Detection of data in a sequence of characters

Patent metadata

Field	Value
Publication number	US-9454522-B2
Application number	US-201414286838-A
Country	US
Kind code	B2
Filing date	May 23, 2014
Priority date	Jun 6, 2008
Publication date	Sep 27, 2016
Grant date	Sep 27, 2016

Abstract

A method of processing a sequence of characters, the method comprising converting the sequence of characters into a sequence of tokens so that each token comprises a lexeme and one of a plurality of token types. Each of the plurality of token types relates to at least one of a plurality of predetermined functions, wherein at least one said token type relates to multiple functions of the plurality of predetermined functions.

First claim

The invention claimed is:

1. A method to process a sequence of characters representing natural or computer language, the method comprising:
receiving, by a lexer executing on a processor, the sequence of characters in a communication between users or in output from one of a compiler and an interpreter;
converting, by a lexer, the sequence of characters into a sequence of tokens, at least one token comprising a proto-lexeme and a proto-token type, the proto-token type representing multiple token types, each of the multiple token types describing one of a plurality of predetermined classes ascribed to the proto-lexeme within the sequence of characters during the conversion,
wherein a parser, executing on the processor, resolves the proto-lexeme into a single lexeme having a single class that is one of the multiple classes represented by the proto-token type, and further resolves the proto-token type into a single token type describing the single class; and
when the single class is associated with a user application, presenting, by the processor, an option to update the user application with the single lexeme, wherein the user application includes one of a calendar and an address book.

2. A method according to claim 1 , wherein the proto-token type indicates that the proto-lexeme relates to the single class.

3. A method according to claim 1 , wherein the predetermined classes describe classes of the lexeme in the context of the sequence of characters.

4. A method according to claim 1 , wherein the parser further parses the tokens to detect predetermined types of data in the sequence of characters.

5. A method according to claim 4 , wherein each said type of data corresponds to at least one of said multiple classes and a predetermined combination of said multiple classes.

6. A method according to claim 4 , wherein the predetermined types of data include at least one of a physical address, an IP address, an e-mail address, a time, a day, a date, and a contact number.

7. A method according to claim 4 , wherein the parser further provides a single parsing path in a decision tree for each token type and each proto-token type in the sequence of token.

8. A method of processing data in a sequence of characters representing natural or computer language, comprising:
receiving, by a parser executing on a processor, a sequence of tokens, at least one token comprising a proto-lexeme and a proto-token type, the proto-token type representing multiple token types, each of the multiple token types describing one of a plurality of predetermined classes previously ascribed to the proto-lexeme within the sequence of characters received by a lexer in a communication between users or in output from one of a compiler and an interpreter;
parsing, by the parser, the sequence of tokens to detect predetermined types of data;
resolving, by the parser, the proto-lexeme into a single lexeme having a single class that is one of the multiple classes represented by the proto-token type, and further resolving the proto-token type into a single token type describing the single class; and
when the single class is associated with a user application, presenting, by the processor, an option to update the user application with the single lexeme, wherein the user application includes one of a calendar and an address book.

9. A method according to claim 8 , the method comprising:
providing, by the parser, a single path for each token type and each proto-token type in the sequence of tokens.

10. A method of processing a sequence of characters representing natural or computer language, the method comprising:
receiving, by a lexer executing on a processor, the sequence of characters in a communication between users or in output from one of a compiler and an interpreter;
converting, by the lexer, the sequence of characters into a sequence of tokens comprising one or more proto-lexemes and one or more corresponding proto-token types, wherein each proto-lexeme is defined as belonging to one of:
a first set comprising one class, wherein the proto-token type represents the one class, and
a second set comprising a combination of classes, wherein the proto-token type represents the classes in the combination;

resolving, by a parser executing on the processor, each proto-lexeme into a single lexeme and the corresponding proto-token type into a single token type;
when the single class is associated with a user application, presenting, by the processor, an option to update the user application with the single lexeme, wherein the user application includes one of a calendar and an address book.

11. A method according to claim 10 , wherein the resolving comprises providing a single analysis path for each said proto-lexeme in a decision tree.

12. A method according to claim 10 , wherein the classes in the combination may include the single class.

13. A method according to claim 10 , further comprising converting the sequence of characters into lexemes and proto-lexemes.

14. An apparatus to detect predetermined data in a sequence of characters representing natural or computer language, the apparatus comprising:
a processor;
a network interface coupled to the processor to receive the sequence of characters in a communication between users or in output from one of a compiler and an interpreter;
a lexer executing on the processor to convert the sequence of characters into a sequence of tokens, at least one token comprising a proto-lexeme and a proto-token type that represents multiple token types, each of the multiple token types describing one of a plurality of predetermined classes ascribed to the proto-lexeme within the sequence of characters during the conversion,
wherein a parser, executing the processor, resolves the proto-lexeme into a single lexeme having a single class that is one of the multiple classes represented by the proto-token type, and further resolves the proto-token type into a single token type describing the single class; and
when the single class is associated with a user application, presenting, by the processor, an option to update the user application with the single lexeme, wherein the user application includes one of a calendar and an address book.

15. An apparatus according to claim 14 , wherein the predetermined classes describe classes of the lexeme in the context of the sequence of characters.

16. An apparatus according to claim 14 , wherein the parser further parses the tokens to detect predetermined types of data in the sequence of characters.

17. An apparatus according to claim 16 , wherein each type of data corresponds to at least one said multiple classes and a predetermined combination said multiple classes.

18. An apparatus according to claim 16 , wherein the predetermined types of data include at least one of a physical address, an IP address, an e-mail address, a time, a day, a date, and a contact number.

19. An apparatus according to claim 16 , wherein the parser comprises a decision tree having a single parsing path for each token type and each proto-token type in the sequence of tokens.

20. A data processing system for processing a sequence of characters representing natural or computer language comprising:
means for receiving the sequence of characters in a communication between users or in output from one of a compiler and an interpreter,
means for converting the sequence of characters into a sequence of one or more proto-lexemes and one or more proto-token types, wherein each proto-lexeme is defined as belonging to o

Assignees

• Apple Inc

Inventors

• Bonnet Olivier
• De Jaeger Frederic
• Goyet Romain

Classifications

• G06F40/40

  Processing or translation of natural language (natural language analysis G06F40/20; semantic analysis G06F40/30) · CPC title

• G06F8/427

  Parsing · CPC title

• G06F16/284

  Relational databases · CPC title

• G06F40/289

  Phrasal analysis, e.g. finite state techniques or chunking · CPC title

• G06F40/284

  Lexical analysis, e.g. tokenisation or collocates · CPC title