Improving data harvesting/collection for AgFlow by leveraging Natural Language Processing (NLP) and Machine Learning (ML) techniques

Requester: AgFlow S.A. (AgFlow)
Anyone following agricultural commodities currently has no other choice but to source key market information from data lakes of news articles, reports, or analyses that come in many different formats and require much effort to process.

AgFlow is changing how the industry gets access to crucial market information by harvesting various datasets out of these data lakes, structuring them, standardizing them, and serving them through an easy-to-use interface that transforms tedious-to-process information into actionable data. So far, we have achieved to streamline data harvesting for tabular formats. However, contextual information is still tricky to extract without employing legions of market analysts/journalists who would spend their day reading agricultural commodities market news articles, reports, or analyses to highlight trends.

Using Natural Language Processing (NLP) techniques within our ecosystem, we will extract key market data in a scalable way, enrich our database, and enhance our extrapolation models powered by Machine Learning (ML) to serve our users through our platform.