Context

Netquest is a company specialising in data collection for the market research industry. They offer a 360-degree view of consumers by collecting and analysing behavioural and declarative data from a single source: their panellists when they browse marketplaces.

Currently, they perform information extraction through rules. This model is not scalable.

Objectives

• Improve data structuring with a solution adaptable to platforms and different languages, reducing dependencies and ensuring maximum transparency in data extraction.
• Automate and optimize product information extraction to provide more detailed insights to retail clients that allow a better understanding of consumer purchasing habits.
• Reduce costs of adapting to new channels and increase efficiency in managing large volumes of data.

How have we done it?

• NLP
• Entity Recognition
• Information Extraction
• Generalization
• Labelling Tool

Results

Improved efficiency, scalability and accuracy in the extraction of relevant information, both in current platforms and in new marketplaces.

Interactive tool to visualise model predictions, access control metrics and solution performance.