社交网络数据分析PPT.ppt

First ,the Basic Analysis about google and twitter. In the google features files ,it clearly show us what kinds of information it inlucde. So we count the number of each feature. * We calculate and analyse several features. For gender , wo suppose that 1 represent male and 2 represent female , 3 represent that the user fill nothing. To my surprise, so many people engaged in photographer. Maybe we can say that it’s the most popular job in foreign countries . The most frequent number User where come from * I think it have a boundary , more than this boundary it will big, less than this boundary the word is small. Regenerate The twitter’s data is not so clear, so we do a roughly statistic about the frequency. 0.Recommend some account for his or her followers every friday 1.Foursquare: Phone service website about user’s Location Information 2.AddThis:A kind of Internet Bookmark, integration website 3.mashable: A Internet news blog 4.GetGlue: a Social recommendation website * The page rank is to show the Connection between pages and pages. Here we use this Algorithm to calculate who is concerned most. Like Sina Big V. The Y axis is the PageRank number. * 1. In?graph theory, a?clustering coefficient?is a measure of the degree to nodes in a graph tend to cluster together. Evidence suggests that in most real-world networks, and in particular?social networks, nodes tend to create tightly knit groups characterised by a relatively high density of ties. There are two kinds definition: the global and the local. 2.The triplet divided two kinds open triplet and closed triplet * This loop list one user’s followers This loop counting the number of the open triplet and closed triplet And then calculate the results * It is show that the Twitter Network is more close. The people of Twitter network The possibility of people know with each other Twitter high than Google. * Running time At the beginning we use the IBM’s cluster. But there are something wrong with the IBM‘s