Analyzing concepts over time

Patent metadata

Field	Value
Publication number	US-10628507-B2
Application number	US-201816119511-A
Country	US
Kind code	B2
Filing date	Aug 31, 2018
Priority date	Sep 22, 2015
Publication date	Apr 21, 2020
Grant date	Apr 21, 2020

Abstract

A method and apparatus are provided for automatically generating and processing first and second concept vector sets extracted, respectively, from a first set of concept sequences and from a second, temporally separated, concept sequences by performing a natural language processing (NLP) analysis of the first concept vector set and second concept vector set to detect changes in the corpus over time by identifying changes for one or more concepts included in the first and/or second set of concept sequences.

First claim

What is claimed is:

1. A method, in an information handling system comprising a processor and a memory, for analyzing concept vectors to detect changes in a corpus over time, the method comprising:
identifying a first set of concept sequences and a second set of concept sequences from the corpus;
generating concept vectors from the first and from the second sets of concept sequences using a neutral network-based vector embedding method, matrix-based vector embedding method, log-linear classifier-based-based vector embedding method or word2vec method; and
analyzing relationship strengths between concepts that persist in the first set of concept sequences and the second set of concept sequences to identify vector changes for one or more concepts included in the first and/or second set of concept sequences, wherein analyzing relationship strengths comprises:
computing, by the system, a first cosine distance between each vector pair Vi, Vj from a first concept vector set V 1 , . . . , Vk derived from the first set of concept sequences over k concepts for all i≠j, 1≤i, j≤k;
computing, by the system, a second cosine distance between each vector pair Vi, V′j from a second concept vector set V′ 1 , . . . , V′k+b derived from the second set of concept sequences over k old and b new concepts for all i≠j, 1≤i, j≤k; and
identifying concept pairs from the first set of concept sequences whose interrelationship has changed by reporting each concept pair Vi, Vj whereby a subtraction of the second cosine distance from the first cosine distance exceeds a first specified reporting threshold.

2. The method of claim 1 , further comprising generating, by the system, the first concept vector set V 1 , . . . , Vk that is derived from the first set of concept sequences over k concepts that are extracted from the corpus and applied to a vector learning component.

3. The method of claim 1 , further generating, by the system, the second concept vector set V′ 1 , . . . , V′k+b that is derived from the second set of concept sequences over k old and b new concepts that are extracted from the corpus and applied to the vector learning component, where the second set of concept sequences is effectively collected after collection of the first set of concept sequences.

4. The method of claim 1 , wherein analyzing relationship strengths comprises performing natural language processing (NLP) analysis of the first and second concept vector sets to detect an appearance of one or more new concepts in the second set of concept sequences that are not present in the first set of concept sequences.

5. The method of claim 4 , wherein detecting the appearance of one or more new concepts comprises:
computing, by the system, a third cosine distance between each vector pair V′i, V′j from a third concept vector set V′ 1 , . . . V′k, V′k+1, . . . , V′k+b derived from a concatenation of the first set of concept sequences over k concepts and a second set of concept sequences over k old and b new concepts for 1<i<k and k<j≤k+b; and
identifying new concept pairs from the second set of concept sequences over k old and b new concepts having a strong interrelationship with concepts in the first set of concept sequences by reporting each concept pair V′i, V′j whereby the third first cosine distance exceeds a second specified reporting threshold.

6. The method of claim 1 , wherein analyzing relationship strengths comprises performing natural language processing (NLP) analysis of the first and second concept vector sets to detect a disappearance of one or more old concepts from the first set of concept sequences that are not present in the second set of concept sequences.

7. The method of claim 1 , wherein analyzing relationship strengths comprises performing natural language processing (NLP) analysis of the first and second concept vector sets to detect an appearance of one or more disruptive concepts in the second set of concept sequences that are related to a specified technology area represented by a sum of a plurality of concept vectors.

8. The method of claim 1 , wherein analyzing relationship strengths comprises performing natural language processing (NLP) analysis of the first and second concept vector sets to detect an appearance of one or more emerging concepts in the second set of concept sequences that are related to a specified topic area.

9. The method of claim 1 , wherein analyzing relationship strengths comprises performing natural language processing (NLP) analysis of the first and second concept vector sets to detect differences in spatial and/or frequency distributions of first and second concept vector sets by identifying changes in values of quantitative geometry and topology features that characterize concept regions associated, respectively, with the first and second concept vector sets.

10. The method of claim 9 , wherein identifying changes comprises computing differences in centroid positions, diameters between extreme points, orientations of the principal axes, number of significant dimensions, aspect ratios between lengths of the principal axes, or number and sizes of clusters computed by standard clustering algorithms.

11. An information handling system comprising:
one or more processors;
a memory coupled to at least one of the processors;
a set of instructions stored in the memory and executed by at least one of the processors to analyze concept vectors to detect changes in a corpus over time, wherein the set of instructions are executable to perform actions of:
identifying, by the system, a first set of concept sequences and a second set of concept sequences from the corpus;
generating, by the system, concept vectors from the first and from the second sets of concept sequences using a neutral network-based vector embedding method, matrix-based vector embedding method, log-linear classifier-based-based vector embedding method or word2vec method; and
analyzing, by the system, relationship strengths between concepts that persist in the first set of concept sequences and the second set of concept sequences to identify vector changes for one or more concepts included in the first and/or second set of concept sequences, wherein analyzing relationship strengths comprises:
computing, by the system, a first cosine distance between each vector pair Vi, Vj from a first concept vector set V 1 , . . . , Vk derived from the first set of concept sequences over k concepts for all i≠j, 1≤i, j≤k;
computing, by the system, a second cosine distance between each vector pair V′i, V′j from a second concept vector set V′ 1 , . . . , V′k+b derived from the second set of concept sequences over k old and b new concepts for all i≠j, 1≤i, j≤k; and
identifying concept pairs from the first set of concept sequences whose interrelationship has changed by reporting each concept pair Vi, Vj whereby a subtraction of the second cosine distance from the first cosine distance exceeds a first specified reporting threshold.

12. The information handling system of claim 11 , wherein the set of instructions are executable to generate the first concept vector set V 1 , . . . , Vk that is derived from the first set of concept sequences over k concepts that are extracted from the corpus and applied to a vector learning component.

13. The information handling system of claim 11 , wherein the set of instructions are executable to generate the second concept vector set V′ 1 , . . . , V′k+b that is derived from the second set of concept sequences over k old and b new concepts that are extracted from the corpus and applied to the vector learning component, where the second set of concept sequences is effectively collected after col

Assignees

• IBM

Inventors

• Ho Tin Kam
• Lastras-Montano Luis A
• Shmueli Oded

Classifications

• G06F16/243

  Natural language query formulation · CPC title

• G06F16/3344

  using natural language analysis · CPC title

• G06F16/3347

  using vector based model · CPC title

• G06Q30/0203 Primary

  Market surveys; Market polls · CPC title

• G06F16/2237

  Vectors, bitmaps or matrices · CPC title