A proposed onto-Apriori algorithm to mine frequent patterns of high quality seafood

Apriori algorithm is a classic data mining algorithm used to generate association rules. This paper proposes onto-Apriori algorithm to generate rules from the ontology data, which is used to determine frequent patterns of quality seafood. In the proposed algorithm, a vertical data layout is used and the generation of candidate set in every step is avoided. The seafood data is represented in ontology and the algorithm is used to mine the ontological data. The proposed algorithm eliminate the costly candidate generation. The dataset obtained from a seafood organization is used to test our approach. The performance of the proposed onto-Apriori algorithm is compared with the existing Apriori algorithm. The results show that the number of transactions required in generating the support count and the candidate items are less when onto-Apriori algorithm is used. The number of times required to check the data to generate candidate 1 item sets is same for both the approaches. Later in onto-Apriori algorithm frequent 1 item set is used as a reference to generate the subsequent candidate item sets. The association rules and patterns obtained by the proposed algorithm are effectively used to discover unknown relationships among seafood.