Cloud computing has been embraced all over the world because of its economic viability and capacity to scale and use resources flexibly, which makes it possible for entities to utilize Database-as-a-Service (DBaaS) tools. This research paper offers two approaches for improving the cloud database efficiency when various processes occur simultaneously. To start with, the tree-based data mining approach will assist in establishing the occurrence of particular sub-expressions in predicate queries and view materialization, which is going to lead to avoiding redundant computations and reducing the time needed for query execution. Finally, the M. Parzen window approach will forecast the data access changes and lead to the creation of adaptive cache, which helps to prioritize blocks that are likely to be reused.