Nowadays, advanced database applications manage two main types of queries: regular queries issued by end users and queries issued automatically by systems such as recommendation systems. The main objective of recommended queries is to suggest new interesting items to users for a large pool of items that may include movies, books, friends, articles, optimization structures, etc. From a DBMS perspective, recommended queries represent an important overhead, since it shall optimize them as it does for regular queries. Usually, queries belonging to these two types are active at the same time. As a consequence, multiple queries can be evaluated more efficiently together than independently, because it is often possible to share state and computation. Motivated by this observation, we propose in this paper, a new approach that dynamically optimizes these two types of queries using materialized views. Our approach first optimizes regular queries according their arrival, and when a recommended query arrives it is optimized based on the optimization history of the regular queries by the means of similarity measures. To ensure the scalability of our approach, we use a hyper-graph structure that capture the interaction between queries regardless of their types. Our approach is evaluated using Star Schema Benchmark and the obtained results are deployed in a commercial DBMS.