(IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 10, No. 4, 2019 Techniques, Tools and Applications of Graph Analytic Faiza Ameer 1 , Muhammad Kashif Hanif 2 , Ramzan Talib 3 , Muhammad Umer Sarwar 4 , Zahid Khan 5 , Khawar Zulfiqar 6 , Ahmad Riasat 7 Department of Computer Science, Government College University, Faisalabad, Pakistan Abstract—Graphs have acute significance because of poly- tropic nature and have wide spread real world big data appli- cations, e.g., search engines, social media, knowledge discovery, network systems, etc. Major challenge is to develop efficient systems to store, process and analyze large graphs generated by these applications. Graph analytic is important research area in big data graphs dealing with efficient extraction of useful knowledge and interesting patterns from rapidly growing big data streams. Tremendously huge and complex data of graph applications requires specially designed graph databases having special data structures and effective features for data modeling and querying. The manipulation of large size of data requires effective scalable and distributed computational techniques for efficient graph partitioning and communication. Researchers have proposed different analytical techniques, storage structures, and processing models. This study provides insight of different graph analytical techniques and compares existing graph storage and computational technologies. This work also assesses the perfor- mance, strengths and limitations of various graph databases and processing models. Keywords—Graph; graph analytic; big data; graph tools; ana- lytical techniques I. I NTRODUCTION Graphs have astute magnitude due to their versatile and expressive nature. Real world big data problems like weather forecasting, geographical changes, large network systems, social networks, semantic search and knowledge discovery, text mining, IOT, cyber security, etc. all can be viewed and modeled as graphs. Graphs can be used to represent and analyze big data. Big data is huge volume, high velocity and a large variety of information asset that demands cost effective, and innovative forms of information processing for enhanced insight and decision making [1]. The term huge volume refers to the large size of static or continuously growing data like Facebook, Twitter, Google, etc. High Velocity represents the required speed of data generation, and analysis. Large variety means the use of various types of structured (e.g., data from relational databases), semi-structured (e.g., XML and JSON documents), and unstructured data (e.g., video, audio, images etc.) [1]. Graph analytic is based on graph theory (a branch of Mathematics). Graph theory was born out of a very practical urban planning problem. The problem started in Konigsberg (old city in Russia). The city had two large islands, connected by seven bridges. Back in 1736, the problem was to device a walkway from one part of city to another by traversing all seven bridges only once. A mathematician, named Euler, proved mathematically that this problem had no solution due to odd number of bridges. He reformulated the problem and solved it by eliminating all features except land masses (termed as vertex) and the connecting bridges (termed as edge). The resulting mathematical structure was called a graph (Fig. 1). By solving this problem, Euler laid down the foundations of whole field of graph theory [2]. Different operations which can be performed on a graph are add/remove a vertex, add/remove an edge, or find the nearest neighbors (i.e., finding nodes connected to the vertex), etc. Graph analytic models large and complex data problems as a set of graphs. It expresses relationship patterns of objects by exploiting the mathematical properties of data and statistical modeling techniques to provide efficient algorithmic solutions and discover meaningful patterns [2]. Many organizations are competing with their peers in market using graph analytic by making accurate and timely decisions [2]. There exists variant techniques for graph analytic like path analytic,connectivity analytic, centrality analytic, and community analytic based on solution to different types of problems [2]. Each one of them use different principles and methods to answer divergent analytical questions. To handle large graphs, new systems incorporate efficient storage and processing technologies. The new database technolo- gies fulfill the growing requirements of current applications and cover the limitations of traditional database models. Modern graph related database management system includes graph databases and graph stores [3]. These databases provide general features for data storage, data modeling, and support for graph queries and query languages. Use of graphs in big data have also generated much interest in the field of large-scale graph data processing. Modern parallel processing systems are based on four different processing models: MPI-Like, Map-Reduce, BSP, and Vertex-Centric Graph Processing. Pregel, Giraph, GPS, Mizan, and GraphLab are important parallel processing models. The remainder of the paper is organized in different sections. Section II gives a brief overview of the applications of graphs. Section III describes different graph analytic techniques. In Section IV, graph storage techniques are discussed. Sections V provides different processing models for large graphs. At the end, we conclude the outcomes. www.ijacsa.thesai.org 354 | Page