Introduction?Graph Analytics is also known as Network Analytics. It is a very important part of data analysis to codify and visualize relationships that exist between objects, people, devices or nodes in a network. We will introduce the background of Graph Analytics, four widely used types of graph analytics include path analysis, connectivity analysis, community analysis and centrality analysis. We will also discuss about the application of graph analytics and how it might be in the future. The aim of this paper is to show the importance of graph analytics, how it works and the connection between graph analytics and other analytical methods. Background:Graph analytics was built on the mathematics of graph theory. Its history can be traced to 18th century, when Swiss mathematician solve the Konigsberg Bridge Problem (whether a route existed that would traverse each of the seven bridges exactly once. In demonstrating that the answer is no, he laid the foundation for graph theory). As used in graph analytics, the term “graph” does not refer to data charts like bar graphs or line graphs. It actually refers to a set of vertices (points, people, objects or nodes) and of lines (called edges) that connect the vertices. When any two vertices are joined by more than one edges, the graph is called a multigraph. A graph without loops and with at most one edge between any two vertices is called a simple graph. When each vertex is connected by an edge to every other vertex, the graph is called a complete graph. When appropriate, a direction may be assigned to each edge to produce what is known as a directed graph, or digraph. By applying those basic methods in graph theory on data, graph analytics develops as graph theory become more systematic. Graph analytics deals fantastically with structured data (nodes) and there are different methods for different types of problems.Method:There are often three way to do the path analysis, which usually use SPSS software to calculate and the d3.js for visualization(SunbrustPartition function). The first is User Typical Path Identification and User Characteristic Analysis, which is often used in user profiling are some demographic data such as gender, geography, or operational data such as order price, number of orders, etc. User access path data opens up another door for us to understand user characteristics. For example, an application for uploading and sharing pictures can be divided into an authoring user who likes to make uploads, an interactive user who likes to like comments, a diving user who silently browses pictures, As well as consumer users who will never download pictures. And the second is Product Design Optimization and Improvement. Path analysis is very helpful to the optimization and improvement of product design. It can be used to monitor and optimize the conversion rate of each module in the expected user path, and can also find out some obscure function points. In a video-sharing sharing app, users often perform a series of editing operations from the time they shoot the video to the final release of the video. Through path analysis, we can clearly see which users are familiar with and love Editing tools, which operations are too long and tedious, this can help us to improve the editing module targeted to optimize the user experience. If the number of users’ authoring in the path analysis process is closely related to the user’s praise, comment and sharing behavior, we can consider enhancing the sociality of this App and enhancing the user’s stickiness and creative desire. (http://blog.csdn.net/qq_39422642/article/details/78782094) The third is Product operation process monitoring. The conversion rate of product key modules is a very important product operation index in itself. Monitoring and verifying the corresponding operating activity results through path analysis can help stakeholders to understand and understand the effects of operational activities.Connectivity analysis can be considered as a algorithm to test whether two points are related. And its algorithm are usually achieved by code in Java or C++. There are some mature algorithm called find and union. Find and union are for the group, which is a set of non-repetitive elements. The simplest representation of a group is that each group of elements are given a specific mark, if the mark is the same, it means that they belong to the same collection, if the mark is different, you can put one of the mark into the other mark, it is simple union operation. When we use quick-find to solve the problem of dynamic connectivity, at least to call N-1 times union (). A union () algorithm requires at least N + 3 accesses to the array. So the total number of times we visited the array was (N-1) (N + 3) times, which is the square level. Absolutely impossible to solve large-scale problems. Therefore, we should consider the optimization of the union () algorithm. By the basic quick union, the size of the two trees is not taken into account. If a tree is large, it is treated as a subtree of another very small tree when combined, and the arrangement is obviously not reasonable, because the combination of two trees will become a very deformity of the tree. Thus, there may be more and more fast and efficient algorithm presented in the future and it is progressing.Community analysis:This is the kind of analysis in which we try to figure out the relationship or pattern that a community follows and the kind of things they talk about, there are likes and dislikes. Sometimes the community can be predictive because the kind of people that form have similar interest. For example: analyses of tweets can give an idea of what is trending among people. It can also help to decide the rating of a movie or an active TV Show. It is the Density of the people interested in a similar kind of thing which gives a picture about what the public demands.For example: ice bucket challenge was started by a small group, but it ended up around the world, the trend can be predictive sometimes and community analysis is all about that. Centrality analysis:This enables to identify relevance to figure out the most influential person or the most viewed page on website. It tells us the connection of one node to another and how many links are present, in many social setting people with more connections tends to have more power and more visibility. It is also regarded as a measure of how long it will take to spread information from one point to all other nodes in a network.Sometimes one person has to spread its information everywhere, finding a path for every person is hard. Centrality provides a easy and shortest way to connect to all the people in a community. It is the highest occurring node in a network.For example: Google’s PageRanker is used to rank the website in a search engine. It basically counts the number and quality of links to a page in order to determine how important the website it. Applications of graph analytics:- There are many applications in which graph analytics plays an important role. GPS(Global Positioning System) use for route optimization:- Nowadays, many people use GPS system of google maps to know the path between two locations. It is undeniable that GPS system makes our driving experience so easy and effective. It becomes very easy to know the shortest path between two locations by using services like GPS system. Graph analytics is being used to know the shortest path between two locations(nodes). There are many algorithms like Dijkstra’s is being used to know the shortest path between two places(nodes). Whenever we search for the location in Google maps it gives the shortest and convenient path to reach to the destination. It has been done by using path analysis which is one of the widely used types of graph analytics.Biological System:-Graph analytics can play a significant role in the biological system. Graph analytics can provide more accurate and effective treatments to the patients by doing an analysis of relationships in DNA, organs, chemical pathways, and cells of a human body. In addition to this, it can be possible to know how human health gets affected by their lifestyle and medical treatment. Graph analysis can be useful to discover the future diseases of the human body as well so one can easily take cure of the future disease.Graph analytics is very useful to find the connection between two or more objects. For instance, the life sciences organization in disease research pairs proteins along with medications as well as chemical pathways.Financial services:-Graph analytics can be used to stop cyber attacks which could take place in financial industries. Graph analytics is very useful to do a comparison of financial data with geographic, social and another kind of data to analyze patterns between diverse data sets that signals cyber attacks.Identifying communities for security:-Graph analytics can be useful to determine communities that come with a certain theme. For instance, the FBI could be interested to analyze a group of people who are communicating with each other about making of Bomb.Moreover, we all use graph analytics in our day to day life. For instance, Linked In and Facebook which helps to make connections which are based on relationships detected by graph analysis. Furthermore, it must be accepted that graph analytics can play a most important role to make sense of massive unstructured data which is being generated through social media.Reference:- https://www.infoworld.com/article/2877489/big-data/how-graph-analytics-delivers-deeper-understanding-of-complex-data.htmlhttp://www.ibmbigdatahub.com/blog/what-graph-analyticshttps://en.wikipedia.org/wiki/Shortest_path_problem#DefinitionReferencehttp://blog.csdn.net/qq_39422642/article/details/78782094http://blog.csdn.net/qq_29169749/article/details/50899788http://blog.csdn.net/D07Qs2KxkH0KkSxEx/article/details/79043173Referanceshttp://pr.efactory.de/e-pagerank-algorithm.shtmlhttps://en.wikipedia.org/wiki/PageRankhttp://www.ibmbigdatahub.com/blog/what-graph-analyticshttp://ieeexplore.ieee.org/stamp/stamp.jsp?tp==6863132http://www2.unb.ca/~ddu/6634/Lecture_notes/Lecture_4_centrality_measure.pdf