March 19 – 21 | PoolParty Summit

From March 19 to 21, the Semantic Web Company is organizing the 3rd PoolParty Summit under the motto ‘Unlocking the Full Potential of LLMs with Knowledge Graphs’. Users, experts and interested parties can be inspired by the PoolParty User Experience over 3 days – with virtual presentations, panels and demos. In line with current topics, the summit will show how PoolParty users are adapting to the rapidly changing requirements of digitalization with semantic technologies and AI.

Users will show how they have successfully overcome information silos, optimized inefficient work processes and overcome the challenges of unstructured data.

We are delighted that Nikhil Acharya, Knowledge Engineer at PANTOPIX, will be taking a look at current customer projects in his presentation on the AI-supported development of product knowledge databases. Our founder and Managing Director Karsten Schrempp will be taking part in the Intelligent Content Panel. Stay tuned and register directly for the PoolParty Summit 2024 free of charge.


March 20 | 6.45 – 7.45 pm
What Makes Intelligent Content Indispensable? An Expert-led Roundtable Discussion
Panelist: Karsten Schrempp and others

With its ability to handle multi-format, multi-channel, multi-lingual, and multi-authored content, intelligent content authoring has become indispensable. Companies do not have to worry about content being lost and can increase the shelf-life of a piece of content, making it all the more valuable. Intelligent content fuses technology with human capability and knowledge so that it is findable (via understanding human search intent), reusable, and machine readable (via structured components and metadata). This roundtable will bring together the different voices of the Content Component Alliance, where experts in the intelligent content space each have their own hand in building and/or using content that better suits the needs of a workplace.


March 21 | 11.15 – 11.45 am
Property Modelling for Product Ontology using Vector Embeddings driven by LLMs and OCR
Speaker: Nikhil Acharya

Identifying Entities and Relations from Heterogeneous data sources related to Technical Documentation is an important part of building a Knowledge Base. Technical Data comprises of Tables, raw text and Images pertaining to different products. We use Pretrained LLM and OCR models to identify Product and Product properties from these sources. This extracted Product information is now disambiguated using vector embeddings and mapped to specific entities and relations in our PIM ontology. This usage of AI tools helps us build a much more concrete Knowledge Base for our customers compared to standard data transformation approaches which only work on structured data and are rule based.

Contact us

Maraike Heim
Head of Marketing

Team Maraike Heim