Explicit Thesauri

A thesaurus provides a list of industry-specific terms and their synonyms to a system so it can recognize these unusual words and phrases when it comes across them. This is useful in environments with a large corpus of industry-specific terms, abbreviations and jargon, such as the medical and scientific fields.

For example, a thesaurus can enable a system to understand that when someone is interested in cancer research, they may also wish to read documents that talk about oncology. However, thesauri are expensive and time-consuming to create and definitions are often not accurate, because words can change their meaning depending on context.

Manual

An expert must compile the lists or thesauri before they can be added into the system. This is a time-consuming and highly manual process. This approach also relies on the ability of a person or group of people to predict and include any term that is likely to be used by anyone using the system, so there is a significant possibility of error. However, it can be rewarding for specific applications where there is a prior knowledge.

No Ability to Learn

The lists are static once entered into the system. They are not able to learn as words change their meaning or as new words are added to the language. Therefore, whenever there is a change or addition, it needs to be added into the system by an administrator, making thesaurus maintenance another time-consuming manual activity. In contast, Autonomy will not only recommend for static datasets, but will also deal differently with constantly changing datasets such as news stories. Finally, words with more than one meaning can also cause problems. For example, a system that uses thesauri for understanding language would interpret the phrase "she's a star" to have the same meaning as "she's a cosmic gas ball."

Autonomy's Approach

Autonomy's understanding of concepts is built from the entire corpus of data analyzed by IDOL. Therefore, as new vocabulary and terms are introduced, or their meaning changes, they will automatically be updated in the system and the thesaurus does not need to be manually updated.

An example is the change in meaning of the term "Ground Zero" after September 11, 2001. With its original meaning being the point of detonation of a nuclear weapon, the phrase took on a new meaning overnight. Autonomy was able to immediately understand the change in meaning and see that it was being used in relation to the terrorist attacks on the United States, and not in relation to a nuclear blast. Autonomy can load and fully utilize any existing thesauri. However, Autonomy can still function without them, unlike solutions that require thesauri to be present.

Further Reference:

The Evolution of Search

Further Reference:

Autonomy's Unique Learning Ability

Forthcoming Events

Archived Events

There do not seem to be any upcoming events related to this page at the moment, please visit our seminar and webinar pages for more information.

There do not seem to be any case studies related to this page available yet, however please visit our publications site at http://publications.autonomy.com/ for more information.