The Cognitive Workbench in a Nutshell

How can I converge knowledge sources? Is there Artificial Intelligence to automatically make sense out of that? How can I apply that new knowledge?

The answer to these questions is the Cognitive Workbench. Three steps:

Continue reading

Intro to the Cognitive Workbench

The Cognitive Workbench is a deep natural language processing enabled knowledge management application and development framework. It is designed to support users in performing activities, accomplishing tasks and organizing the information and resources they use every day. The Cognitive Workbench is capable to work with any data, be it the personal data (e.g. emails and texts), or be it professional data (news stories, scientific data, etc.), finding structure and regularity in their everyday activities. Features:

Continue reading

The Cognitive Workbench’s Graph Database

The heart of the Cognitive Workbench is its innovative Graph Database, a flexible, typed, semi-structured database that supports complex and intelligent interaction while remaining small enough to run on mobile platforms. Graph databases have a number of advantages for applications, and a long history in artificial intelligence and natural language processing. They reflect more naturally the structure of human linguistic information than more traditional relational and record-oriented databases. Continue reading

Graph nodes: Entities

The Cognitive Workbench stores knowledge in the form of entities and relations between them. At this basic level of analysis, there are facts which are observable in the user’s data, e.g. in emails or calendar entries, in news stories, in scientific papers or patents. or any other kind of user-generated record.

Continue reading

Entity Relations

The Cognitive Workbench Graph Database stores knowledge in the form of entities and relations between the entities. With this it is possible to incorporate any type of knowledge representation. For example it is possible to model the user his private life into four basic entity types: An email account has a type Place and can intuitively been seen to contain a repository of emails called “Inbox”, also of type Place which contains a specific email of type Thing. An email is naturally associated with a send-time of type Time and a sender or recipient of type Person. Given this entity modeling it is then possible to introduce untyped and types relations between these entities, such as beneficiary or agent.

Continue reading

Named Entities in the Cognitive Workbench

The Cognitive Workbench database goes beyond just storing real-world entities and their meta-information. Using natural language processing technologies, the Cognitive Workbench extracts from the unstructured text itself references to times and people, as well as geographic names like cities, company names, diseases, drugs, genes, and potentially many other categories. These too are stored in the Cognitive Workbench, with relations to the documents from which they are extracted and to each other if they appear together in the text.

Continue reading