Event matching by analysis of text characteristics (e-match)

A system for event matching, the system comprising:

• at least one processor;
• and
• a memory storing instructions that, when executed by the at least one processor, cause the at least one processor to perform functions comprising:
• acquiring a plurality of documents, wherein each document of the plurality of documents comprises respective metadata that is indicative of an event attribute,
• identifying a document subset comprising multiple documents of the plurality of documents using the respective metadata for each document of the multiple documents,
• extracting a first salient text feature from a first document of the multiple documents and a second salient text feature from a second document of the multiple documents,
• determining, based on a comparison between the first salient text feature and the second salient text feature, an event similarity score for the first document and the second document;
• and
• upon determining that the event similarity score satisfies a threshold condition, including the first document and the second document as a pair of documents in a common event document list, wherein the common event document list identifies pairs of documents whose respective event similarity scores satisfy the threshold condition, and
• providing, for display, the common event document list.

A non-transitory computer-readable medium having stored therein instructions that, when executed by at least one processor, cause the at least one processor to perform functions comprising:

• acquiring a plurality of documents, wherein each document of the plurality of documents comprises respective metadata that is indicative of an event attribute,
• identifying a document subset comprising multiple documents of the plurality of documents using the respective metadata for each document of the multiple documents,
• extracting a first salient text feature from a first document of the multiple documents and a second salient text feature from a second document of the multiple documents,
• determining, based on a comparison between the first salient text feature and the second salient text feature, an event similarity score for the first document and the second document;
• and
• upon determining that the event similarity score satisfies a threshold condition, including the first document and the second document as a pair of documents in a common event document list, wherein the common event document list identifies pairs of documents whose respective event similarity scores satisfy the threshold condition, and
• providing, for display, the common event document list.