Ping Jiang, Jiejie Chen, and Haixia Li
Genetic algorithm (GA), machine learning, big data analytics, model selection, model optimisation
In the era of big data, optimising analysis models to improve the efficiency and accuracy of data processing is crucial. This study innovatively proposes combining genetic algorithms (GAs) and machine learning to select and optimise big data analysis models. By constructing an initial population containing 10 different big data analysis models, using the crossover, mutation, and selection operations of GAs, optimising model parameters, and using machine learning algorithms to evaluate performance, the experiment has achieved remarkable results: the prediction accuracy in data set A is improved by 5%, while the training time in data set B is reduced by 15%. In addition, this study also tested the unbalanced data set C. By optimising the evaluation criteria, the classification accuracy was significantly improved, with the G-Mean value reaching 0.89 and the F-value of the subcategory increasing to 0.92. By integrating GAs with machine learning, we have greatly improved the optimisation efficiency of big data analysis models, enhanced their generalisation ability and prediction accuracy, and demonstrated significant practical value and cutting-edge innovation for large- scale dataset processing. This innovative solution provides a solid and efficient answer to the analysis of large amounts of data, demonstrating its profound practical significance.
Important Links:
Go Back
