Document processing device, document processing method, and storage medium recording program therefor

• US 20060062492A1
• Filed: 03/16/2005
• Published: 03/23/2006
• Est. Priority Date: 09/17/2004
• Status: Abandoned Application

• Current Assignee: Fuji Xerox Company Limited (Xerox Holdings Corp.)
• Original Assignee: Fuji Xerox Company Limited (Xerox Holdings Corp.)
• Inventors: Itoh, Atsushi, Masuichi, Hiroshi, Tashiro, Kiyoshi, Sato, Naoko, Tamune, Michihiro, Tagawa, Masatoshi, Liu, Shaoming, Ishikawa, Kyosuke
• Application Number:
  US11/080,924
• Publication Number:
  US 20060062492A1
• Time in Patent Office: Days
• US Class Current: 382/305
• CPC Class Codes: G06F 40/211 Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars

Abstract

The invention provides a document processing device including: a memory that stores syntax data expressing syntax of character strings whose probability of being a title of a document is high or-character strings whose probability of being a title of a document is low; an input unit that inputs document data obtained by digitizing a document; an extraction unit that analyzes the input document data and extracts character string data expressing character strings; a syntax analyzing unit that analyzes the extracted character string data and specifies the syntax of each character string contained in the document corresponding to the document data; and a specifying unit that specifies, from among the extracted character string data, character string data expressing a title of the document corresponding to the document data, based on results of specification by the syntax analyzing unit and content stored in the memory.

Claims

1. A document processing device comprising:

a memory that stores syntax data which expresses syntax of character strings whose probability of being a title of a document is high or character strings whose probability of being a title of a document is low;

an input unit that inputs document data obtained by digitizing a document;

an extraction unit that analyzes document data input by the input unit and extracts character string data which expresses character strings;

syntax analyzing unit that analyzes the character string data extracted by the extraction unit and specifies the syntax of each character string contained in the document corresponding to the document data; and

specifying unit that specifies, from among the character string data extracted by the extraction unit, character string data that expresses a title of the document corresponding to the document data, based on results of specification by the syntax analyzing unit and content stored in the memory.

2. The document processing device according to claim 1,

wherein weight data expressing levels of probability that a character string with syntax expressed by the syntax data is the title of a document, is associated with the syntax data stored in the memory, and

wherein the specifying unit specifies the character string data expressing the title of the document based on the weight data stored in the memory in association with the syntax data expressing the syntax specified by the syntax analyzing unit.

3. The document processing device according to claim 2, wherein the specifying unit narrows down the character string data extracted by the extraction unit to character string data with a probability of being the title of a document, in accordance with the result of specification by the syntax analyzing unit and content stored in the memory, presents a user with this narrowed-down character string data, and specifies character string data selected by the user as the character string data expressing the title of the document.

4. The document processing device according to claim 2, wherein the specifying unit deletes, from the character string data extracted by the extraction unit, character string data that has a low probability of being the title of a document, in accordance with the result of specification by the syntax analyzing unit and content stored in the memory, presents a user with the remaining character string data, and specifies character string data selected by the user as the character string data expressing the title of the document.

5. The document processing device according to claim 1, wherein the extraction unit extracts, from among the document data obtained by analyzing the document data input by the input unit, only character string data that expresses character strings with a high probability of being a title of the document corresponding to the document data, depending on the presence or absence of formatting of the character strings corresponding to this character string data or based on distances from character strings positioned above or below those character strings.

6. A document processing method comprising:

storing in a memory, syntax data which expresses syntax of character strings whose probability of being a title of a document is high or character strings whose probability of being a title of a document is low;

inputting document data obtained by digitizing a document;

extracting character string data which expresses character strings by analyzing the input document data;

specifying a syntax of each character string contained in the document corresponding to the document data by analyzing the extracted character string data; and

specifying, from among the extracted character string data, character string data that expresses a title of the document corresponding to the document data, based on a result of the specification and content stored in the memory.

7. The document processing method according to claim 6,

wherein weight data expressing levels of probability that a character string with syntax expressed by the syntax data is the title of a document, is associated with the syntax data stored in the memory, and

wherein the character string data specifying step includes specifying the character string data expressing the title of the document based on the weight data stored in the memory in association with the syntax data expressing the specified syntax.

8. The document processing method according to claim 7, wherein the character string data specifying step includes:

narrowing down the extracted character string data to character string data with a probability of being the title of a document, in accordance with a result of the specification and content stored in the memory;

presenting a user with the narrowed-down character string data; and

specifying character string data selected by the user as the character string data expressing the title of the document.

9. The document processing method according to claim 7, wherein the character string data specifying step includes:

deleting, from the extracted character string data, character string data that has a low probability of being the title of a document, in accordance with a result of the specification and content stored in the memory;

presenting a user with remaining character string data;

specifying character string data selected by the user as the character string data expressing the title of the document.

10. The document processing method according to claim 6, wherein the extraction step includes extracting, from among the document data obtained by analyzing the input document data, only character string data that expresses character strings with a high probability of being a title of the document corresponding to the document data, depending on a presence or absence of formatting of the character strings corresponding to this character string data or based on distances from character strings positioned above or below those character strings.

11. A computer-readable storage medium recording a program for causing a computer to function as:

extraction means that, when document data obtained by digitizing a document is input, analyzes the document data and extracts character string data expressing character strings;

syntax analysis means for analyzing the character string data extracted by the extraction means and specifying the syntax of each character string contained in the document corresponding to the document data; and

specifying means for specifying, from among the character string data extracted by the extraction means, character string data that expresses a title of the document corresponding to the document data, based on results of specification by the syntax analysis means and syntax data stored in advance in the computer as data expressing the syntax of character strings whose probability of being a title of a document is high or character strings whose probability of being a title of a document is low.