PRODUKTOSKOP – Development of an information system using AI to identify consumer opinions on product safety and quality

Frame 1 (2)

Project co-financed by NCBiR under the 3rd INFOSTRATEG call (contract number INFOSTRATEG-III/0003/2021-00)

Acronym: PRODUKTOSKOP

Type: R&D

Grant: PLN 2 834 369.23

Total value: PLN 2 834 369.23

Start date: 01.08.2022

Project completion date: 31.01.2026

Aim of the project:

The aim of the Produktoskop project is to create an innovative system for finding and reporting on products about which there are doubts concerning their quality, with a special focus on cases of Dual Quality practices used by traders, based on the analysis of consumer opinions posted on the Internet. Artificial intelligence methods will be developed and used to build the system, which is in line with the main objective of the Infostrateg programme.

The results of the project will be used to implement solutions in the activities of the OCCP (Office of Competition and Consumer Protection) to assess whether there are differences in the quality of products in different countries, and will make it possible to monitor the situation in Poland and the EU in this regard.

The project is being carried out in a consortium comprising two research units: Łukasiewicz Research Network – Poznan Institute of Technology and Poznan University of Technology.

 

Results:

The outcome of the project will be an application using artificial intelligence, capable of searching Internet resources to collect current consumer opinions on a wide range of consumer products and, based on these, suggesting products for laboratory testing and suggesting accredited laboratories capable of performing the necessary analyses. In particular, the products of the project include:

  • A database of labelled opinions on a variety of products and from many sources, and intended for training and evaluation of machine learning algorithms
  • A database of accredited laboratories
  • Language processing models (e.g. Polish, English, German) based on deep neural networks adapted to the specificity of the language appearing in opinions and comments posted on the Internet
  • A system functioning in the form of a web application with an implementation manual integrating the developed components

 

Benefits:

The results of the project will be implemented by the OCCP. The results of the research will be made available to consumers to increase their awareness and knowledge and to protect their interests, and the material collected will be used in proceedings to protect the collective interests of consumers.

 

Left arrowALL CURRENT PROJECTS