Technical Features

GRACE is based on the principle that a content management system must not replace or unnecessarily enlarge the customer's existing resources, but rather allow the customer to maximize their use.

This is accomplished by two major advantages of the system:

integration of the existing content sources and
the use of the Grid technology

Organizations typically use various document management systems and document repositories, and are not keen with a new one. They may rather face an awkward situation where several different document management systems are installed, or where there are multiple document repositories installed on various locations. With GRACE, organizations can seamlessly integrate these varying and multiple internal content sources, and achieve a single point of access to all of them in parallel.

Moreover, GRACE allows organizations to integrate these internal content sources with additional, external resources, such as web based document repositories, databases and search engines.

Innovative Approach

GRACE introduces an innovative approach to the integration of multiple content sources: it systematically harvests relevant information from these documents, applying very strong natural language processing methods in order to re-index them into Knowledge Domains. Knowledge Domain is not only a complete virtualization of multiple relevant content sources, but also incorporates the underlying semantics encapsulated in related ontologies.

These ontologies are used to:

Query the content sources, and than index them by associating them with the key terminology
This allows GRACE to extract from the extensive content sources only the portion that is indeed relevant for a particular interest
The ontologies are further used for querying, just like a normal index, browsing, and presentation of the search results

Constant Update of the Knowledge Domains

Since GRACE systematically harvests the relevant content sources, it offers a constant update of the Knowledge Domains with new and relevant information. The users do not need to repeatedly query the content sources in order to stay updated: the fresh information is automatically integrated into the existing Knowledge Domain, and made instantly available.

Unprecedented Information Retrieval Functionalities

The utilization of the strong natural language processing methods allows GRACE to offer unprecedented Information Retrieval functionalities to the knowledge workers. These functionalities include

Semantic Indexing
Named Entity Extraction
Automatic Extraction of Relevant Lexical Patterns
Inference Search
Automatic Creation of Semantic Maps

Semantic Indexing indexes documents by systematically detecting relevant concepts. Named Entity Extraction allows detection of critical information, such as personal and organizational names, email addresses, and dates. Extraction of Relevant Lexical Patterns elucidates additional key terms typical for a set of documents. Inference Search allows user to formulate complex queries that require logical reasoning to be applied to the search results. Semantic Maps allow users to browse the search results by their semantic relations.

Multilinguality

GRACE provides multilingual abilities for several Western European languages. Currently multilingual text processing is supported for English, German, Italian, but other languages can be added by integrating suitable lexical databases.
Why GRACE utilizes the Grid Technology ?The application of very strong natural language processing methods for indexing purposes is a computationally intensive task that requires significant resources. This is why GRACE utilizes the Grid Technology. The Grid is an emerging technology that allows organizations to perform more efficiently computationally intensive tasks without increasing their existing resources, but rather maximizing their use. This is accomplished by secure and dynamic sharing of the existing computational and storage resources. GRACE employs the Grid mainly for text normalization and categorization, due to the amount of processing respources required by these tasks.

Use of the Grid by GRACE

A special case of this technology are the so called Data Grids that assume that data consumed by these computations must be shared across an organization, and that it is thus more efficient to keep these data stored in a central location, accessible to various front end applications targeting various organizational information needs.

GRACE utilizes the Data Grid in order to make the Knowledge Domains securely accessible throughout an organization, regardless of the geographical location or the strength of the locally available resources. GRACE complements the Database Federation systems that utilize Grid technology. They also serve as a virtualization of multiple information sources, and allow a seamless single point access to all the relevant information in parallel. Nevertheless, while the Database Federation systems focus on the structured data, GRACE is the first working system that allows integration of the unstructured, textual information.

GRACE Applications

There are many applications that can be built on top of GRACE. They include:

information update alarms
graphic visualization of semantic maps
multilingual applications, etc.

The variety of available functionalities allows the users to design a strong and reliable content management system that is the most suitable to their particular information needs.