Method for calculating semantic similarities between messages and conversations based on enhanced entity extraction
AssigneePalo Alto Research Center Incorporated
InventorsOliver Brdiczka, Petro Hizalev
Life of this patent
Estimating the term…
Abstract
One embodiment of the present invention provides a system for estimating a similarity level between documents. During operation, the system receives a number of documents, extracts a number of semantic entities from the documents, calculates the similarity level between the documents based on the extracted semantic entities, and produces a result indicating similar documents based on the calculated similarity level.
Claims25
A computer-executable method for estimating a similarity level between documents, the method comprising:
- receiving a number of documents;
- extracting a number of semantic entities from the documents;
- calculating the similarity level between the documents based on the extracted semantic entities;
- and
- producing a result indicating similar documents based on the calculated similarity level.
A computer-readable storage medium storing instructions that when executed by a computer cause the computer to perform a method for estimating a similarity level between documents, the method comprising:
- receiving a number of documents;
- extracting a number of semantic entities from the documents;
- calculating the similarity level between the documents based on the extracted semantic entities;
- and
- producing a result indicating similar documents based on the calculated similarity level.
A system for estimating a similarity level between documents, comprising:
- a receiving mechanism configured to receive a number of documents;
- an entity-extraction mechanism configured to extract a number of semantic entities from the documents;
- a similarity-calculation mechanism configured to calculate the similarity level between the documents based on the extracted semantic entities;
- and
- a result-generation mechanism configured to generate a result indicating similar documents based on the calculated similarity.
Dependencies are inferred from the claim text.