprehensive dataset. To further refine model performance, a meticulous GridSearchCV approach exhaustively explored optimal parameter values within predefined parameter grids, significantly enhancing predictive capabilities. Additionally, leveraging a bagging method, particularly Random Forest (RF), Logistic Regression (LR), and Gradient Boosting, further bolstered predictive accuracies. These bagging models demonstrated exceptional achieving accuracy of 94.02%, 93.22% and 92.28%. The systematic approach involving clustering, dimensionality reduction, ensemble learning, and parameter optimization resulted in highly accurate predictive models for fetal health assessment. Our research not only highlights the potential of machine learning in healthcare but also enhanced predictive models, yielding highly accurate outcomes for fetal health classification.

Cattle diseases can significantly impact on livestock health and agricultural productivity is substantial. Timely detection and prognosis of these diseases are essential for prompt interventions and preventing their spread within the herd. This study delved into employing machine learning models to anticipate cattle diseases based on relevant parameters. These parameters encompass milk fever, milk clots, milk watery, milk flake, blisters, lameness, stomach pain, gaseous stomach, dehydration, diarrhea, vomiting, abdominal issues, and alkalosis. A dataset of 2,000 samples from diverse cattle populations was amassed, each tagged with the presence or absence of specific diseases. The primary goal was to compare the efficacy of five well-known machine learning models: Naïve Bayes multinomial (NBM), lazy-IBk, partial tree (PART), random forest (RF), and support vector machine (SVM). The findings underscored the consistent superiority of RF in comparison to the other models, boasting the highest accuracy in predicting cattle diseases. The RF model exhibited an accuracy rate of 88% on the test dataset. This achievement can be ascribed to its capacity to handle intricate interactions among input features and mitigate over fitting through ensemble learning. These insights can furnish valuable information about early indicators and risk factors associated with diverse cattle diseases.

In this paper, an approach of multi-view learning, with multilayer perceptron (MLP) and radial basis functions (RBF) with dynamic decay adjustment (DDA), has been proposed. Three different categories of semi-supervised learning are multi-view training, co-training and self-training. Here we have only used self-training and multi-view learning mechanisms to train the classifier. To test the accuracy of the algorithms, we have taken five real-time datasets from UCI Machine Learning Repository. The classifier is trained using the perceptron learning rule with its supervised and semi-supervised (self-training) versions and MLP with RBF (multi-view learning). The average classification accuracies have been compared and the proposed algorithm outperforms the former versions on the specified training sets. The significant improvement in performance obtained using multi-view learning can be used for various fields such as detecting changes of images, speech recognition and biometric identification.

Weather monitoring for different parameters is very common nowadays but accurate forecasting of weather or atmosphere for a particular location is a very challenging task. Odisha is a state in eastern India which is very much prone to climatic hazards like cyclones, super cyclones and earthquakes. This state has seen many cyclones in its history. The prediction of cyclones and preparedness as per the prediction can save the life and property of people in mass for which the Government has taken such steps in the past and could be able to save the lives of many people from impending climatic disasters. Therefore, to predict the future weather, here we collected 15 years of weather data set of the Odisha region from a Data access viewer by giving different parameters and implemented the collected dataset in python programming to train the machine to predict the future condition by analysing the huge amount of data given. The machine can analyse and predict the data by using different methods such as Decision Tree (DT), Naïve Bayes (NB) and Support Vector Machine (SVM), then finally calculated that out of these 3 algorithms the accuracy rate of SVM is highest among the other two algorithms and the accuracy is 77

Prognosis of illnesses is a difficult problem these days throughout the globe. Elder people of twenty years and over are taken into consideration to be laid low with this sickness now a days. For example, human beings having HbA1c level more than 6.5% are diagnosed as infected with diabetic diseases. This paper uses IoT to evaluate threat factors which have been similar to heart diseases which are not treated properly. Diagnosis, prevention of heart disease may be done by use of machine learning (ML). There has been an extensive disconnect among Machine Learning architects, health care researchers, patients and physicians in their technology. This paper intends to perform an in-intensity evaluation on Machine Learning to make us of new advance technologies. Latest advances within the development of IoT implanted devices and other medicine delivery gadgets, disease diagnostic methods and other medical research have considerably helped human beings diagnosed heart diseases. New soft computing models can be helpful for remedy of various heart diseases. The Food and Drug Administration (FDA) employs several particularly creative thoughts to get their capsules to the client. Artificial Neural Community offers a first-rate chance to deal with heart diseases with advance IoT and cloud applications.

The term SMS stands for short message service. It is a message networking method that uses smartphones and cell phones. It is a text messaging system that lets smart phones to communicate with one another. SMS has risen in prominence in current history, with firms adapting and optimizing SMS to be used as a service tool mostly because of its numerous benefits. We are using a unique technique that uses data mining and machine learning method to identify and filter spam communications in this study. We analyzed the characteristics of spam SMS in depth and discovered ten criteria that may effectively distinguish short message (or messaging) service spam from ham. If some adjustments are needed to identify a new scam, they were made manually, either by applying changes to current algorithms or by inventing new algorithms. In this method, as the quantity of clients and data grows, so does the amount of human work, which may also effectively separate spam text messages and ham SMS messages. The random forest classification method, we proposed, is the methodology that produced 99.9

Prediction of weather is a challenging task for all researchers of weather and the meteorological department. Many techniques are evolved in time for the prediction of weather since last many years. The advancement of science and technology has helped the researchers to perform the prediction of weather simply and with less error rates. The soft computing techniques are the new technologies in computer science which are capable of making the weather prediction with promising output and less error rates. Weather prediction is done traditionally by use of many historical data in many models of physics. This prediction of weather is unsteady due to the change in weather condition. Because of change of weather system, the prediction is unstable. In this paper, we present different machine learning models that will make use of the historical data to train the models and then the model will be used to predict the weather whose accuracy is better than the traditional models. The evaluation of the models on the basis of accuracy shows that the models outperform and can be used as state-of-art technique to predict the weather in smarter way in less time.

The art of weather forecasting is a challenging task of predicting the state of the atmosphere at a future time for a specified location. Climate change and weather prediction is a highly nonlinear phenomenon which is called butterfly effect. The soft computing techniques are now capable of replacing the conventional weather prediction methods. The proposed new hybrid soft computing models are designed by exploiting the positive features of the constituent soft computing techniques and suppressing their disadvantages and also this research work intends to design the hybrid models by making use of favourable properties of Support Vector Machine, Multi-Layer Perceptron and Fuzzy Logic considering the weather of Delhi. The new hybrid soft computing models are used here to forecast the weather at Delhi by training the models using weather data of Delhi.

Climate modelling and weather prediction, the application of science and technology to predict the state of the atmosphere for a given location is a challenging task for the researchers in this modern age.Now-a-days the soft computing techniques have been evolved which can be used for the prediction of weather with more accuracy and less errors. But, Soft computing technique is a new approach to construct computationally intelligent systems that are supposed to possess humanlike expertise within a specific domain, adapt themselves and learn to do better in changing environments. Hence, this paper tries to analyze soft computing techniques i.e Rough Set and SVM for weather prediction of Delhi and reveals that Rough set exhibits promising result that SVM.