KantanQES - More effective communications with A.I.

Quality Estimation

KantanQES empowers global organizations to understand the quality of machine translation, and to use machine translation in a more intelligent way.

Predictive Quality Estimation

KantanQES uses cognitive computing and deep learning to recommend high-quality machine translation texts. This can be used to prioritize translation effort, reduce costs and accelerate the delivery of multilingual content. This helps international organizations increase their competitiveness in the global market by providing customized multilingual content rapidly. KantanQES uses neural modelling to determine the quality of machine translation output which can be helpful in determining resource requirements, translation budgets and project timelines.

Improve communication effectiveness

By combining artificial intelligence and neural machine translation, KantanQES provides an objective confidence score or quality estimation for each source sentence.

How KantanQES Works

Integrate

KantanQES is seamlessly integrated into KantanMT neural machine translation engines.

Measure

Source sentences are processed by a neural network to determine language fluency and overall quality.

Analyze

A detailed analysis is produced to help Project Managers calculate translation resources, costs and timelines.

Prioritize translation activities

KantanQES (Quality Estimation Score) uses advanced neural network modelling to predict the translation quality of every sentence. These predictive scores can be used to plan project schedules, determine project costs and assign suitable resources in order to deliver projects on-time and within budget.

Translation Quality Estimation

Score each segment to determine translation quality using KantanAnalytics, a unique segment quality estimation technology

Project Costs

Calculate project costs using the KantanAnalytics report - a detailed report designed for Project Managers, which helps them predict project costs

Project Schedule

Predict the turn-around time of a project based on the overall post-editing effort required - a key feature in KantanAnalytics

Monitor Engine Quality

Track the improvements of your KantanMT engine at the segment level and see immediate results

Ready to learn more?

KantanAI works with some of the world’s largest organization to improve their products and services and deliver smoother customer journeys with A.I.

KantanQES Report

KantanQES creates a detailed project management report of all segments within a KantanMT project. This includes segment-by-segment quality estimation scores in addition to other useful project statistics such as word, character, placeholder and tag counts.

Tiered Cost Model

Tied to the quality estimation scores, higher scoring segments can be charged at a lower post-editing rate. This creates a tiered business model which more accurately predicts the actual cost of post-editing for a machine translation project.

Predict Project Schedule

As higher scoring segments are proven to take less time to post-edit, a more accurate and precise project schedule can be determined using the KantanQES statistics.

KantanQES Reports

These reports provide detailed analysis of the performance of KantanMT engines - including segment level quality scores, word, character, tag and placeholders counts. Think 'Fuzzy Match' report for MT and you've got the general idea!

Monitor Quality Performance

The better the KantanQES score - the better the quality performance of a KantanMT engine. An engine with a high distribution of low-scores should be retrained or replaced with a better more domain specific engine. KantanQES helps Project Managers prioritize translation activities.

How can KantanQES improve translation velocity?

Challenge

A client requires a large technical manual to be translated quickly, but has a limited budget. How can the project manager use machine translation to accelerate the project delivery while staying within tight budgetary constraints?

Solution

KantanQES provides quality estimation scores for every source sentence, ensuring translators only post-edit low-quality sentences leading to faster translation turn-around times.