47.4% Better Than Google NMT For Spanish-to-English IT Security

This Spanish-to-English IT Security translator uses a Slate Desktop personalized SMT that’s 47.4% more productive than Google NMT.

A Slate Rocks customer (a translator) created a translation engine with his or her translation memories (TMs) using a personal computer. This page describes the engine and compares the translator’s Slate Desktop experience to an experience using Google’s new-and-improved neural machine translation (NMT) technology. You can read the entire report with thirth (30) more customer experiences by downloading the full report Study of Machine Translated Segment Pairs.

Slate Desktop Engine Details

The customer started with translation memories in the language pair and industry of his or her work, totaling the estimated corpus size number of segments. Slate Desktop cleaned the TMs, prepared a training corpus and built the engine. Note that these processes typically runs overnight. During that processing, Slate Desktop also extracted a representative set consisting of randomly selected segments from the training corpus.

sourcees
targeten
subject domainit-security
estimated corpus size80,000
segments per representative set2,336
words per source segment20
words per target segment18

Benchmark Score Comparison

The segment pairs in the representative set are representative of the translator’s daily work. By focusing on one translator’s experience, these scores indicate a level of work reduction this customer will likely experience in his or her daily work using the respective MT system (Google or Slate Desktop) with 95% confidence.

 Google
NMT
Slate Desktop
SMT
words per MT segment1918
BLEU score49.2382.11
exact MT match (count)1981211
exact MT match (percent)8.5%51.8%
words per exact MT match (count)814
filtered BLEU score (no exact MT matches)47.6870.58
segments requiring edit (count)2,1381,125
character edits per segment3327
total character edits70,55430,375

These scores indicate this customer using Slate Desktop will likely spend significantly less time editing MT suggestions than if he or she were using Google for this work. This is because Slate Desktop creates engines with the customers translation memories and optimizes them to predict how the customer translates. While on the other hand, Google optimizes its NMT service for millions of customers with countless demands.

Google’s three (3) longest exact MT matches

The exact MT match (count) in the Benchmark Scores table (above) is the number of segments that Google NMT successfully matched to the translator’s actual work, i.e. Google successfully predicted the translator’s actions. The three segments in this table are the exact MT match segments with the longest length. This translator can expect to experience these kinds Google NMT results while translating these kinds of project.

esen (Google and translator)
En la configuración del proveedor de identidad, configure el dominio de identidad y, opcionalmente, el identificador corto del proveedor de identidad.In the configuration of the identity provider, configure the identity domain and, optionally, the short identifier of the identity provider.
Registre las entidades de confianza del sistema (CA, VA, TSA, proveedores de identidad, validadores de autenticación).Register the trusted entities of the system (CA, VA, TSA, identity providers, authentication validators).
La actualización 3.0.10S2R1_T_KI_C13 añade la siguiente funcionalidad al servicio de verificación de firmas.The 3.0.10S2R1_T_KI_C13 update adds the following functionality to the signature verification service.

Slate’s three (3) longest exact MT matches

The exact MT match (count) in the Benchmark Scores table (above) is the number of segments that this translator’s Slate Desktop engine successfully matched to the translator’s actual work, i.e. Slate Desktop successfully predicted the translator’s actions. The three segments in this table are the exact MT match segments with the longest length. This translator can expect to experience these kinds Slate Desktop results while translating these kinds of project.

esen (Slate and translator)
Dentro de la actividad en el grupo BIG, y con la finalidad de verificar la interoperabilidad entre las tecnologías SPOC/EAC de otros países que despliegan infraestructuras de pasaporte de segunda generación, Safelayer ha abierto un área de pruebas de SPOC para pasaporte electrónico conforme las especificaciones del BIG.As part of the BIG’s activity and with the aim of testing the interoperability between the SPOC/EAC technologies of the countries deploying infrastructures for the second generation ePassport, Safelayer has created a SPOC test area for electronic passports as per the BIG’s specifications.
Indica el nombre completo de la clase Java que actuará en el “nivel de presentación” para intercambiar el valor del elemento o atributo XML (dentro del documento XML global virtual que maneja el servicio EP) con el de la columna de la base de datos al que corresponda.Specifies the complete name of the Java class that is to act at the presentation level to exchange the value of the XML element or attribute (within the virtual, global XML document that the EP service handles) with the value of the column of the database to which it corresponds.
La solución permite utilizar diferentes repositorios de identidad que se encuentren en producción, basados en estándares como LDAP/AD o basados en bases de datos, mapeando los atributos de identidad y el formato de servicio que se proporciona basado en OAuth/OpenID Connect y SAML.The solution supports using different identity repositories that are found in production, based on standards such as LDAP/AD or databases, by mapping the identity attributes and the format of the service provided based on OAuth/OpenID Connect and SAML.
Advertisements