Corpus feature list

In short Corpus is a search and structured information engine that provide: XML interface to information queries suitable for most Web and application needs Easy to use administration via the built in Corpus web server Platform independence by using Java 1.4 as platform and HTTP/XML as interface!

The big list

This is one list of functions provided with Corpus ? any suggestions or questions can be posted to: info@illuminet.se

And please do ask us, since we are constantly working on new functions and features to meet your needs!

Queries

To better understand your search needs, we offer language adaptive and context sensitive searching - more sensitive than Google(tm) or Altavista(tm).

• Near phrase matching finds pages with word sequences near the query
• Boolean search with +/- for excluding or exclusively finding pages with words
• Metadata queries with boolean expressions such as ?author:Andrew?
• Filtering for regular expressions or date/metadata ranges
• Sorting by property/metadata in alphabetical or numerical order (and reverse)
• Using conceptual networks for statistical searching
• API and open infrastructure to explore dictionary and index information

Language and linguistics

• Stemming for increased precision on word-forms (Swedish, English etc..)
• Spelling alternatives and suggestions sorted by word-frequency
• Frequency and inference statistics
• Dictionaries for more than 15 languages

Document formats

• HTML-2,3,4/XHTML And some scripts
• XML with open XSLT-infrastructure for easy access
• Microsoft Office formats using MS-Components or wvWare with structured OLE2-format parser
• PDF parsing using PDF-to-HTML converter
• Add any other converter by command prompt extension
• Add any other converter by open Corpus API

Agents

• Regular harvesting spider (multi thread)
• Web log (Used with WEB-DAV)
• File system (act on changes)
• Summarizer - automatic short descriptions
• LDAP-Spider - create links connecting people, roles and documents
• Ref Parser (heuristics agent)
• Remote Warehouse (version control of remote information)
• Error-reporting - keep track of links and syntax (using mail or web-page)
• Console (for interfacing with corpus output or debug)
• API for any new agents

Interfaces

• HTTP/XML
• HTTP/RSS-Query (includes overviews, spelling and concept hints)
• HTTP/RDF-Query
• HTTP/Dictionary
• RMI/Query
• RMI/RemoteCorpus
• Java/CorpusService (Servlet-like-runtime-pluggable-services)
• Command line

• Application Programmers Interface (java doc etc)

Runtime environment

We seek to accommodate stability, robustness, platform independence and performance with our server. All of this is included in the environment.

• Java 1.4 server environment
  (some Corpus versions also support earlier versions of java)
• Nt-service or Unix sys-V script
• Adjustable system logging
• Access control and users
• Massive built in storage and database solution
• Optimized cache and object persistence

Administration

• Command line interaction
• Web-based services view screen-shots!
• Dictionary control (list/add/remove)
• Domain control (list/add/remove)
• Filter control (list/add/remove)
• Agents control (list/add/remove)
• Web-explorer