Canine Disease Prediction using Multi-Directional Intensity Proportional Pattern with Correlated Textural Neural Network

Data optimization is crucial for enhancing prediction accuracy and similarity identification in texture learning systems, especially for predicting canine diseases. Traditional data retrieval methods often struggle with accuracy and efficiency, particularly when dealing with large datasets. This study presents a novel approach combining Multi-Directional Intensity Proportional Pattern (MDIPP) with a Similarity Measure (SM) system to improve data relevance and similarity estimation. The model organizes data into a paged database structure, which speeds up search operations. A neural network, Correlated Textural Neural Network (CTNN), forecasts the relevance of feature attributes and sorts matching indexes to predict canine diseases based on Test data. The CTNN model incorporates a correlation factor among features to enhance prediction accuracy. The relevance of data is determined using an upgraded neural network that accounts for these correlations. The study evaluates performance based on precision, recall, F1-score, and data retrieval accuracy, comparing the results with state-of-the-art techniques. By improving the organization and indexing of data and refining the prediction process, this approach aims to advance data validation and the prediction of canine diseases in large-scale texture learning systems.