Emergent Semantics

Peer-to-peer systems are evolving with new information-system architectures, leading to the idea that the principles of decentralization and self-organization will offer new approaches in informatics, especially for systems that scale with the number of users or for which central authorities do not prevail. This book describes a new way of building global agreements (semantic interoperability) based only on decentralized, self-organizing interactions.

Preface
Chapter 1 Introduction
1.1 On Syntax, Semantics and Syntactic Semantics
1.2 Emergent Semantics
in Distributed Information Systems
1.3 Scope of the Research
1.4 What this Book is not about
1.5 Outline
1.6 Contributions
Chapter 2 On Integrating Data
in the Internet Era
2.1 Federated Databases
2.2 XML, RDF and the Semantic Web
Chapter 3 Peer-to-Peer Information Management
3.1 From unstructured to structured P2P Systems
3.2 Peer Data Management
Chapter 4 Semantic Gossiping
4.1 On Uncertain Schema Mappings
in Decentralized Settings
4.1.1 Mapping Completeness
4.1.2 Mapping Soundness
4.2 The Model
4.2.1 The Data Model
4.2.2 The Network Model
4.3 Overview
4.4 Syntactic Similarity
4.5 Semantic Similarity
4.5.1 Cycle Analysis
4.5.2 Result Analysis
4.6 Gossiping Algorithm
4.7 Case Study
4.8 Related Work
4.9 Conclusions
Chapter 5 Self-Repairing Semantic Networks
5.1 Experimental setup
5.2 Cycle Analysis
5.3 Result Analysis
5.4 Combined Analysis
5.5 Related Work
5.6 Conclusions
Chapter 6 Probabilistic Message Passing
6.1 Introduction
6.2 Problem Definition
6.2.1 An Introductory Example
6.3 Modeling PDMSs as Factor-Graphs
6.3.1 A Quick Reminder on Factor-Graphs
and Message Passing Schemes
6.3.2 On Factor-Graphs
in Undirected PDMSs
6.3.3 On Factor-Graphs in Directed PDMSs
with Containment Mappings
6.4 Embedded Message Passing
6.4.1 On Feedback Variables
in PDMS Factor-Graphs
6.4.2 On Cycles in PDMS Factor-Graphs
6.4.3 Embedded Message Passing Schedules
6.4.4 Prior Belief Updates
6.4.5 Introductory Example Revisited
6.5 Performance Evaluation
6.5.1 Performance Analyses
6.5.2 Performance Evaluation
on Random PDMS Networks
6.5.3 Applying Message Passing
on Real-World Schemas
6.6 Conclusions
Chapter 7 Analyzing Semantic Interoperability in the Large
7.1 Introduction
7.2 The Model
7.2.1 The Peer-to-Peer Model
7.2.2 The Peer-to-Schema Model
7.2.3 The Schema-to-Schema Model
7.3 Semantic Interoperability In the Large
7.3.1 Semantic Connectivity
7.4 A Necessary Condition
for Semantic Interoperability
7.4.1 Undirected Model
7.4.2 Directed Model
7.5 Semantic Component Size
7.6 Weighted Graphs
7.6.1 Connectivity Indicator
7.6.2 Giant Component Size
7.7 Semantic Interoperability
in a Bioinformatic Database Network
7.7.1 The Sequence Retrieval System (SRS)
7.7.2 Graph analysis of an SRS repository
7.7.3 Applying the Heuristics
to the SRS Graph
7.7.4 Generating a Graph with
a given Power-Law Degree Distribution
7.8 Use Case Scenarios
7.9 Conclusions
Chapter 8 GridVine:Building Internet-Scale
Semantic Overlay Networks
8.1 Introduction
8.2 Overview of our Approach
8.2.1 Data Independence
8.2.2 Decentralized Semantics
8.3 The P-Grid P2P System
8.4 Semantic Support
8.4.1 Metadata Storage
8.4.2 Schema Definition And Storage
8.5 Resolving Queries in GridVine
8.5.1 Resolving Atomic Queries
8.5.2 Resolving Conjunctive Queries
8.6 Semantic Interoperability
8.6.1 Schema Inheritance
8.6.2 Semantic Gossiping
8.7 Implementation
8.7.1 Architectural Overview
8.7.2 Querying
8.7.3 Query Reformulation
8.7.4 Experimental Evaluation
8.8 Related Work
8.9 Conclusions
Chapter 9 PicShark: Sharing Semi-Structured Annotations
in the Large
9.1 Introduction.
9.2 Sharing Semi-Structured Metadata
9.2.1 On Semi-Structured Metadata
9.2.2 On the Difficulty
of Sharing Semi-Structured Metadata
9.2.3 Opportunities for Reducing
Metadata Scarcity Collaboratively
9.3 Formal Model
9.3.1 Metadata Entropy
9.4 Recontextualizing Semi-Structured Metadata
9.4.1 Exporting Local Metadata
through Data Indexing
9.4.2 Dealing with Metadata Incompleteness
through Intra-Community Metadata
Imputation
9.4.3 Dealing with Metadata Heterogeneity
through Pairwise Schema Mappings
9.4.4 Dealing with Metadata Incompleteness
through Inter-Community Metadata
Propagation
9.4.5 Possible Answers and User Feedback
9.5 PicShark: Sharing Annotated Pictures
in the Large
9.5.1 Information Extraction in PicShark
9.5.2 Performance Evaluation
9.6 Related Work
9.7 Conclusions
Chapter 10 idMesh: Graph-Based Disambiguation
of Online Identities
10.1 Introduction
10.2 Contributions and Outline
10.3 Related Work
10.4 Problem Definition
10.5 idMesh Constructs
10.6 Making Sense of It
10.6.1 An Introductory Example
10.6.2 Deriving a Factor-Graph
to Retrieve Equivalent Identities
10.6.3 Deriving a Factor-Graph
to Retrieve Up-to-date Identities
10.6.4 Query Answering
10.7 System Perspective
10.7.1 Architectural Overview
10.7.2 Distributed Probabilistic Inference
10.8 Performance Evaluation
10.8.1 Performance of the Inference Network
10.8.2 Scale-Up
10.9 Conclusions
Chapter 11 Conclusions
List of Frequently Used Symbols and Abbreviations
Bibliography