The education domain offers a fertile ground for many interesting and challenging data mining applications. These applications can help both educators and students, and improve the quality of education. In this paper, we present a real-life application for the Gifted Education Programme (GEP) of the Ministry of Education (MOE) in Singapore. The application involves many data mining tasks. This paper focuses only on one task, namely, selecting students for remedial classes. Traditionally, a cut-off mark for each subject is used to select the weak students. That is, those students whose scores in a subject fall below the cut-off mark for the subject are advised to take further classes in the subject. In this paper, we show that this traditional method requires too many students to take part in the remedial classes. This not only increases the teaching load of the teachers, but also gives unnecessary burdens to students, which is particularly undesirable in our case because the GEP students are generally taking more subjects than non-GEP students, and the GEP students are encouraged to have more time to explore advanced topics. With the help of data mining, we are able to select the targeted students much more precisely.

In many data mining applications, the objective is to select data cases of a target class.