NATURAL LANGUAGE GENERATION
Do you need to write a report describing data? To explain income statement lines or prepare an executive summary of the quarterly sales data? No mistakes allowed — precisely the type of text writing which should be automated.
Are you looking at Business Intelligence dashboards and finding it hard to figure out what is important among all those pie charts and tables? Would a line of text drawing attention to the essential aspects help? That's where automatically generated outlier descriptions would be of great use.
E-shop owners need accurate product descriptions underlying the main attributes of the product. Yet often, there are no resources available to manage the ever-changing inventory. There, product description automation would be of great help.
Accelerated Text is a Low-code Natural Language Generation platform that enables you to automatically add narrative to your data.
Automate report generation
We augment data with tables, charts, and pictures to help us understand raw numbers and facts. Yet, it is the narrative — a story explaining it — which truly helps us understand the data and make informed decisions.
Accelerated Text is the natural language generation engine. It will automatically create narratives highlighting outliers, describing trends, and providing explanations in the style and wording most suitable for the intended readers.
Accelerated Text is the open source “data to text” natural language generation engine that allows you to define data descriptions and then generates versions of those descriptions varying in wording and structure. It can work with much of your data:
Customer interaction data
You can generate multi-language text for a range of cases. From your business reports to the e-commerce platform to customer support applications.
Accelerated Text follows the principle of strict adherence to data-bound text generation. It provides instruments to define how the data should be described. Its natural language generation engine produces texts that are always and only about the data provided.
Data descriptions require precision. For instance, a description of weather conditions cannot invent things beyond the provided data: Temperature: -1C, Humidity: 40%, Wind: 9km/h. The report can only state those facts — nothing more, nothing less. The expression about the temperature may vary: cold, just below freezing, or -1°C but the fact must be stated. A "data to text" system is not the one to elaborate on the story adding lines about the freezing lake's serenity — it was not in the supplied data.