Frequent pattern mining, known as Apriori algorithm is a simple yet powerful mining algorithm that can be applied in databases to find the frequent itemsets. This paper implements the Apriori algorithm for predicting flood in the study area in Terengganu which contains 7 districts, each has attributes and instances that need to test. The attributes which are flood area, water level and flood status are supplied by the Malaysian Drainage and Irrigation Department from 2009 until 2014, for the month if November until March. The best rules generated from the datasets using Data Mining tools (WEKA) is then indicated to develop a new flood area prediction system. This research aims to help people especially in risky areas to get early warnings and save life and valuable belongings.