Smart Prediction of Water Quality System for Aquaculture using Machine Learning Algorithms

This article focuses on the importance of the continuous collection of water parameters data from the sensors and also the prediction of water quality using the latest different Machine learning algorithms like Logistic Regression, Random Forest, Support Vector Machine, Decision Tree, K-nearest Neighbour, XGBoost, Gradient Boosting and Naive Bayes. These Machine learning models are implemented and tested to validate and achieve a satisfactory result of water quality prediction in terms of different attributes like pH, hardness, Solids, Chloramines, Sulfate, Conductivity, organic carbon, trihalomethanes, Turbidity and potability.

Analysis and Prediction of Parkinson's Disease using Machine Learning Algorithms

Despite advancements in medical science, some conditions remain incurable, requiring accurate diagnosis to slow patient deterioration. AI and Machine Learning (ML) now enable precise diagnosis, aiding doctors significantly. Parkinson’s Disease, an incurable condition affecting the Central Nervous System (CNS), causes tremors, slowed movement, muscle rigidity, and speech loss. This article explores using ML models—K-Nearest Neighbour (KNN), Logistic Regression (LR), Random Forest Classifier (RFC), and XGBoost Classifier (XGBC)—to diagnose Parkinson’s using the Oxford Parkinson's Disease Detection Dataset. The XGBoost Classifier showed the best performance, with an accuracy of 96.7%, precision of 0.8571, recall of 1.00, and an F1-Score of 0.923.

Machine Learning Aided Breast Cancer Classification

Breast cancer, a significant concern for women, often begins with benign lumps that can turn malignant over time. Types of breast cancer include Lobular, Ductal, Paget disease, Phyllodes tumor, and Angiosarcoma. Early detection is vital, and ML/AI can enhance this process. This research implements eight ML algorithms—SVM, KNN, LR, DTC, RFC, XGBC, GBC, and NB—on a dataset following AMA guidelines to automate breast cancer detection. Performance metrics used are accuracy, precision, recall, F1-score, AUC-ROC, and AUC-PR curves. XGBoost and Gradient Boosting Classifiers excelled, achieving 98.84% accuracy, 0.9688 precision, 1.00 recall, and a 0.9841 F1-score.

Deep Learning based prediction and monitoring of Parkinson's Disease using Voice Data

Detecting Parkinson's disease (PD) early is crucial for treatment. This study uses deep learning to predict and monitor PD via voice data and UPDRS scores. PD affects the nervous system, causing tremors, stiffness, and speech issues. Early detection through speech recognition is vital. Using two datasets (5.3 MB and 1.01 MB) with acoustic features (jitter, shimmer, intensity, pitch) and UPDRS scores, the proposed Residual Neural Network (ResNet) outperformed ten other models. It achieved 98% accuracy, 0.98 precision, and a 0.98 F1-Score with data from 64 healthy and 188 early PD patients.