Knowledge Based Recommender Systems

This webinar provides a comprehensive overview of possible uses of recommender systems, and why they should be central building blocks in the digital workplace and in almost all enterprise information systems, such as knowledge and content hubs, customer experience platforms, or support systems.

From a technical point of view, we will further address how recommender systems essentially work and how they can be optimized. For this purpose, different approaches will be compared and we will specifically address those recommenders that are based on knowledge graphs and semantic knowledge models.

We will explain the methodology for knowledge-based recommendations and show how we can apply taxonomies, ontologies and rules to provide recommendations with different levels of expressivity. We also show how this methodology is realized in the PoolParty platform architecture using different software components.

Especially in knowledge-intensive workflows, where decisions have to be made continuously, users benefit from accurate suggestions. The business objects (documents, experts, products, suppliers, etc.) that best fit the respective context are automatically brought into the spotlight. The possible applications are manifold and we will go into numerous examples and demos and point out, among others, applications in the environment of HR management, drug approval, technical support for software or legal.

The webinar is aimed at AI and information architects, knowledge managers, UX specialists, but also directly at department managers who are responsible for better support, more efficient process management or accurate risk management in their areas.

The core topics and questions that Robert David (CTO, Semantic Web Company) and Andreas Blumauer (Co-founder & CEO, Semantic Web Company) will address in this webinar are:

– Why are recommender systems essential for a successful digital transformation?
– What are their potential and possible applications?
– What methods are such systems based on, and how can they be optimized?
– How do you start a project to establish a recommender in your own environment?