An Architecture for Linguistic and Semantic Analysis on the ARXMLIV Corpus


Abstract
The ARXMLIV corpus is a remarkable collection of text containing scientific mathematical discourse. With more than half a million documents, it is an ambitious target for large scale linguistic and semantic analysis, requiring a generalized and distributed approach. In this paper we implement an architecture which solves and automates the issues of knowledge representation and knowledge management, providing an abstraction layer for distributed development of semantic analysis tools. Furthermore, we enable document interaction and visualization and present current implementations of semantic tools and follow-up applications using this architecture. We identify five different stages, or purposes, which such architecture needs to address, encapsulating each in an independent module. These stages are determined by the different properties of the document formats used, as well as the state of processing and linguistic enrichment introduced so far. We discuss the need of migration between XML representations and the challenges it would pose on our system, revealing the benefits and trade-off of each format we employ. In the heart of the architecture lies the Semantic Blackboard module. The Semantic Blackboard comprises a system based on a centralized RDF database which can facilitate distributed corpus analysis of arbitrary applications, or analysis modules. This is achieved by providing a document abstraction layer and a mechanism for storing, reusing and communicating results via RDF stand-off annotations deposited in the central database. Achieving a properly encapsulated and automated pipeline from the input corpus document to a semantically enriched output in a state-of-the-art representation is the task of the Preprocessing, Semantic Result and Output Generation modules. Each of them addresses the task of format migration and enhances the document for further semantic enrichment or aggregation. The fifth module, targeting Visualization and Feedback, enables user interaction and display of different stages of processing..
Keywords No keywords specified (fix it)
Categories (categorize this paper)
Options
Edit this record
Mark as duplicate
Export citation
Find it on Scholar
Request removal from index
Translate to english
Revision history

Download options

Our Archive


Upload a copy of this paper     Check publisher's policy     Papers currently archived: 45,328
External links

Setup an account with your affiliations in order to access resources via your University's proxy server
Configure custom proxy (use this if your affiliation does not provide a proxy)
Through your library

References found in this work BETA

No references found.

Add more references

Citations of this work BETA

No citations found.

Add more citations

Similar books and articles

Will the Neural Blackboard Architecture Scale Up to Semantics?Michael G. Dyer - 2006 - Behavioral and Brain Sciences 29 (1):77-78.

Analytics

Added to PP index
2010-12-22

Total views
59 ( #146,666 of 2,280,259 )

Recent downloads (6 months)
2 ( #561,475 of 2,280,259 )

How can I increase my downloads?

Downloads

My notes

Sign in to use this feature