الفهرس 
Title : Analyzing Attributed Social Networks
AbstractOnly 14 pages are availabe for public view
 |  | from | 101 |  |  |
| | | from | 101 | | |
Abstract
Graph models are widely used to model different types of data. Graph model is a semi-structured model where data are represented as a set of vertices and a set of edges. Analyzing and exploring graph characteristics have gained recent attention in the literature. This thesis focuses on discovering groups of vertices that share a special phenomena together such as a community of linked users in social networks which are highly interacting. Discovering such communities can be done by either community detection or community search. Community detection is to find the groups of vertices that are highly connected to each others and weakly connected to outside vertices. Due to the changes in the graph structure and other features such as link weight, and vertex attributes which led to more complex graphs, another task emerged which is community search. Community search is to discover communities given a set of query parameters where the community vertices share the same attributes. The main contribution of the thesis is to develop efficient and effective community search algorithms to discover top-r weighted communities where the top-r-weighted is considered as the search query parameter.