Data Mining and Knowledge Engineering

Varsha Namdeo ¹, Ramjeevan Singh Thakur ¹, Ghanshyam Singh Thakur ¹

Affiliations
1 Department of Computer Applications, Maulana Azad National Institute of Technology, Bhopal, IN

Abstract

Classification is one of the most important tasks in data mining. Researchers are focusing on designing classification algorithms to build accurate and efficient classifiers for large data sets. Classification at multiple levels helps in finding more specific and relevant knowledge.
There are several applications, where it is necessary to classify data at different abstraction level due to sparsity of data. This paper presents an approach for classifying multilevel data from simplified Neural Networks. This work is very useful in the applications where data is spread in multilevel hierarchy and required classification of data at all abstraction levels of data.

Keywords

Multilevel Classification, Multilevel Abstraction, Artificial Neural Network, Classification by WEKA.

Full Text

     

Vivek Tiwari ¹, Ramjeevan Singh Thakur ²

Affiliations
1 Maulana Azad National Institute of Technology (MANIT), Bhopal, IN
2 Maulana Azad National Institute of Technology, Bhopal, IN

Abstract

In many database applications, information stored in a database has a built-in hierarchy consisting of multiple levels of concepts. This is also called ontological driven data. The problems of finding frequent item sets are basic in multi level association rule mining and fast algorithms for solving problems are needed. This paper presents an efficient version of FP-Growth algorithm for mining multi-level (FPGM:FP-Growth in Multilevel) association rules in large databases. The proposed method is designed in such a way that it can find maximum frequent itemset at lower level of abstraction in ontological categorical data. The proposed model adopts a top down progressively deepening approach to derive large itemsets. Proposed algorithm works well comparison with general approach of multilevel association rules in term of execution time and throughput. Proposed algorithm is flexible in term of supply different support value for different level. This paper discus some important and crucial issue regarding support value and dataset. There are also some special cases discussed. These cases reveal the behavior of proposed algorithm in different circumstances.An example is also given to demonstrate and support that the proposed mining algorithm can derive the multiple-level association rules under different supports in a simple and effective manner.

Keywords

Multilevel Association Rules, FP-Tree, Ontology Data.

Full Text