The disambiguation tool uses the Cortical.io Retina API to identify contexts in which an input term is often found. A cluster of dots on the fingerprint grid of an input term corresponds to a particular context. The Retina API provides a programmatic approach to identify such clusters on fingerprints and to extract the corresponding contexts in which the term is found.
The Cortical.io Retina technology is accessed via a REST API. To disambiguate terms and create semantic fingerprints, you can include the following API calls in your application:
/expressionsendpoint: Generates a semantic fingerprint for an input term or expression (input expressions include minus signs or plus signs; for example, apple + computer)
/expressions/contextsendpoint: Returns a list of contexts in which the input term or expression is found
/expressions/similarterms endpoint: Returns terms that are similar to the input term or expression
To begin to integrate Cortical.io services into your application, register for a free API key.
To discover contexts that are associated with terms, follow the instructions on the screen. The output consists of two lists:
The meaning of the input is numerically encoded as a semantic fingerprint, which is graphically displayed as a square grid. Each blue dot on the grid contains part of the meaning of the text.
To view a larger fingerprint, above the fingerprint, click the double-arrow icon.
Every dot on a semantic fingerprint represents a collection of terms that are associated with one another. You can view some of the terms that are located at any of the blue dots on the grid by hovering over that dot. Dots that are close to one another on the grid are also close in meaning. Together, all blue dots on the grid represent the meaning of the inputted text or web page.
On the fingerprint grid, you can highlight dots that are associated with a particular context. You can also reduce the list of similar terms to only those that are associated with that context. In the output area, click the context that you want to focus on.
If you clicked a context, you can redisplay all similar terms by clicking the selected context again.
If you input, for example, the term apple, the resulting semantic fingerprint represents various different meanings of the term, including a type of fruit, an information technology company, and a record label. You can remove ambiguity by refining the semantic fingerprint. To do so, include
minus signs in the input; for example, apple – record – fruit; that is, apple minus record minus fruit.
Similarly, you can broaden the semantic fingerprint by including
plus signs between input words; for example, apple + computer.
You can export the disambiguation results (the associated contexts and similar terms) as separate .csv files. In the output are, click Export contexts or Export similar terms.