We are pretty positive about Natural Language Generation (NLG). It would be a surprise, frankly, if we were not. We honestly believe that NLG solutions can be vital to answering questions facing companies and organisations today.
There are numerous benefits to companies that decide to implement NLG products as a support to their everyday operations. In e-commerce, an online retailer can take a dataset and use it to produce an unlimited number of product descriptions, hundreds per second, that can be updated automatically. Those texts produced with one dataset can be transformed automatically by adding another dataset, saving the time, effort, and resources needed to make such changes manually.
In the newsroom, breaking developments can be reported immediately while flesh-and-blood reporters race to gather facts, conduct interviews, and give events their proper context. There are numerous, real possibilities, many of which we have already been—or are—working on. Think about all the straightforward texts that you read in your day-to-day life that could be automated from a dataset: weather reports, real estate listings, financial reports, school reports, medical reports, etc… If something can be measured, it can be rendered as data. And if it is rendered as data, then it can be automated.
That is not to say that the sailing has always been smooth. There are issues and challenges that companies and organisations find when they try to implement an NLG solution.
I will address that in a second. However, at this point, it is instructive to take a step back and outline the NLG products that Retresco offers.
The first is known as textengine pro, which is Retresco’s managed-service product. Customers opting for this solution work with Retresco’s team of data architects, programmers, developers, and linguistic data architects to develop and deliver bespoke, tailored solutions.
The second solution is textengine.io, which comes out of beta this week. This revolutionary self-service product puts the power and expertise to create NLG content into the hands of our clients.
Both services are built using the rtr textengine, one of the world’s most-powerful pieces of NLG software. The rtr textengine is built to generate text in up to five languages (English, German, French, Dutch, and Italian).
Textengine.io currently supports NLG in English and German. We will add Dutch soon and aim, over the coming months, to expand the offering to include Italian, Spanish, and French.
We developed textengine.io for a number of reasons, but mainly to facilitate organisations that want total control in developing an NLG solution. This is because we have noticed issues with the implementation of NLG that we think can be easily overcome with textengine.io.
Firstly, many organisations find that the texts they wish to automate are often intricate and complicated. A weather project, for example, relies on masses of complicated data, including temperature, humidity, time, air pressure, precipitation, cloud cover, time of day, and duration. Those things all work in combination, and any change in one necessitates a whole new set of text, plus its many variations. The same goes for financial services, where data such as EBIT, EBITDA, multiple rates of taxation, profit, and loss often has to be calculated and put into context. While any robust system designed to do such work can fire out hundreds of texts per second, it is important from the outset that they are set up with the knowledge of how to interpret this data correctly. Even though our experts are experts at what they do, they are not experts on every business sector. So, many companies may feel more comfortable if they have more control of the NLG project.
This moves us to the second issue that companies have in implementing an NLG solution. Occasionally, with textengine pro, some companies find themselves off-balance as they feel that they have lost ‘control’ of their project as it is put together internally by Retresco staff. It is a disconcerting feeling for many when they see their project come along in spates and spats, and not having overall management of its development. With textengine.io, however, companies are able to develop their own solutions in-house.
Retresco is proud of textengine.io. It is a product that we have been developing for over two years, and is built on the groundbreaking work that Retresco has done in Natural Language Processing (NLP) since the company’s inception in 2008. It comes out of beta with a host of features. These include:
- An intuitive interface that makes creating your own NLG projects straightforward and quick.
- Automatic text adaptation through intelligent linguistic analysis takes the data provided and adjusts your text automatically.
- Adaptable and easily-changeable project management meaning that you can continually hone your NLG work, maintaining control at every step of the process.
- Comprehensive onboarding and support from Florian, our customer success manager, to guarantee a successful project.
We built textengine.io to address the needs of customers looking to implement NLG solutions at a reasonable cost and with little friction. It is an aim that we think we have succeeded at. We hope you think so, too.
Founded in Berlin in 2008, Retresco has become one of the leading companies in the field of natural language processing (NLP) and machine learning. Retresco develops semantic applications in the areas of content classification, recommendation, as well as highly innovative technology for natural language generation (NLG). Through nearly a decade of deep industry experience, Retresco helps its clients accelerate digital transformation, increase operational efficiencies, and enhance customer engagement.