Difference between revisions of "Timo Lahtinen"

From Nordterm Wiki
Jump to navigationJump to search
[checked revision][checked revision]
(Created page with "== CONEXOR NAVITERM - FACTS FROM DOCS == === Timo Lahtinen === NaviTerm is a tool that recognizes terms and proper names from texts automatically. In addition, proper names are ...")
 
 
(One intermediate revision by the same user not shown)
Line 1: Line 1:
 
== CONEXOR NAVITERM - FACTS FROM DOCS ==
 
== CONEXOR NAVITERM - FACTS FROM DOCS ==
=== Timo Lahtinen ===
+
 
  
 
NaviTerm is a tool that recognizes terms and proper names from texts automatically. In addition, proper names are classified to geographical names, the names of organisations, and the names of individuals.
 
NaviTerm is a tool that recognizes terms and proper names from texts automatically. In addition, proper names are classified to geographical names, the names of organisations, and the names of individuals.
Line 10: Line 10:
 
At the moment, the supported languages are English, Finnish, Swedish, Spanish, French and German.
 
At the moment, the supported languages are English, Finnish, Swedish, Spanish, French and German.
  
Timo Lahtinen
+
Timo Lahtinen<br/>
Conexor oy
+
Conexor oy<br/>
Helsinki Science Park
+
Helsinki Science Park<br/>
Koetilantie 3, 00710 Helsinki, Finland
+
Koetilantie 3, 00710 Helsinki, Finland<br/>
mobile: +358 50 3077803
+
mobile: +358 50 3077803<br/>
tel: +358 9 37468500
+
tel: +358 9 37468500<br/>
fax: +358 9 37468502
+
fax: +358 9 37468502<br/>
 
http://www.conexor.fi -Timo
 
http://www.conexor.fi -Timo

Latest revision as of 10:43, 20 October 2010

CONEXOR NAVITERM - FACTS FROM DOCS

NaviTerm is a tool that recognizes terms and proper names from texts automatically. In addition, proper names are classified to geographical names, the names of organisations, and the names of individuals.

The output of NaviTerm is up to the needs of users. It can be a list of terms and names, or a hyper text index with links to the source documents. It is also possible that terms and names are marked up into texts in their context by tags (e.g. SGML/HTML/XML).

The recognition of terms and names is based on automatic content analysis that includes syntactical parsing and detection of word distributions. The content analysis produces an estimation of what is important and what is less important in a given text. The importance is measured by weights.

At the moment, the supported languages are English, Finnish, Swedish, Spanish, French and German.

Timo Lahtinen
Conexor oy
Helsinki Science Park
Koetilantie 3, 00710 Helsinki, Finland
mobile: +358 50 3077803
tel: +358 9 37468500
fax: +358 9 37468502
http://www.conexor.fi -Timo