A Review on Different Techniques for Mining Frequent Patterns in Unordered Trees

Abstract

Numerous fields, such as bioinformatics, web mining, and social network analysis, now require frequent pattern mining in unordered trees. Finding repeating patterns and substructures inside a collection of unordered trees is the key to unlocking this technique's potential to provide important information about the underlying data. This paper gives a thorough overview of various methods for mining frequent patterns in unordered trees, highlighting their advantages, disadvantages, and practical uses.The review starts out by defining the basic terms and concepts related to frequent pattern mining in unordered trees. The discussion then moves on to a number of widely used algorithms in this setting, such as graph-based strategies, bottom-up tree traversal techniques, and depth-first search-based techniques. Each approach is thoroughly explained, including its underlying concepts, computational complexity, and applicability to different kinds of tree datasets.The paper also examines recent developments in the subject, including distributed frameworks and scalable parallel algorithms for mining common patterns in big unordered tree collections. In order to improve the mining process and the calibre of patterns found, it also looks at the incorporation of extra constraints and measurements like weighted support and tree edit distance.The paper also talks about recent developments in the subject, namely scalable parallel algorithms and distributed frameworks for finding common patterns in enormous unordered tree collections. In order to strengthen the mining process and raise the calibre of patterns found, it also looks at the incorporation of extra restrictions and metrics like tree edit distance and weighted support.