This paper presents a holistic system for semantic annotation, indexing, and retrieval of documents based on real-world entities. The authors propose a system called KIM, which uses a light-weight upper-level ontology to define entity types and their attributes, and a knowledge base to store entity descriptions. The system aims to provide efficient and reusable metadata by decoupling annotations from the content they refer to, allowing dynamic and user-specific annotations. The paper discusses the design and implementation of the KIM platform, including its architecture, ontology, knowledge base, information extraction, indexing, and retrieval capabilities. The KIM platform is demonstrated through a browser plug-in that highlights named entities and provides hyperlinks to their corresponding instances in the knowledge base. The paper also reviews related work and outlines future challenges, such as developing evaluation metrics for semantic annotation systems.This paper presents a holistic system for semantic annotation, indexing, and retrieval of documents based on real-world entities. The authors propose a system called KIM, which uses a light-weight upper-level ontology to define entity types and their attributes, and a knowledge base to store entity descriptions. The system aims to provide efficient and reusable metadata by decoupling annotations from the content they refer to, allowing dynamic and user-specific annotations. The paper discusses the design and implementation of the KIM platform, including its architecture, ontology, knowledge base, information extraction, indexing, and retrieval capabilities. The KIM platform is demonstrated through a browser plug-in that highlights named entities and provides hyperlinks to their corresponding instances in the knowledge base. The paper also reviews related work and outlines future challenges, such as developing evaluation metrics for semantic annotation systems.