Semantic computing

Content sourced from Wikipedia, licensed under CC BY-SA 3.0.

Semantic computing is a field that combines ideas from semantic analysis, natural language processing, data mining, and knowledge graphs to understand and use meaning in data. It focuses on extracting meaning from unstructured, semi-structured, and structured data and linking it across sources.

It goes beyond the traditional idea of the Semantic Web by including multimedia, services, and data that aren’t strictly web pages. It aims to access, use, synthesize, integrate, and analyze data and services.

Semantic computing brings together many technologies, such as software engineering, user interfaces, NLP, artificial intelligence, programming languages, grid computing, and pervasive computing, into one cohesive approach.

The field was named by Phillip Sheu in 2007, when he launched the IEEE International Conference on Semantic Computing and the International Journal of Semantic Computing. The early work described the field as technologies and interactions used to extract or process the contents and semantics of active services and passive data, regardless of how the data is stored.

A common way to think about semantic computing is as a layered architecture. A four-layer model includes:
- Semantic Analysis: interpreting signals like pixels or words to get meaning.
- Semantic Integration: combining content and semantics from different sources.
- Applications: using these meanings to solve problems and provide services.
- Semantic Interface: letting users access and manipulate semantic data across sources.

This was later expanded to a five-layer model by splitting Applications into:
- Semantic Services: solving specific problems with tools like web search, question answering, multimedia retrieval, and semantic synthesis.
- Service Integration: coordinating multiple semantic services to offer more capable and interoperable solutions.