The semantic matching service determines whether one or more word senses are present in a document. A Sense Analysis is first performed on the document. The senses found are then compared to those provided by the appication using an advanced matching logic that considers equivalencies.
Typical applications of this service are:
- Social media monitoring: Matching can eliminate most of the noise in Tweets or other social media. An application can identify an initial analysis set based on keywords and then refine the analysis set using the matching service.
- Improving precision of searches: Matching can be used to improve precision by removing false hits (i.e., where the search keyword has the wrong sense).
- Profanity filtering: Matching can identify if a document contains profanities by matching all the senses against a set of a few hundred profane word senses.
Using the Service
The service request includes a filter for one or more sensekey. The sensekey value can be obtained from the Language Graph Browser. It will look something like “Tide/N8″ (the sensekey for the popular detergent.)
Some applications may wish to perform additional processing for documents that match using the results of the sense analysis (e.g., sentiment analysis). This result can be specified by adding parameter wsdMime to the request. The default output format is an XML file in Semdoc format. Learn more…
The conf property included with each element of matches (i.e., occurrence of a detected keyword) indicates the confidence of the sense analysis result. Applications must consider whether to trust low confidence decisions. For example, an application could decide to fallback to normal keyword logic if the confidence is less than 50%. Learn more…