If the composition of the same washing powder sold in Poland and Germany is different, we are dealing with so-called dual quality. Although it is an unlawful practice, exterminating it constitutes a challenge. The Produktoskop, a tool developed by researchers from the Łukasiewicz – Poznań Institute of Technology and the Poznań University of Technology, will be helpful in efforts to protect consumer rights.
Any product introduced into the European Union market, under the same name and in the same packaging, must have identical composition and characteristics (e.g. taste, behaviour, shelf life). If they are different, we are dealing with so-called dual quality. This is a practice prohibited in the EU because it misleads consumers. They have the right to expect a particular juice to be made 100% from fruit in Spane and Romania, and a washing liquid to leave the same long-lasting fragrance in Germany and Poland.
These rules are guarded by consumer rights institutions in EU countries. The problem is that they only respond to customer reports. Not all consumers who perceive a difference in quality report this problem.
Produktoskop – to the rescue
So how to deal with this? Produktoskop comes to the rescue – a software based on artificial intelligence, created by researchers from Łukasiewicz – PIT, the project leader, and Poznań University of Technology. It was commissioned by the Office of Competition and Consumer Protection and is intended to serve it.
The researchers analysed great datasets available online, including reviews concerning search engines and price comparison sites, online shops, and martketplaces, blogs or social media from across Europe. They were not limited to textual feedback – they also recognised images, audio recordings and audiovisual content posted on YouTube, Instagram, or TikTok. Obtaining unstructured data from websites and linking specific products to each other also proved a challenge.
- The key was to produce a tool capable of analysing diverse types of information in a single ecosystem. We have developed a solution to process text, images, and audio using advanced artificial intelligence algorithms. The analysed content is anonymised, ensuring the protection of personal data, broken down into found products and comments on them, and then assessed in a multi-stage manner in terms of the nature and impression of the opinions in question. This approach makes it possible to more accurately identify cases where a product introduced into different markets has a different composition or characteristics.
Maciej Niemir z Łukasiewicz – PIT, kierownik projektu
- This project was made possible through collaboration with a number of scientists and the involvement of a team of annotators who prepared the data for the machine learning processes. The development of dedicated, multilingual AI modules that provide high speed and effective analysis was also a key element. This will enable the Office for Competition and Consumer Protection to accurately identify potential cases of dual quality.
Mikołaj Sobczak z Politechniki Poznańskiej, kierownik naukowy projektu
Produktoskop will respond immediately when a review appears online suggesting the use of dual quality. This will enable the Office for Competition and Consumer Protection to investigate the product in question immediately, without waiting for consumer reports.
The software was developed under a grant from the National Centre for Research and Development (NCBR) under the INFOSTRATEG III programme.