# US 12,216,987 B2 · Generating heading based on extracted feature words > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,216,987 B2 - Document ID: US12216987B2 - Status: Granted - Kind: B2 - Application number: 18/037,889 - Filed: 2020-11-25 - Published: 2023-12-21 - Granted: 2025-02-04 - Earliest claimed: 2020-11-25 - Assignees: NEC Corporation - Applicants: NEC Corporation - Inventors: Ayako Hoshino - Web page: https://patentgrep.com/patent/US12216987B2 ## CPC - G06F 40/166: Editing, e.g. inserting or deleting - G06F 40/103: Formatting, i.e. changing of presentation of documents (automatic justification G06F40/189; automatic line break hyphenation G06F40/191) - G06F 40/126: Character encoding - G06F 40/258: Heading extraction; Automatic titling; Numbering - G06N 20/00: Machine learning - G06V 30/19127: {Extracting features by transforming the feature space, e.g. multidimensional scaling; Mappings, e.g. subspace methods} - G06V 30/416: Extracting the logical structure, e.g. chapters, sections or page numbers; Identifying elements of the document, e.g. authors - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract The information processing device generates headings from structured documents. The acquisition means acquires a structured document including headings and texts. The feature word extraction means extracts feature words from subordinate elements of the heading for the headings included in the structured document. The heading generation means generates a new heading corresponding to the subordinate elements based on the extracted feature words. ## Claims ### Claim 1 An information processing device comprising: a memory storing instructions; and one or more processors configured to execute the instructions to: acquire a structured document including a heading and text; generate, for the heading included in the structured document, a matrix of frequency of occurrence of words appearing in documents of subordinate elements of the heading; extract, for the heading, feature words by reducing dimensions of the words appearing in the documents using principal component analysis (PCA); generate a new heading based on the extracted feature words; and generate a corrected structured document by replacing the heading with the new heading. ### Claim 2 (depends on 1) The information processing device according to claim 1, wherein the subordinate elements include at least one of subordinate headings below the heading and the text below the heading in the structured document. ### Claim 3 (depends on 1) The information processing device according to claim 1, wherein the one or more processors extract the feature words using a TF-IDF value. ### Claim 4 (depends on 1) The information processing device according to claim 1, wherein the one or more processors are further configured to convert the words appearing in the documents into vectors, wherein the one or more processors reduce the dimensions of the words by reducing the dimension of the vectors using autoencoding, and wherein the one or more processors extract the feature words by converting the vectors with the reduced dimensions into the feature words. ### Claim 5 (depends on 1) The information processing device according to claim 1, wherein the one or more processors are further configured to detect, as the heading, an inappropriate heading from a plurality of headings included in the structured document. ### Claim 6 (depends on 5) The information processing device according to claim 5, wherein the one or more processors generate the corrected structured document by replacing the inappropriate heading in the document with the new heading. ### Claim 7 (depends on 5) The information processing device according to claim 5, wherein the one or more processors generate the corrected structured document by adding at least a part of the new heading to the inappropriate heading in the document. ### Claim 8 (depends on 5) The information processing device according to claim 5, wherein the inappropriate heading is that of a character string identical to another heading in a parallel relationship in the structured document. ### Claim 9 (depends on 5) The information processing device according to claim 5, wherein the inappropriate heading is in a parallel relationship in the structured document and includes a number or a symbol without meaning or content. ### Claim 10 (depends on 1) The information processing device according to claim 1, wherein the one or more processors are further configured to convert an inputted document into the structured document. ### Claim 11 An information processing method performed by a computer and comprising: acquiring a structured document including a heading and text; generating, for the heading included in the structured document, a matrix of frequency of occurrence of words appearing in documents of subordinate elements of the heading; extracting, for the heading, feature words by reducing dimensions of the words appearing in the documents using principal component analysis (PCA); generating a new heading based on the extracted feature words; and generating a corrected document by replacing the heading with the new heading. ### Claim 12 A non-transitory computer-readable recording medium storing a program executable by a computer to perform processing comprising: acquiring a structured document including a heading and text; generating, for the heading included in the structured document, a matrix of frequency of occurrence of words appearing in documents of subordinate elements of the heading; extracting, for the heading, feature words by reducing dimensions of the words appearing in the documents using principal component analysis (PCA); generating a new heading based on the extracted feature words; and generating a corrected document by replacing the heading with the new heading.