Martinus: Identifying similarities in books

Discovering an optimal book to read next is a typical problem of almost every book lover. However, often it is tricky to find similar books. Different AI and NLP techniques can greatly help and speed up this process. Our aim is to research methods for extracting relevant book information, and to uncover the hidden similarities and relations among them.

KInIT joined forces with Martinus – the biggest Slovak e-commerce bookstore. Martinus was awarded as the “best” or “most favorite” e-commerce store in Slovakia several times. Their web and also the backend is developed by internal Martinus teams, which brings a great potential for innovations and research collaborations. This is fully inline with their goal – to have the trust of the readers and customers.

To support Marintus in improving their products and bringing innovations to the market, KInIT team contributes with its extensive know-how in Natural language processing and AI. In the joint project we are researching methods for discovering hidden relations and connections among the books. This is based on the processing of the books’ metadata.

Coupled with their advanced user interface, users will be empowered to find a relevant book for their taste, interests and needs. Helping them to discover serendipitous book suggestions for what to read next.

We are always happy to collaborate with established and successful Slovak companies, and are especially pleased about our project with Martinus, as we have numerous passionate readers in our team.

tomas ulej

We extend our sincere gratitude to the team at KInIT for their outstanding collaboration in our experiments and exploration of how AI might in the future assist readers in selecting books. Their professionalism, reliability, and agility are highly appreciated.

Tomáš Ulej

Explore team, Martinus

Project team

Andrej Ridzik
Research Engineer
Kris Farrugia
Research Engineer 09/2022-05/2024
Rastislav Papšo
PhD Student 09/2022-12/2023
Michal Kompan
Lead and Researcher