Explore the vast range of titles available.

Ornamental fish keeping is the second most preferred hobby in the world and it provides a great opportunity for entrepreneurship development and income generation. Controlling the environment in ornamental fish farm is a considerable challenge because it is affected by a variety of parameters like water temperature, dissolved oxygen, pH, and disease occurrences. One particular interesting ornamental fish species is goldfish ( Carassius auratus ). Machine learning (ML) and deep learning technique have significant potential in analysing voluminous data collected from fish farm. Through this technique, the fish farmers can get insight on feeding behaviour, fish growth patterns, predict diseases/stress, and environmental factors affecting fish health. The aim of the study is to analyze the behavioural changes in goldfish due to alterations in environmental parameters (water temperature and dissolved oxygen). Decision tree, Naïve Bayes classifier, K-nearest neighbour (KNN), and linear discriminant analysis (LDA) were used to analyse the behavioural change data. To compare the performance between all four classifiers, cross validation and confusion matrix used. The cross-validation error of LDA, Naïve Bayes classification, KNN and decision tree was 19.86, 28.08, 30.14 and 13.78 respectively. Decision tree was proved to be the most accurate and effective classifier. Different temperature and DO range were taken to predict fish behaviour. Some findings are, the behaviour of fish was rest between temperature 37.85 °C and 40.535 °C, erratic when temperature was greater than or equal to 40.535 °C, gasping when temperature was between 37.85 and 40.535 °C and when DO concentration was less than 6.58 mg/L. Blood parameter analysis has been done to validate the change in external behaviours with change in physiological parameters.

In data mining, classification is one of the most important steps in predicting the target class. Classification is performed by an improved model in existing work in which feature selection is performed based on the bat optimization method to increase the classification accuracy. And an Enhanced Neural Network is used for classification which includes Intuitive, Interpretable Correlated-Contours fuzzy rules. And an effective model is created based on the extraction of fuzzy rules, where data partitioning is performed via a similarity-based directional component. However, the dataset used for experimentation is noisy as well as incomplete data values. Due to incompleteness, knowledge discovery is obstructed and the result of classification is affected as well. And bat provides very slow convergence and easily falls into local optima. To solve this issue, an improved framework is introduced in which missing value imputation is performed by using k means clustering, and then for feature selection, an improved cuckoo search optimization is used. An enhanced classifier based on fuzzy logic and Alex Net neural network structure (F-ANNS) is used for classification and hybrid Ant Colony Particle Swarm Optimization (HASO) is used for optimizing parameters of the AlexNet neural network classifier. The results show that the proposed work is more effective in precision, recall, accuracy, and f-measure as shown by experimental results.