Iron deficiency is the most known form of nutritional deficiency. It is most common in undernourishment and is most prevalent in young children, women of childbearing age, and pregnant women. Iron deficiency in children causes developmental delays and behavioral disruptions, and in pregnant women it raises the risk of premature labor and delivery of a baby with low birth weight. There was an awareness in the past three decades, about the intake of iron supplement for infants that was resulted due to this childhood iron-deficiency anemia all over the world. Even though, it is always better to detect the disease at an earlier stage of life to prevent further harmful effects and to devise proper treatment. In this study, the anemia is taken into consideration for early prediction and diagnosis of the disease by using the data mining technique to analyze the data. In healthcare organizations the volume of data is more. To get knowledge from those data we need an efficient technique. Data mining is used for the purpose of discovering knowledge from vast amount of database. To classify the stages of anemia, classification technique which is one of the data mining technique is used. The data is collected from 200 household of students from Public Health College in Jazan University. The research work is done with WEKA open-source software under Windows7 environment. An experimental study is carried out using data mining techniques such as J48, Random Forest tree and hoeffding tree. As a result, the performance is evaluated for three classification techniques and their accuracy compared through confusion matrix. It has been concluded that Random Forest tree gives better accuracy than the J48 and Hoeffding tree technique.

Anemia Data Mining Classification Technique J48 Random Forest tree and Hoeffding Tree