The visit highlighted and strengthened our relationship even further by focusing on the expansion and deployment of our PangeaMT customized MT solution for Sony Europe in several languages and exploring the integration of our technologies in tailor-made, corporate globalization management environments.

PangeaMT provides industry specific statistical machine translation (SMT) engines for automotive, consumer electronics, and industrial sectors. The service was launched in 2009 at a recent TAUS User Conference with an offer to train engines for free for companies seriously looking into deploying open source MT with a TMX workflow. If you would like to know more about PangeaMT´s current Spring campaign, please contact us.

* Worth pointing out that Pangeanic leads the TDA data downloaders´ list, well ahead companies, such as Lionbridge, Oracle or WeLocalize. Downloaded data: 302,334,953 words. Info collated by TAUS Data Association and distributed to their partners at the end of March 2010.

Pangeanic relies heavily on the previous and current international research and expertise in Machine Translation and related areas of some of their team members. This is why conferences such as EAMT are no new to us at all. We are now happy to devote as much effort as possible to keep abreast of and benefit from what the MT research arena has to offer to companies like us. It is worth pointing out that EAMT 2010 was an event catering for academics, researchers and corporate practitioners alike. Although PangeaMT was represented but not formally discussed in any presentation, conference attendants, especially those coming from the industry and some universities, with whom we collaborate, were fully aware of our advancements in the field.

An example of this would be the improved version that PangeaMT has been made available today to Sybase, one long-standing client for the English-German language pair.
This version makes use of special tokenization techniques and post-processing modules to reach a considerably better output.

The version is customization of Moses that integrates features such as
* TMX generator (for TMX input and output)
* TXT data handling
* Inline parser to handle tags and formatting information contained in TMX directed to documentation (HTML, FrameMaker, InDesign, Word, etc)
* Cygwin integration

If you are considering the implementation of MT in your workflow, please contact Ms Elia Yuste at our Business Development Department.

The Tokyo summit will be the first of its kind in the country and it will have a strong focus on Use Cases and MT practical applications to localization workflows for Japanese industries or the Japanese language.

PangeaMT will feature as a leader in open standards implementation, with a strong focus in compatibility via its TMX and XLIFF workflows.

If you would like to speak to a representative of PangeaMT, please email us.

The award number is IMIDTA/2009/741.  For Pangeanic, this marks the beginning of a series of developments into other European languages and different combinations to service both industry and institutions.

The award corroborates the company’s long-term drive to implement, develop and offer customized translation automation solutions that accelerate and cut multilingual translation costs.

The strategy of investing in people who can bring new skills and customize fit-for-purpose solutions to particular machine-translation applications is highlighted in the report, which places importance in access to data via initiatives like TDA.

Whilst large sets of data do not automatically translate in perfect statistical machine translation engines, the selection and customization of TM as well as other data is one of the key points in developments designed to accelerate language transfer, bringing time and cost savings to companies.

The study also reports on large LSPs like SDL and Lionbridge and the possible MT ”lock” strategies behind their marketing.

Pangeanic.com.MT is the statistical machine translation division, with a mission to build and adapt SMT solutions that will work effectively in particular applications and domains, with client data and customized sets.

The report can be downloaded from the link below.

http://ec.europa.eu/dgs/translation/publications/studies/size_of_language_industry_en.pdf

Pangeanic conducted a series of tests with PangeaMT for specific language domains by combining its own statistical data with data obtained from TAUS’sTDA during late September. The aim of the test was to prove that increased amounts of trustable, regular data from TDA would help Pangeanic’s own technologies to improve output percentage quality, and to open up new domain developments.

Background

>Version 1 was a development concerned mainly with technical/engineering, electronics and automotive industries for general, user-manuals and scientific journal publication. Version 2 (PangeaMT) builds on that experience and adds several new areas: Software (SOF), Consumer and Professional Electronics + Computer Hardware (ECH), Marketing-Business-Economics (MBE), Legal-Pro (LEG), Healthcare-Pharma-Life Sciences (HEALTH).

PangeaMT is based on a Moses engine enhanced with an applied set of heuristics according to each language in question. The translation process is fully TMX-based. The concept is to have SMT acting as a plug-in to existing systems, not as an alternative solution or technology. It also integrates a parser that can interpret code/tags in the TMX and place it in the resulting translated segment. Post-editing can take place in any environment, thus resulting in an application-agnostic SMT plug-in.

Data

Three domains were selected for the test in the English-Spanish language pair (no distinction as to Lat.Am/EU), with the following number of files:

• ECH (Electronics-Computer Hardware): 800 tmx
• MBE (Marketing-Business-Economics): 76 tmx
• SOF (Software): 80 tmx

Data sets were selected according to the following criteria.

a) Language Model to follow

b) TDA data availability

c) Subject field

ELECTRONICS – COMPUTER HARDWARE

The aim was to improve on existing engines (Electronics). To this end, TDA data from Intel and Dell in Spanish was added to existing sets coming from Sony. Not all data available from TDA from particular donors was used as fit for the customized training. Some was discarded for a variety of reasons. Client-specific terminology was applied to original donor’s data sets for terminology standardization purposes. Pangeanic contributed with small sets of self-generated data. The result was a medium size 3,9M word engine specifically designed for the field of application and with the client’s terminology applied through donor’s TMX files in order to ease post-editing.

The data set for electronics was:

SOFTWARE

The aim of this development was to build a fresh engine with TDA data only in the subject field of a potential client to offer a solution which would show enough ROI for our SMT as a plug-in. To this end, we selected TDA data from several software donors in a subject field related to the product lines. We did not include Microsoft data initially as the size of the TM would have created a bias towards Microsoft terminology. However, engine enhancement is not discarded in future or more general releases. Again, not all data available from TDA from particular donors was used in the customized training. Some data was discarded and Pangeanic contributed with small sets of self-generated data.

The data set for software was:

MARKETING-BUSINESS-ECONOMICS

The aim of this development was to build a first test-bench engine serving as a business case within an uncontrolled, general field that has usually been “a work of literature” and out of the scope of traditional MT systems (particularly Rule-Based MT). Marketing and Economics are above natural speech and can be elaborate, complex texts and sometimes flowery or metaphorical. Again, the aim is to offer a solution which would show enough ROI for our SMT as a plug-in. The client did not provide enough training data and TDA did not offer enough bulk related material for this purpose. In this case, to show some results was more essential than to finalize a large engine.

The data set for marketing-business-economics was:

Process

The tables below describe the processes followed in the training. We can see that sentence length increases from domain to domain, that 2,000 representative segments (just over 20,000 words in all three cases) were not incorporated in the training so they could be used in the tests (BLEU/Meteor scores). Some sentences happened to be common (identical) to the training (18, 12, 2 respectively) mostly because of the nature of the source files (user manuals, software strings/commands in some cases which contain certain repetitions).

Perplexity is a measure that gives us an idea of the complexity of the task and how similar the test is to the training.??The higher the perplexity, the higher the difficulty.

Results

Model training + optimization: Moses+MERT

Language models: 5-grams

# TMX files for each category

ECH: 800

MEB: 76

SOF: 80

Translation results English->Spanish

BLEU: ECH: 49.98

MEB: 24.39

SOF: 47.78

Meteor 0.8.3

ECH: 0.4312

MEB: 0.2610

SOF: 0.4377

The best scoring domain is Electronics-Computer Hardware, with almost 50% scoring in BLEU and 43 in METEOR.

Results in Software are also very high (47,78% and 43,7% respectively).

This is a new domain for our development and we have used TDA data almost exclusively.

Marketing-Business-Economics lags behind with around 25% in both. Specific, “imaginative” marketing TMs weigh a lot here, and there is less content from TDA. Marketing literature is, by definition, not necessarily as accurate as the other two fields, which are fairly controlled languages. The engine was a first step, a test development still to be enhanced with further data.

Nevertheless, the results surpass our expectations. A 50% BLEU-Meteor scoring can translate in large increases in language production. Even the 25%, as an initial result for marketing leaves a lot of room for improvement once even more data is available.