Blog Standard

Lorem ipsum dolor sit

Today, Sue Feldman, Hadley Reynolds and I officially launch the Cognitive Computing Consortium, a forum for researchers, developers and practitioners of cognitive computing and its allied technologies. This consortium has been more than 18 months in the making, during which time we've talked to companies, academics, practioners and analysts about what the kind of organization that can be of most help to them as they...

Read More
Linear classifier

Author’s note: This is the second in a multi-part series talking about current work on demystifying the black box nature of machine learning. Read part one here. Part 2. The BlackBoxitude of machine learning   In part 1 of this series, I described the various ways people use the term confidence in machine learning. Technologists tend to think of confidence in statistical terms, while laymen often use the...

Read More

Author's note: This is a multi-part series talking about current work on demystifying the black box nature of machine learning. Part 1. Confidence:  Probability vs Trust As a computer scientist working with data classification, I often get the question "what’s your confidence as a percentage, in this classification result from the software?" It’s taken me some time and a number of false starts to work out what...

Read More

This post covers autoclassification using a classification model such as a taxonomy or ontology. Read here to see the differences between systems that use machine learning and those that use the type of rulebase system I talk about below.   Historically, people manually have applied classification to documents and books to help users organize and consistently retrieve information. Libraries use the Dewey Decimal System to shelve and...

Read More

Providers primarily use two methods to metatag and categorize content. The first is rule-base systems that use classification schemes such as controlled vocabularies, taxonomies and ontologies. The second are machine learning systems.   Rulebase systems tend to be rigid, and they need a lot of maintenance, but their results are also more predictable. Rulebased systems tend to work very fast and be unsophisticated; consequently, that makes them...

Read More