Nominal Terminology, Translation Terminology and Sentence-Level Memory – The Best of All Worlds

MultiTrans TermBase provides comprehensive tools to extract, generate, manage, share and use mono or multilingual terminology effectively and consistently across your language management value chain. TermBase leverages terminology information contained in MultiTrans-generated corpora and makes it easy for you to import terminology information from many other sources.

The repository structure of the TermBase allows you to simultaneously insert various forms of translations into your project and capture the terms in a repository for additional management and future reuse. Use TermBase to seamlessly integrate the following types of terminology:

• Nominal Terminology: regular, more ‘formal’ terminology usually presented in the nominal form.
• Translation Terminology: a recurring expression, typically 5 words or less, entered in the terminology database in the same form as it is used in the text, to be replaced by a pre-translation process.
• Sentence-level Translation Memory (TM): databases that have been previously created with traditional TM tools.

In addition to building nominal and translation terminology repositories on the fly, you can use the corpus expression extraction and alignment tools to rapidly build these repositories independent of a translation project.

Click to view a larger screenshotTerminology Extraction In The Blink of an Eye

Once a desired term is found in a corpus, it can be manually copied from the corpus, along with any translated equivalents, and added to the terminology repository with a single mouse click. This simple process also automatically fills in source, context, created-by, and created-date fields in the terminology file.

In a batch process, the corpus-building process automatically extracts all unique words and statistical expressions from referenced corpora that may contain millions of words of monolingual or multilingual text. Purely statistical extraction algorithms adapt to any language. User-defined “exclude lists” reduce terminology extraction noise by eliminating specified expressions from consideration.

Optional, advanced auto-generation algorithms analyze the terminology extraction from the source language of the corpus and compare it to the entire indexed multilingual corpus to identify and retrieve the corresponding translations. The system then displays a list of possible translations per expression, and a terminologist or a translator may validate the suggestions. This unique corpus-based approach enables you to quickly leverage the information contained in your multilingual documentation into large and specialized terminology databases.

Tight integration with the MultiTrans terminology repositories provides easy, automated capture of approved terms from the corpus. Flexibility is also provided to export terms into additional formats. Intelligent comparison algorithms identify the extracted words and expressions that already exist in the terminology databases and allow for convenient filtering so you are able to focus specifically on new terms.

Proven technology provides fast, scalable performance. In real applications, these algorithms have extracted over 60 000 expressions with their corresponding translations from a 6 million word corpus in a matter of hours.

Shared Terminology Management Across Your Language Management Value Chain

MultiTrans TermBase C/S (Client/Server) is MultiCorpora’s optional solution for building, sharing and distributing terminology across the many users in your language management value chain, which is accessible via a network, intranet or the Internet. It is built on an n-tier Client/Server architecture that provides high scalability and easy integration into large multi-server networks.

The architecture of MultiTrans TermBase C/S enables an organization’s terminology to be accessible, shared, and updated in real time, across geographical boundaries. The databases are centralized on a server within an organization and accessed in real-time via an internal network, Intranet or the Internet by a MultiTrans workstation or web browser. This architecture enables easy and cost effective deployment of terminology, while creating a synergy between all participants in the information processes, from authoring and translation, through revision and information usage.