Araya Web Server 
Dienstag, Dezember 30, 2008, 10:16 AM - Araya, Translation memory, XLIFF, TMX
Araya-Web-Server

Der Araya-Web-Server bietet eine einfache Möglichkeit TMX Datenbanken und Terminologie-Datenbanken unternehmensweit lesend verfügbar zu machen. Der Benutzerkann über seinen Webbrowser einfach und effizient auf die TMX Datenbanken und Terminologiedatenbanken zugreifen. Serverseitig muss nur TOMCAT oder eine ähnliche Java Servlet Maschine installiert sein. Die Suchoberfläche kann an die Bedürfnisse des Unternehmens angepasst werden. Es wird dabei die wortbasierte (termbasiert) die segmentbasierte Suche (unscharfe Suche) sowie Suche basierend auf regulären AUsdrücken unterstützt. - http://www.heartsome.de/de/araya.php#Araya-Web-Server

Araya-Web-Server
The Araya-Web-Server offers a simple yet powerful opportunity to access TMX and terminology databases in the whole enterprise. The user can easily query TMX and terminology databases from within a web browser. This only requires TOMCAT as servlet container to be installed on the server side. The browser interface can be easily adapted for user or enterprise needs. The browser supports both word based and segment based search. Segment based search uses the proveb Araya fuzzy retrieval approach. It also supports regular expression search. - http://www.heartsome.de/en/araya.php#Araya-Web-Server

  |  0 Trackbacks   |  Permalink   |  Related Link

Bilingual Terminology Extraction  
Samstag, Mai 26, 2007, 08:18 PM - Terminology Extraction, TMX
The bilingual terminology extraction has been extended with additional functions, esp. searching terms in the extracted term list. In addition terms can now be sorted in different orders (e.g. by score, by source term, target term, number). Extraction also has been heavily improved with reagrd to speed. A progress dialog is shown now too.

The tests done with the Europarl corpus have been continued for terms consiosting of several words und the minimum frequency has been lowered.
  |  0 Trackbacks   |  Permalink   |  Related Link

Europarl Corpus as test bed 
Donnerstag, Mai 3, 2007, 05:37 PM - Araya, Terminology Extraction, TMX
The Europarl Corpus is currently used to evaluate the bilingual term extractor. Europarl is a corpus of texts translated into various European languages based on meetings of the European Parliament.

Several of the extracted term files are available for download.

For more information see either:

http://www.heartsome.de/de/europarl.php
or
http://www.heartsome.de/en/europarl.php
  |  0 Trackbacks   |  Permalink   |  Related Link


Zurück