International Journal of Soft Computing and Engineering (IJSCE) ISSN: 2231-2307, Volume-1, Issue-3, July 2011 19 Abstract — Graphs become increasingly important in modeling complicated structures, such as circuits, images, chemical compounds, protein structures, biological networks, social networks, the web, workflows, and XML documents. Many graph search algorithms have been developed in chemical informatics, computer vision, video indexing and text retrieval with the increasing demand on the analysis of large amounts of structured data; graph mining has become an active and important theme in data mining. Index Termes: Subgraphs, Graph Mining, gSpan I. INTRODUCTION In mathematics, computer science and related subjects an algorithm is an effective method for solving a problem expressed as a finite sequence of instructions. Algorithms are used for calculation data processing and many other fields. Meaning1. An algorithm operating on data that represent continuous quantities, even enough this data is represented by discrete approximation such algorithm are studied in numerical analysis. Meaning2. An algorithm in the form of different equations that operates continuous on the data running an analog computer. A. Apriori-Based Approach Apriori based frequent substructure mining algorithm share similar characteristics with Apriori-based frequent item set mining algorithms. The search for frequent groups starts with graphs a small “size” and proceeds in a bottom-up manner by generating candidate having an extra vertex, edge or path. The definition of graph size depends on algorithm used. B. Pattern-Growth Approach The Apriori-based approach has to use the breadth-first search (BFS) strategy because of its level-wise candidate generation. II. SURVEY OF TECHNIQUES AND ALGORITHMS Various algorithms on graph mining were developed by many researchers. Some of them are reviewed in this section. Ullmann [1] in 1976 developed an algorithm for subgraph isomorphism. Subgraph isomorphism determined by means Manuscript received June 4, 2011. Vijender Singh, Department of Computer Science and Engineering, Thapar University, Patiala (Punjab), India, Mobile No. +91-9255074702, (e-mail: vijender_bhar@hotmail.com ). Dr. Deepak Garg, Professor, Department of Computer Science and Engineering, Thapar University, Patiala (Punjab), India (e-mail: deep108@yahoo.com ). of a brute-force tree search procedure. This algorithm attains efficiency by inferentially eliminating successor’s nodes in the tree search. Agarwal and Srikant [2] in 1994 considered the problem of discovering association rules between items in a large database of sales transaction. They presented two new algorithms for solving this problem that are fundamentally different from the known algorithm. Cook and Holder [14] in 1994 discovered a new version of their SUBDUE substructure discovery system is based on minimum description length principle. Holder, Cook and Djoko [3] in 1994 described the SUBDUE system which the minimum description length (MDL) principle is discovered substructures that compress the database and represent structural concepts in the data. In this paper they described the application of SUBDUE and also discussed the minimum description length principle and background knowledge used by SUBDUE can guide substructure discovery in a variety of domain. Blockeel and Raedt [6] in 1998 introduced a first-order framework for top-down induction of logical decision tree. Top-down induction of decision trees is the best known and most successful machine learning technique. It has been used solve numerous practical problems. It employs a divide-and conquers strategy, and in this it differs from its rule-based competitors which are based on covering strategies. Chakrabarti, Dom and Indyk [7] in 1998 developed a new method for automatically classifying hypertext into a given topic hierarchy, using an iterative relaxation algorithm. After bootstrapping off a text-based classifier, they used both local texts in a document as well as the distribution of the estimated classes of other documents in its neighborhood, to refine the class distribution of document being classified. They discussed three area of research: text and hypertext information retrieval, machine learning in context other text or hypertext, and computer vision and pattern recognition. Inokuchi,Washio and Motoda [9] in 1998 proposed a novel approach name AGM to efficiently mine the association rule among the frequently appearing substructure in a given graph dataset. A graph is represented by adjacency matrices and the frequent patterns appearing in the matrices are mined through the extended algorithm of the basket analysis. Calders and Wisen [10] in 2001 Presented on monotone data mining layer a simple data-mining logic (DML) that can express common data mining tasks like “find Boolean association rules” or “Find inclusion dependencies”. Kramer, Raedt, and Helma [11] in 2001 presented the application of feature mining techniques to the developmental therapeutics program’s AIDS antiviral screen database. Kuramochi and Karypis [12] in 2001 presented a computationally efficient algorithm for finding all frequent subgraphs in large graph databases. They evaluated the performance of the algorithm by experiments with synthetic datasets as Survey of Finding Frequent Patterns in Graph Mining: Algorithms and Techniques Vijender Singh, Deepak Garg