Tabular data is a broad term that includes structured data that generally fits into a specific row and column. It can be a SQL database, a spreadsheet, a .csv file, etc.
Although enormous progress has been made in the field of artificial intelligence that has been applied to unstructured and successive data, these large language models are blurry due to design. They are built to manipulate inpoertokens to generate a coherent exit without necessarily following a fixed structure. The best LLMs are also expensive to gain access through an API or duration to run on your own cloud infrastructure.
And yet many companies already have a data strategy with a data warehouse or data cake to centralize all important data and some data scientists who can use this data to improve the company’s strategy.
French startup Neuralk-ai An artificial intelligence company worked on AI models aimed at table data. The company announced $ 4 million in financing this week.
“Data with real value for companies are data that was identified long ago, structured in the form of a table, and used by the data scientists of these companies to make all their machine learning algorithms,” co-founder of Nueralk-AI and Chief Scientist Officer Alexandre Pasquiou told Techcrunch.
Neuralk-AI thinks there is a chance to re-view the development of AI model, but with a specific focus on structured data. In the first instance, it is planning to offer his model as an API of data scientists who work for Commerce companies because these companies Love Data – Think of product catalogs, customer databases, trends in shopping carts, etc.
“Nowadays, LLMS are great for searching, natural user interaction and answering questions based on unstructured documents. But it has some restrictions when we go back to Classic Machine Learning, which is really based on classic table data, “said Pasquiou.
With Neuralk-AI, Retailers can automate complex data workflows with smart deduplication and enrichment. But they can also use the models of the company to detect fraud, optimize the product recommendations and generate sales forecasts that can be used for stock management and product prices.
Fly Ventures led the Tour of $ 4 million of the company with Steamai, which also participated. Various business angels also invested in the startup, such as Thomas Wolf from hugging face, Charles Gorintin from Alan, and Philippe Corrot and Nagi Letaifa from Mirakl.
The team still works actively on its models. It is planning to test with a group of leading French retailers and startups of the trade, such as E.Leclerc, Auchan, Mirakl and Lucky Cart.
“Within three or four months we will release the first version of our model and the public benchmark on which we can rank our model compared to the state-of-the-art in this room,” Pasquiou said. “And in September the idea is to be the best Tabular Foundation model in everything related to learning representation.”