Systems and methods for semantic information retrieval

• US 9,280,520 B2
• Filed: 08/02/2012
• Issued: 03/08/2016
• Est. Priority Date: 08/02/2012
• Status: Active Grant

• Current Assignee: American Express Travel Related Services Company, Inc. (American Express Company)
• Original Assignee: American Express Travel Related Services Company, Inc. (American Express Company)
• Inventors: Guenigault, Nicola J., Powell, Jordan Rhys, Chelst, Azriel L.
• Primary Examiner(s): Shah, Paras D
• Assistant Examiner(s): THOMAS-HOMESCU, ANNE L
• Application Number:
  US13/565,036
• Publication Number:
  US 20140039877A1
• Time in Patent Office: 1,314 Days
• Field of Search: 704 1- 10, 704/250, 704/240, 704/270, 704/257, 704/255, 704/232, 704/231, 704/254, 707/739, 707/706, 707/722, 707/794, 707/737, 707/769, 707/749, 707/741, 707/723, 707/705, 707/738, 707/755, 707/736, 707/746, 707/780, 707/771, 707/759, 715/230, 715/839, 715/854
• US Class Current: 1/1
• CPC Class Codes: G06F 3/14 Digital output to display device ; Cooperation and interconnection of the display device with other functional units
• CPC Class Codes: G06F 40/10 Text processing natural language analysis G06F40/20; semantic analysis G06F40/30; processing or translation of natural language G06F40/40
• CPC Class Codes: G06F 40/117 Tagging; Marking up details of markup languages G06F40/143; Designating a block; Setting of attributes style sheets, e.g. eXtensible Stylesheet Language Transformation [XSLT], G06F40/154
• CPC Class Codes: G06F 40/131 Fragmentation of text files, e.g. creating reusable text-blocks; Linking to fragments, e.g. using XInclude; Namespaces
• CPC Class Codes: G06F 40/194 Calculation of difference between files
• CPC Class Codes: G06F 40/211 Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
• CPC Class Codes: G06F 40/253 Grammatical analysis; Style critique
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates
• CPC Class Codes: G06F 40/295 Named entity recognition
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 40/35 Discourse or dialogue representation

Abstract

A semantic tagging method may add context to a sentence in order to increase search efficiency. Regardless of an author'"'"'s writing style, translating semantic concepts into tags may increase search efficiency. Automatic semantic tagging of documents may allow semantic search and reasoning. Text for semantic tagging may include an email, a website chat room, an internet forum, or a text message. Additional texts may include aggregating general consensus of an emailed topic across multiple emails, whether in the same email chain or separate emails. To increase search efficiency, the analysis of prior communications within the body of text may comprise analyzing structured contextual information to facilitate with homophora resolution. The structured contextual information may include at least one of a sender email address, one or more recipient email addresses, a subject field, a message date and time stamp, and an attachment title.

Claims

1. A method comprising:

receiving, by a computer-based system, a body of text from a data source,

wherein the body of text is an electronic text and is one of an email, a website chat room, an internet forum, or a text message;

parsing, by the computer-based system, the body of text by determining a language and structure of the body of text;

determining, by the computer-based system, a known format of the body of the text;

wherein the known format is based on the data source from which the body of text was received,

identifying, by the computer-based system and in response to the determining, structured contextual information based on the known format of the body of the text,

wherein the structured contextual information includes at least one of a sender email address, one or more recipient email addresses, a subject field, a message date and time stamp, or an attachment title;

tokenizing, by the computer-based system and in response to the parsing, the body of text by splitting the body of text into individual tokens;

generating, by the computer-based system and based on the tokenizing, a tagged body of text,

wherein the generating comprises assigning each individual token a part-of-speech tag indicating a grammatical role of the individual token,

wherein the part-of-speech tag may include custom terminology from a tagging database, and

wherein the grammatical role includes one of a noun, a pronoun, a verb, an adverb, an adjective, a conjunction, a preposition, an article, an auxiliary verb, an infinitive, an interjection, modal verb, an object, a participle, a phrase, or a predicate;

splitting, by the computer-based system, the tagged body of text into grammatical chunks;

identifying, by the computer-based system, named entities within the body of text;

resolving, by the computer-based system and based on the tokenizing, the individual tokens having a pronoun grammatical role with corresponding noun phrases;

wherein the resolving the individual tokens comprises weighting the individual tokens having a pronoun grammatical role based on the structured contextual information,

deciding, by the computer-based system and in response to the resolving, a context and purpose of the body of text,

translating, by the computer-based system and in response to the deciding, semantic concepts of the body of text into one or more semantic tags;

identifying, by the computer-based system and in response to the translating, one or more communication topics and presuppositions of the body of text,

wherein the identifying the one or more communication topics and presuppositions comprises analysis of prior communications within the body of text to facilitate the tokenizing the body of text,

wherein the analysis of prior communications within the body of text comprises, in response to the identifying the structured contextual information comprises;

analyzing structured contextual information to facilitate a homophora resolution; and

integrating, in response to the analyzing and in response to the weighting of the individual tokens having a pronoun grammatical role based on the structured contextual information, the homophora resolution into an anaphora resolution algorithm by substituting the structured contextual information into the body of text to help interpret the body of text;

generating, by the computer-based system and in response to the translating, a list of the one or more semantic tags; and

conducting, by the computer-based system, in response to the translating and using the one or more semantic tags, semantic reasoning to facilitate pattern identification within a group of documents,

wherein the pattern identification includes analyzing implied relationships of the text within the group of documents to identify a specific topic, wherein the pattern identification is based on at least one of progress or consensus of the text within the group of documents; and

displaying, by the computer-based system, in response to the conducting and to a user interface, the specific identified topic of the body of text.

2. The method of claim 1, wherein the tokenizing further comprises splitting the body of text into sentences, and further splitting each sentence into words and punctuation.

3. The method of claim 2, wherein the grammatical chunks include noun-phrase chunks and pronoun chunks.

4. The method of claim 3, wherein the identifying named entities within the body of text is configured to identify people, organizations, and locations from the body of text.

5. The method of claim 4, wherein the generating the one or more semantic tags includes the one or more semantic tags being at least one of text topic and ideas, authors, people mentioned in text, document type or place of origin.

6. The method of claim 5, further comprising linking the one or more semantic tags into a network of one or more semantic tags.

7. The method of claim 6, wherein the tagging accuracy is 95% or greater.

8. The method of claim 7, wherein weighting the individual tokens includes giving more importance to the structured contextual information.

9. The method of claim 8, wherein the body of text is from a data source.

12. The method of claim 1, further comprising identifying a header and an email body of an email, wherein the body of text is the email, and wherein the known format is a Multipurpose Internet Mail Extension (MIME) format.

10. An article of manufacture including a non-transitory, tangible computer readable storage medium having instructions stored thereon that, in response to execution by a computer-based system, cause the computer-based system to perform operations comprising:

receiving, by the computer-based system, a body of text from a data source,

wherein the body of text is an electronic text and is one of an email, a website chat room, an internet forum, or a text message;

parsing, by the computer-based system, the body of text by determining a language and structure of the body of text;

determining, by the computer-based system, a known format of the body of the text;

wherein the known format is based on the data source from which the body of text was received,

identifying, by the computer-based system and in response to the determining, structured contextual information based on the known format of the body of the text,

wherein the structured contextual information includes at least one of a sender email address, one or more recipient email addresses, a subject field, a message date and time stamp, or an attachment title;

tokenizing, by the computer-based system and in response to the parsing, the body of text by splitting the body of text into individual tokens;

generating, by the computer-based system and based on the tokenizing, a tagged body of text,

wherein the generating comprises assigning each individual token a part-of-speech tag indicating a grammatical role of the individual token,

wherein the part-of-speech tag may include custom terminology from a tagging database, and

wherein the grammatical role includes one of a noun, a pronoun, a verb, an adverb, an adjective, a conjunction, a preposition, an article, an auxiliary verb, an infinitive, an interjection, modal verb, an object, a participle, a phrase, or a predicate;

splitting, by the computer-based system, the tagged body of text into grammatical chunks;

identifying, by the computer-based system, named entities within the body of text;

resolving, by the computer-based system and based on the tokenizing, the individual tokens having a pronoun grammatical role with corresponding noun phrases;

wherein the resolving the individual tokens comprises weighting the individual tokens having a pronoun grammatical role based on the structured contextual information,

deciding, by the computer-based system and in response to the resolving, a context and purpose of the body of text,

translating, by the computer-based system and in response to the deciding, semantic concepts of the body of text into one or more semantic tags;

identifying, by the computer-based system and in response to the translating, one or more communication topics and presuppositions of the body of text,

wherein the identifying the one or more communication topics and presuppositions comprises analysis of prior communications within the body of text to facilitate the tokenizing the body of text,

wherein the analysis of prior communications within the body of text comprises, in response to the identifying the structured contextual information comprises;

analyzing structured contextual information to facilitate a homophora resolution; and

integrating, in response to the analyzing and in response to the weighting of the individual tokens having a pronoun grammatical role based on the structured contextual information, the homophora resolution into an anaphora resolution algorithm by substituting the structured contextual information into the body of text to help interpret the body of text;

generating, by the computer-based system and in response to the translating, a list of the one or more semantic tags; and

conducting, by the computer-based system, in response to the translating and using the one or more semantic tags, semantic reasoning to facilitate pattern identification within a group of documents,

wherein the pattern identification includes analyzing implied relationships of the text within the group of documents to identify a specific topic, wherein the pattern identification is based on at least one of progress or consensus of the text within the group of documents; and

displaying, by the computer-based system, in response to the conducting and to a user interface, the specific identified topic of the body of text.

11. A system comprising:

a tangible, non-transitory memory communicating with a processor,

the tangible, non-transitory memory having instructions stored thereon that, in response to execution by the processor, cause the processor to perform operations comprising;

receiving, by the processor, a body of text from a data source,

wherein the body of text is an electronic text and is one of an email, a website chat room, an internet forum, or a text message;

parsing, by the processor, the body of text by determining a language and structure of the body of text;

determining, by the processor, a known format of the body of the text;

wherein the known format is based on the data source from which the body of text was received,

identifying, by the processor and in response to the determining, structured contextual information based on the known format of the body of the text,

wherein the structured contextual information includes at least one of a sender email address, one or more recipient email addresses, a subject field, a message date and time stamp, or an attachment title;

tokenizing, by the processor and in response to the parsing, the body of text by splitting the body of text into individual tokens;

generating, by the processor and based on the tokenizing, a tagged body of text,

wherein the generating comprises assigning each individual token a part-of-speech tag indicating a grammatical role of the individual token,

wherein the part-of-speech tag may include custom terminology from a tagging database, and

wherein the grammatical role includes one of a noun, a pronoun, a verb, an adverb, an adjective, a conjunction, a preposition, an article, an auxiliary verb, an infinitive, an interjection, modal verb, an object, a participle, a phrase, or a predicate;

splitting, by the processor, the tagged body of text into grammatical chunks;

identifying, by the processor, named entities within the body of text;

resolving, by the processor and based on the tokenizing, the individual tokens having a pronoun grammatical role with corresponding noun phrases;

wherein the resolving the individual tokens comprises weighting the individual tokens having a pronoun grammatical role based on the structured contextual information,

deciding, by the processor and in response to the resolving, a context and purpose of the body of text,

translating, by the processor and in response to the deciding, semantic concepts of the body of text into one or more semantic tags;

identifying, by the processor and in response to the translating, one or more communication topics and presuppositions of the body of text,

wherein the identifying the one or more communication topics and presuppositions comprises analysis of prior communications within the body of text to facilitate the tokenizing the body of text,

wherein the analysis of prior communications within the body of text comprises, in response to the identifying the structured contextual information comprises;

analyzing structured contextual information to facilitate a homophora resolution; and

integrating, in response to the analyzing and in response to the weighting of the individual tokens having a pronoun grammatical role based on the structured contextual information, the homophora resolution into an anaphora resolution algorithm by substituting the structured contextual information into the body of text to help interpret the body of text;

generating, by the processor and in response to the translating, a list of the one or more semantic tags; and

conducting, by the processor, in response to the translating and using the one or more semantic tags, semantic reasoning to facilitate pattern identification within a group of documents,

wherein the pattern identification includes analyzing implied relationships of the text within the group of documents to identify a specific topic, wherein the pattern identification is based on at least one of progress or consensus of the text within the group of documents; and

displaying, by the processor, in response to the conducting and to a user interface, the specific identified topic of the body of text.