A superkey SK is a subset of attributes of a relation schema R, such that for any two distinct tuples t1 and t2 in relation state r of R, we have t1 [SK] ≠ t2 [SK]. For example, consider a relation schema BOOK with three attributes ISBN, Book_title, and Category. The value of ISBN is unique for each tuple; hence, {ISBN } is a superkey. In addition, the combination of all the attributes, that is, {ISBN, Book_title, Category} is a default superkey for this relation schema.
Generally, all the attributes of a superkey are not required to identify each tuple uniquely in a relation. Instead, only a subset of attributes of the superkey is sufficient to uniquely identify each tuple. Further, if any attribute is removed from this subset, the remaining set of attributes can no longer serve as a superkey. Such a minimal set of attributes, say K, is a candidate key (also known as irreducible superkey). For example, the superkey {ISBN, Book_title, Category} is not a candidate key, since its subset {ISBN} is a minimal set of attributes that uniquely identify each tuple of BOOK relation. So, ISBN is a candidate key as well as superkey. Hence, it is concluded that all candidate keys are superkeys, whereas all superkeys are not candidate keys.
First read the answer fully, then try to explain it in your own words. After that, open a few related questions and compare the concepts. This method helps you remember the topic for a longer time and improves exam preparation.