Cloud computing, a new and promising distributed computing technology, offers a pay-per-use foundation for large-scale scientific and business process applications. With the advancement of cloud computing, artificial intelligence, and big data, demands for cloud resources, particularly certain unclear and emergent resource demands, are fast increasing. Classical cloud resource allocation approaches do not assist the emergent mode in terms of ensuring resource allocation timeliness and optimization. The proposed method employs an Adaptive Neuro-Fuzzy Inference System Quality of Service Aware Genetic Algorithm (ANFIS-QoSGA) method for quickly determining the best VM for each job. This paper proposed an improved resource allocation optimization technique that takes into account the goals of reducing deployment costs, balancing the load, and increasing the Quality of Service performance. Cloud customers' main challenge is deciding the resources to use for the deployment of their applications without negotiating the Quality of Service requirements. The proposed algorithm takes into account 'n' Number of customers Quality of Service needs and resources allocated inside the limited cost constraints.The proposed approach's major goal is to reduce task computational time, cost, and energy consumption while maximizing meaningful resource use. Extensive studies show that the suggested method outperforms other similar scheduling techniques in terms of energy cost and has a better outcome in terms of total execution time which is reduced by 4%, 8%, and 11% when compared with RAA-PI-NSGAII, SFWOA,NSGA-III and makespan, degree of imbalance, and security value under high load conditions.