System and method for parsing a document
View Patent ↗A parsing system and method are provided in which the break characters in the document are used to rapidly parse the document and extract one or more key phrases from the document which characterize the document. The break characters in the document may include explicit break characters, such as punctuation, soft stop words and hard stop words. The determination of which phrases in the document are extracted depends upon the type of break character appearing after the phrase in the document. The parser may also be used to parse a foreign language document into one or more phrases.
1. A system including a computer processor for parsing a piece of text into one or more phrases which characterize the piece of text, the system comprising:
a buffer, configured for storing one or more words from the piece of text until a break character is identified;
a parser that extracts at least one of the one or more words from the buffer when the break character is identified and which stores at least one of the at least one extracted words as a phrase in a memory based on a type of the break character identified, wherein the parser further extracts additional occurrences of the at least one extracted words from the piece of text after the buffer is parsed.
2. The system of claim 1 , wherein the buffer further comprises a means for flushing the buffer when the at least one of the one or more words are extracted from the buffer.
3. The system of claim 1 , wherein the parser only extracts noun phrases from the buffer.
4. A method for parsing a piece of text into one or more phrases which characterize the piece of text, the method comprising:
reading one or more words from the piece of text into a buffer until a break character is identified;
extracting at least one of the one or more words from the buffer when the break character is identified using a parser;
storing at least one of the at least one extracted words as a phrase in a memory based on a type of the break character identified; and
extracting, by the parser, additional occurrences of the at least one extracted words from the piece of text after the buffer is parsed.
5. The method of claim 4 , further comprising flushing the buffer when the at least one of the one or more words are extracted from the buffer.
6. The method of claim 4 , wherein the parser only extracts noun phrases from the buffer.
7. A system including a computer processor for parsing a piece of text into one or more phrases which characterize the piece of text, the system comprising:
a parser, comprising:
a first pass processing means for extracting at least one of one or more words from a buffer when a break character is identified; and
a means for storing at least one of the at least one extracted words as a phrase in a memory based on a type of the break character identified; and
a second pass processing means for retrieving additional occurrences of the at least one extracted words from the piece of text after the piece of text is parsed.
8. A method for parsing a piece of text into one or more phrases which characterize the piece of text, the method comprising:
performing a first pass through the piece of text by a parser, the first pass comprising:
extracting at least one of one or more words from a buffer when a break character is identified; and
storing at least one of the at least one extracted words as a phrase in a memory based on a type of the break character identified; and
performing a second pass through the piece of text by the parser, the second pass comprising retrieving additional occurrences of the at least one extracted words from the piece of text after the piece of text is parsed.
9. A method comprising:
receiving a first word from a text stream into a buffer;
receiving a first break character that follows the first word within the text stream into the buffer;
temporarily retaining the first word within the buffer, based on the first break character;
receiving a second word from the text stream into the buffer, the second word following the first break character in the text stream;
receiving a second break character into the buffer, the second break character following the second word in the text stream;
extracting the first word and the second word from the buffer by a parser and storing the first word and the second word as a phrase in a memory prior to receiving a third word from the text stream at the buffer; and
extracting additional occurrences of the first word and second word from the text stream after the text stream is parsed by the parser.