Mining Frequent Itemsets from a transaction database is an very important and most widely used task for analyzing data in any business. It is the preliminary step to find the correlation between the items which are called Association Rules. Closed Frequent Itemsets are the compact representation of the Frequent Itemsets which can save memory and time for large, dense data. It is very challenging to mine Closed Frequent Itemsets in uncertain data streams due to concept drifts. This paper presents a detailed survey on various popular algorithms developed for mining closed frequent itemsets from data streams.