Εξόρυξη γνώσης σε κοινωνικά δίκτυα
View/ Open
Subject
Δίκτυα πληροφόρησης -- Κοινωνικές απόψεις ; Κοινωνικά δίκτυα ; Online social networks ; Data mining -- Methodology ; Εξόρυξη δεδομένωνAbstract
Since the conception of the Internet, WWW has become the most well-known sharing information network. Recently, a new class of information networks has gained tremendous popularity and now rivals the traditional WWW in terms of usability. Social networks like MySpace (over 125 million users), Facebook (over 800 million users), Orkut (over 100 million users), and LinkedIn (over 110 million “professionals”) are examples of wildly popular networks used to find and manage contacts. Other social networks such as Flickr, YouTube, and Google Video provide multimedia content sharing; on the other hand networks such as LiveJournal and BlogSpot are used to share blogs. The study and analysis of social networks provide us with important information about the network structure, its associated properties and issues related with trust and information distribution. The main problem we face with this study is to survey and analyze various mining techniques that can be used to draw useful conclusions based on the data of the social networks. This dissertation consists of two parts: a) the theoretical one which contains a detailed survey of data mining techniques (community detection, classification nodes, link prediction, finding experts, trust, etc.) and b) the practical part which presents an effort to draw conclusions from a social network database using the data mining techniques provided by SQL Server Analysis Services.