Knowledge-based Recommendation Systems as a Cornerstone of the Digital Workplace

There is a great need for intelligent assistants in digital workplaces, and this more than ever, considering the increasing degree of decentralization in the post-Corona era.||||Recommendation systems are well known through their use in e-commerce platforms, but there they often achieve only too low a degree of precision and relevance, since they only follow the principle of "more of the same" based on "collaborative filtering". In other words: in the digital workplace, we need smarter recommenders so that they are actually accepted and used.||||Recommender systems often have too little contextual information and background knowledge, so they can't really respond to individual user or customer requests.||||So it takes other methods to provide employees and users with truly relevant information, suggestions and "next best actions" that users actually rate as helpful.||||To stay with the e-commerce example: A better recommendation would also mean that after buying a couch, the next thing recommended is a coffee table to complete the living room, rather than yet another couch. A recommendation system based on semantic AI and knowledge models is able to intelligently link user preferences with knowledge or products, resulting in overall solutions.||||In this presentation, I review different use cases for recommender systems based on semantic AI, discuss scenarios from different industries (retail, manufacturing, software industry, and pharma), and look at how recommender systems serve as building blocks for broader concepts such as 'enterprise 360' or 'knowledge hub'.||||This talk addresses the following questions:||||* How do recommender systems support the automatic generation of holistic and personalized views of business objects such as products, customers or employees, and build by that the core of any digital workplace infrastructure?||* What are the specific use cases for recommender systems and what are the benefits?||* How are precise, explainable, controllable, and adaptive recommender systems built, and what is the role of semantic knowledge models and knowledge graphs?||||To this end, I will show several demos and use cases to give ideas for the use of recommender systems in enterprises.


Available material for this talk.