Results 1  10
of
11
Visual analysis of large heterogeneous social networks by semantic and structural abstraction
 IEEE Transactions on Visualization and Computer Graphics
"... ..."
Learning to rank typed graph walks: Local and global approaches
 In WebKDD/KDDSNA workshop. Einat Minkov
, 2007
"... We consider the setting of lazy random graph walks over directed graphs, where entities are represented as nodes and typed edges represent the relations between them. This framework has been used in a variety of problems to derive an extended measure of entity similarity. In this paper we contrast t ..."
Abstract

Cited by 15 (6 self)
 Add to MetaCart
(Show Context)
We consider the setting of lazy random graph walks over directed graphs, where entities are represented as nodes and typed edges represent the relations between them. This framework has been used in a variety of problems to derive an extended measure of entity similarity. In this paper we contrast two different approaches for applying supervised learning in this framework to improve graph walk performance: a gradient descent algorithm that tunes the transition probabilities of the graph, and a reranking approach that uses features describing global properties of the traversed paths. An empirical evaluation on a set of tasks from the domain of personal information management and multiple corpora show that reranking performance is usually superior to the local gradient descent algorithm, and that the methods often yield best results when combined.
Using Ontological Information to Accelerate PathFinding in Large Semantic Graphs: A Probabilistic Approach
"... Many realworld graphs contain semantics. That is, they represent meaningful entities and relationships as vertices and directed edges, respectively. Moreover, such graphs (called semantic graphs) have meaningful types associated with their vertices and edges. These types produce an ontology graph, ..."
Abstract

Cited by 2 (0 self)
 Add to MetaCart
Many realworld graphs contain semantics. That is, they represent meaningful entities and relationships as vertices and directed edges, respectively. Moreover, such graphs (called semantic graphs) have meaningful types associated with their vertices and edges. These types produce an ontology graph, which specifies the types of vertices that may be connected via a given edge type. Pathfinding in large realworld semantic graphs can be a nontrivial task since such graphs typically exhibit smallworld properties. In this paper, we use ontological information, probability theory, and heuristic search algorithms to reduce and prioritize the search space between a source vertex and a destination vertex. Specifically, we introduce two probabilistic heuristics that utilize a semantic graph’s ontological information. We embed our heuristics into A * and compare their performances to breadthfirst search and A * with a simple nonprobabilistic heuristic. We test our heuristics on both unidirectional and bidirectional search algorithms. Our experimental results on two realworld semantic graphs illustrate the merits of our approach.
The Words of Warcraft: Relational text analysis of quests in an MMORPG
"... As massively multiplayer online games and virtual worlds grow in popularity, the investigation of culture in synthetic environments has become of interest. One representation of culture in games is the narrative provided in MMORPGs ’ quest sets: tasks given to players that provide a view of the arti ..."
Abstract

Cited by 2 (0 self)
 Add to MetaCart
(Show Context)
As massively multiplayer online games and virtual worlds grow in popularity, the investigation of culture in synthetic environments has become of interest. One representation of culture in games is the narrative provided in MMORPGs ’ quest sets: tasks given to players that provide a view of the artificial cultures... Researchers have used specific quests to advance arguments about game cultures. We expand on this previous work by examining the complete quest set for the MMORPG World of Warcraft, subdividing it into three corpora: two for the quests intended for players in one of the two ingame factions, one for those that can be completed by members of either faction. In order to determine salient cultural and narrative elements highlighted in the text, we performed relational text analysis on these corpora, looking for shared textual relationships. We find that while all three corpora possess diverse and unique terms the only relationships present in the corpora at least 5 % of the time are those emphasizing the relationships between players, enemies, and quest giving computercontrolled characters. From this we suggest that it’s important that game designers leverage extraquest data if they want to get messages about the game world to the player.
unknown title
"... How much reference resolution matters for entity extraction, relation extraction, and social network analysis ..."
Abstract
 Add to MetaCart
(Show Context)
How much reference resolution matters for entity extraction, relation extraction, and social network analysis
(Dr. Peter F. PatelSchneider) Second Supervisor
, 2008
"... When intelligence analysts are required to understand a complex uncertain situation, one of the techniques they use most often is to simply draw a diagram of the situation. The diagrams, also called attributed relational graphs or semantic graphs, generally capture the meaning about the situation in ..."
Abstract
 Add to MetaCart
(Show Context)
When intelligence analysts are required to understand a complex uncertain situation, one of the techniques they use most often is to simply draw a diagram of the situation. The diagrams, also called attributed relational graphs or semantic graphs, generally capture the meaning about the situation in their nodes and edges, where the nodes represent concepts/entities and the edges represent the relations/connectivity between the nodes. An important research problem in the area of semantic knowledge discovery and pattern analysis is to identify common/uncommon patterns and instances on these diagrams. Finding patterns and anomalies in data has important applications in intelligence analysis domains such as crime detection and homeland security. The intelligence community’s focus over many years on improving intelligence collection has come at the cost of improving intelligence analysis. The problem today is often not a lack of information, but instead, information overload. Analysts lack tools to locate the relatively few bits of relevant information and tools to support reasoning over that information. Graphbased algorithms can help intelligence analysts solve this problem by sifting through a large amount of data to find the small subset that is indicative of suspicious or abnormal activity.
Visual Hints for Semantic Graph Exploration
, 2009
"... Semanticgraphs contain typed nodes and typed links between nodes. They impose greater demands on a visualization system as they can represent more than one graph and contain a great amount of information as attributes of graph entities. Semantic graphs need stronger tools that combine statistical a ..."
Abstract
 Add to MetaCart
(Show Context)
Semanticgraphs contain typed nodes and typed links between nodes. They impose greater demands on a visualization system as they can represent more than one graph and contain a great amount of information as attributes of graph entities. Semantic graphs need stronger tools that combine statistical and topological analysis and provide the link to the correct information context whenever possible. In this work we present visual hints as a method to reveal semantic information and assist both navigation and exploration of semantic graphs. Visual hints are defined by specific queries on the elements of the graph or their data. We define three types of visual hints: topological, statistical and contextual and show how these are used effectively
ACKNOWLEDGEMENTS
"... I would like to thank my supervising professor, Dr. Lawrence Holder, whose continued support and encouragement made this possible. Throughout every phase of this research, he challenged me to the best of my ability. His guidance and mentoring during my entire graduate studies have been an inspiratio ..."
Abstract
 Add to MetaCart
I would like to thank my supervising professor, Dr. Lawrence Holder, whose continued support and encouragement made this possible. Throughout every phase of this research, he challenged me to the best of my ability. His guidance and mentoring during my entire graduate studies have been an inspiration to me, for which I hope some day to be able to instill in other students. I would like to express my gratitude to Dr. Lynn Peterson, who supervised my Master’s work and not only encouraged me to pursue my PhD, but was always a source of encouragement throughout the graduate school process. I would also like to thank the rest of the committee, Dr. Diane Cook, Dr. Sharma Chakravarthy and Dr. Gautam Das, whose input and encouragement were instrumental in the completion of this work. I would like to thank my parents who have always encouraged me in my academic endeavors. Their examples of scholarly pursuits have always been an inspiration to me. I would also like to thank the rest of my family and friends for their encouragement during this process. And last, but not least, I would like to thank my beautiful wife Robin. Without her love and encouragement through what has been a tough three years, this work and all of my graduate studies would not have been possible. ii
Distributed BreadthFirst Search with 2D Partitioning ∗
"... Many emerging largescale data science applications require searching large graphs distributed across multiple memories and processors. This paper presents a scalable implementation of distributed breadthfirst search (BFS) which has been applied to graphs with over one billion vertices. The main co ..."
Abstract
 Add to MetaCart
Many emerging largescale data science applications require searching large graphs distributed across multiple memories and processors. This paper presents a scalable implementation of distributed breadthfirst search (BFS) which has been applied to graphs with over one billion vertices. The main contribution of this paper is to compare a 2D (edge) partitioning of the graph to the more common 1D (vertex) partitioning. For Poisson random graphs which have low diameter like many realistic information network data, we determine when one type of partitioning is advantageous over the other. Also for Poisson random graphs, we show that memory use is scalable. The experimental tests use a levelsynchronized BFS algorithm running on a large Linux cluster and BlueGene/L. On the latter machine, the timing is related to the number of synchronization steps in the algorithm. 1
Paper
"... The prediction of terrorist threat on the basis of semantic association acquisition and complex network evolution ..."
Abstract
 Add to MetaCart
The prediction of terrorist threat on the basis of semantic association acquisition and complex network evolution