Granted Patent B2

US 12,332,944 · App. 17/230,475 · Granted Jun 17, 2025

Inventors: June-Ray Lin (Taipei, TW); Nan Chen (Beijing, CN); Ju Ling Liu (Beijing, CN); Li Na Wang (Beijing, CN); Shun Xian Wu (Beijing, CN)

Assignee: International Business Machines Corporation

G06F16/93G06F16/90332G06F16/9035G06F16/9038G06F18/214G06N3/08

Abstract

Claims (77)

1. A computer-implemented method for identifying equivalent technical terms, the method comprising:

training a deep learning model to identify equivalent technical terms using training data comprising structured and unstructured data comprising documents annotated with technical terms that are identified as being equivalent, wherein said equivalent technical terms correspond to technical terms having a same semantic meaning within a threshold degree, wherein said training of said deep learning model to identify equivalent technical terms comprises:

annotating technical terms in text of a first document regarding a first product and a second document regarding a second product with tags corresponding to matching technical terms listed in a data structure; and

annotating said tagged technical terms with entity types using an entity identification engine to assist said deep learning model in identifying equivalent technical terms, wherein said entity identification engine provides components for term or entity detection using maximum entropy models trained from annotated data, wherein said entity identification engine further provides a trainable co-reference component for grouping detected terms in a document that correspond to a same entity, and a trainable relation extraction system;

applying said deep learning model to a first document;

analyzing each sentence of said first document to identify technical terms;

analyzing text preceding or succeeding a first technical term identified in said first document to determine a meaning of said text preceding or succeeding said first technical term;

reviewing a glossary list to determine if said meaning of said analyzed text preceding or succeeding said first technical term in said first document matches a meaning in said glossary list linked to a second technical term, wherein said glossary list comprises a list of meanings associated with equivalent technical terms in designated products;

identifying said second technical term equivalent to said first technical term from said glossary list in response to determining said meaning of said analyzed text preceding or succeeding said first technical term matching said meaning in said glossary list linked to said second technical term; and

annotating said first document by replacing said first technical term with said second technical term or tagging said first technical term with said second technical term.

2. The method as recited in claim 1 , wherein said meaning of said analyzed text preceding or succeeding said first technical term is deemed to match said meaning in said glossary list linked to said second technical term in response to said meaning of said analyzed text preceding or succeeding said first technical term having a semantic similarity to said meaning in said glossary list linked to said second technical term that exceeds a threshold value.

3. The method as recited in claim 1 , wherein said training of said deep learning model to identify equivalent technical terms further comprises:

analyzing text preceding or succeeding said tag annotations of said technical terms to identify a meaning of said analyzed text, wherein said meaning of said analyzed text is determined using part-of-speech tagging using a stochastic part-of-speech tagging algorithm.

4. The method as recited in claim 1 , wherein said training of said deep learning model to identify equivalent technical terms further comprises:

analyzing text preceding or succeeding said tag annotations of said technical terms to identify a meaning of said analyzed text, wherein said meaning of said analyzed text is determined using part-of-speech tagging using a rule-based tagging algorithm.

5. The method as recited in claim 1 , wherein said training of said deep learning model to identify equivalent technical terms further comprises:

receiving a second document regarding a second product and a third document regarding a third product;

annotating technical terms in text of said second and third documents with tags;

receiving further annotations of technical terms in said text of said second and third documents with tags from a user;

analyzing text preceding or succeeding said tag annotations of said technical terms in said second and third documents to determine a meaning of said text preceding or succeeding said tag annotations of said technical terms in said second and third documents;

identifying a third technical term in said second document with a same meaning as a fourth technical term in said third document; and

adding said third and fourth technical terms in said glossary list as being equivalent technical terms;

wherein said third technical term is deemed to have said same meaning as said fourth technical term in response to a meaning of said analyzed text preceding or succeeding said third technical term having a semantic similarity to a meaning of said analyzed text surrounding said fourth technical term that exceeds a threshold value.

6. The method as recited in claim 1 , wherein said entity identification engine provides a syntactic breakdown of said text in said first and second documents, identification of entities detected in said text, co-reference resolution, and relationship detection between entities, wherein said entity identification engine provides logic for searching indices augmented with co-referenced terms discovered by said entity identification engine and for generating a relational database that stores entities found by said entity identification engine, an entity type, terms related to said entities, a frequency of a particular relationship within a corpus or across corpora, a time of publication of a document that is a source of said relationship, a corpus identifier, and a document identifier of a document that is said source of said relationship.

7. A computer program product for identifying equivalent technical terms, the computer program product comprising one or more computer readable storage mediums having program code embodied therewith, the program code comprising programming instructions for:

training a deep learning model to identify equivalent technical terms using training data comprising structured and unstructured data comprising documents annotated with technical terms that are identified as being equivalent, wherein said equivalent technical terms correspond to technical terms having a same semantic meaning within a threshold degree, wherein said training of said deep learning model to identify equivalent technical terms comprises:

annotating technical terms in text of a first document regarding a first product and a second document regarding a second product with tags corresponding to matching technical terms listed in a data structure; and

annotating said tagged technical terms with entity types using an entity identification engine to assist said deep learning model in identifying equivalent technical terms, wherein said entity identification engine provides components for term or entity detection using maximum entropy models trained from annotated data, wherein said entity identification engine further provides a trainable co-reference component for grouping detected terms in a document that correspond to a same entity, and a trainable relation extraction system;

applying said deep learning model to a first document;

analyzing each sentence of said first document to identify technical terms;

analyzing text preceding or succeeding a first technical term identified in said first document to determine a meaning of said text preceding or succeeding said first technical term;

reviewing a glossary list to determine if said meaning of said analyzed text preceding or succeeding said first technical term in said first document matches a meaning in said glossary list linked to a second technical term, wherein said glossary list comprises a list of meanings associated with equivalent technical terms in designated products;

identifying said second technical term equivalent to said first technical term from said glossary list in response to determining said meaning of said analyzed text preceding or succeeding said first technical term matching said meaning in said glossary list linked to said second technical term; and

annotating said first document by replacing said first technical term with said second technical term or tagging said first technical term with said second technical term.

8. The computer program product as recited in claim 7 , wherein said meaning of said analyzed text preceding or succeeding said first technical term is deemed to match said meaning in said glossary list linked to said second technical term in response to said meaning of said analyzed text preceding or succeeding said first technical term having a semantic similarity to said meaning in said glossary list linked to said second technical term that exceeds a threshold value.

9. The computer program product as recited in claim 7 ,

wherein said training of said deep learning model to identify equivalent technical terms further comprises the programming instructions for:

analyzing text preceding or succeeding said tag annotations of said technical terms to identify a meaning of said analyzed text, wherein said meaning of said analyzed text is determined using part-of-speech tagging using a stochastic part-of-speech tagging algorithm.

10. The computer program product as recited in claim 7 ,

wherein said training of said deep learning model to identify equivalent technical terms further comprises the programming instructions for:

analyzing text preceding or succeeding said tag annotations of said technical terms to identify a meaning of said analyzed text, wherein said meaning of said analyzed text is determined using part-of-speech tagging using a rule-based tagging algorithm.

11. The computer program product as recited in claim 7 , wherein said training of said deep learning model to identify equivalent technical terms further comprises the programming instructions for:

receiving a second document regarding a second product and a third document regarding a third product;

annotating technical terms in text of said second and third documents with tags;

receiving further annotations of technical terms in said text of said second and third documents with tags from a user;

analyzing text preceding or succeeding said tag annotations of said technical terms in said second and third documents to determine a meaning of said text preceding or succeeding said tag annotations of said technical terms in said second and third documents;

identifying a third technical term in said second document with a same meaning as a fourth technical term in said third document; and

adding said third and fourth technical terms in said glossary list as being equivalent technical terms;

wherein said third technical term is deemed to have said same meaning as said fourth technical term in response to a meaning of said analyzed text preceding or succeeding said third technical term having a semantic similarity to a meaning of said analyzed text surrounding said fourth technical term that exceeds a threshold value.

12. The computer program product as recited in claim 7 , wherein said entity identification engine provides a syntactic breakdown of said text in said first and second documents, identification of entities detected in said text, co-reference resolution, and relationship detection between entities, wherein said entity identification engine provides logic for searching indices augmented with co-referenced terms discovered by said entity identification engine and for generating a relational database that stores entities found by said entity identification engine, an entity type, terms related to said entities, a frequency of a particular relationship within a corpus or across corpora, a time of publication of a document that is a source of said relationship, a corpus identifier, and a document identifier of a document that is said source of said relationship.

13. A system, comprising:

a memory for storing a computer program for identifying equivalent technical terms; and

a processor connected to said memory, wherein said processor is configured to execute program instructions of the computer program comprising:

training a deep learning model to identify equivalent technical terms using training data comprising structured and unstructured data comprising documents annotated with technical terms that are identified as being equivalent, wherein said equivalent technical terms correspond to technical terms having a same semantic meaning within a threshold degree, wherein said training of said deep learning model to identify equivalent technical terms comprises:

annotating technical terms in text of a first document regarding a first product and a second document regarding a second product with tags corresponding to matching technical terms listed in a data structure; and

annotating said tagged technical terms with entity types using an entity identification engine to assist said deep learning model in identifying equivalent technical terms, wherein said entity identification engine provides components for term or entity detection using maximum entropy models trained from annotated data, wherein said entity identification engine further provides a trainable co-reference component for grouping detected terms in a document that correspond to a same entity, and a trainable relation extraction system;

applying said deep learning model to a first document;

analyzing each sentence of said first document to identify technical terms;

analyzing text preceding or succeeding a first technical term identified in said first document to determine a meaning of said text preceding or succeeding said first technical term;

reviewing a glossary list to determine if said meaning of said analyzed text preceding or succeeding said first technical term in said first document matches a meaning in said glossary list linked to a second technical term, wherein said glossary list comprises a list of meanings associated with equivalent technical terms in designated products;

identifying said second technical term equivalent to said first technical term from said glossary list in response to determining said meaning of said analyzed text preceding or succeeding said first technical term matching said meaning in said glossary list linked to said second technical term; and

annotating said first document by replacing said first technical term with said second technical term or tagging said first technical term with said second technical term.

14. The system as recited in claim 13 , wherein said meaning of said analyzed text preceding or succeeding said first technical term is deemed to match said meaning in said glossary list linked to said second technical term in response to said meaning of said analyzed text preceding or succeeding said first technical term having a semantic similarity to said meaning in said glossary list linked to said second technical term that exceeds a threshold value.

15. The system as recited in claim 13 ,

wherein the program instructions of said training of said deep learning model to identify equivalent technical terms further comprises:

analyzing text preceding or succeeding said tag annotations of said technical terms to identify a meaning of said analyzed text, wherein said meaning of said analyzed text is determined using part-of-speech tagging using a stochastic part-of-speech tagging algorithm.

16. The system as recited in claim 13 ,

wherein the program instructions of said training of said deep learning model to identify equivalent technical terms further comprises:

analyzing text preceding or succeeding said tag annotations of said technical terms to identify a meaning of said analyzed text, wherein said meaning of said analyzed text is determined using part-of-speech tagging using a rule-based tagging algorithm.

17. The system as recited in claim 13 , wherein the program instructions of said training of said deep learning model to identify equivalent technical terms further comprises:

receiving a second document regarding a second product and a third document regarding a third product;

annotating technical terms in text of said second and third documents with tags;

receiving further annotations of technical terms in said text of said second and third documents with tags from a user;

analyzing text preceding or succeeding said tag annotations of said technical terms in said second and third documents to determine a meaning of said text preceding or succeeding said tag annotations of said technical terms in said second and third documents;

identifying a third technical term in said second document with a same meaning as a fourth technical term in said third document; and

adding said third and fourth technical terms in said glossary list as being equivalent technical terms;

wherein said third technical term is deemed to have said same meaning as said fourth technical term in response to a meaning of said analyzed text preceding or succeeding said third technical term having a semantic similarity to a meaning of said analyzed text surrounding said fourth technical term that exceeds a threshold value.

Assignments (1)

Continuity (1)

References Cited (19)