# US 2020/0401659 A1 · Contextual disambiguation of an entity in a conversation management system

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2020/0401659 A1
- Document ID: US20200401659A1
- Status: Published application
- Kind: A1
- Application number: 16/444,801
- Filed: 2019-06-18
- Published: 2020-12-24
- Earliest claimed: 2019-06-18
- Assignees: International Business Machines Corporation
- Applicants: International Business Machines Corporation
- Inventors: Sunhwan Lee; Shun Jiang; Chung-hao Tan; Lei Huang; Pawan Chowdhary
- Web page: https://patentgrep.com/patent/US20200401659A1

## CPC

- G06F 40/295: Named entity recognition
- G06F 40/30: Semantic analysis
- G06N 20/00: Machine learning
- G10L 25/27: characterised by the analysis technique
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING

## Abstract

Systems, computer-implemented methods, and computer program products that can facilitate word entity disambiguation are provided. According to an embodiment, a system can comprise a memory that stores computer executable components and a processor that executes the computer executable components stored in the memory. The computer executable components can comprise a language model component that employs an artificial intelligence model to generate a profile vector of an entity based on one or more binary values representing profile data of the entity and a word vector of a word entity in a dialogue based on one or more second word entities adjacent to the word entity in the dialogue. The computer executable components can further comprise a dialogue management component that disambiguates the word entity based on the profile vector and the word vector.

## Claims

### Claim 1

A system, comprising:
a memory that stores computer executable components; and
a processor that executes the computer executable components stored in the memory, wherein the computer executable components comprise:
a language model component that employs an artificial intelligence model to generate a profile vector of an entity based on one or more binary values representing profile data of the entity and a word vector of a word entity in a dialogue based on one or more second word entities adjacent to the word entity in the dialogue; and
a dialogue management component that disambiguates the word entity based on the profile vector and the word vector.

### Claim 2 (depends on 1)

The system of claim 1, wherein the computer executable components further comprise:
a data collection component that collects at least one of the profile data of the entity or one or more conversation logs comprising at least one of the dialogue, the word entity, or the one or more second word entities.

### Claim 3 (depends on 1)

The system of claim 1, wherein the dialogue management component disambiguates the word entity based on at least one of a distance between the profile vector and the word vector or distances between the profile vector and multiple word vectors.

### Claim 4 (depends on 1)

The system of claim 1, wherein the artificial intelligence model comprises at least one of a language model, a document to vector model, a word to vector model, a natural language model, a term vector model, a vector space model, a natural language processing model, a statistical language model, or a feature learning model.

### Claim 5 (depends on 1)

The system of claim 1, wherein the word entity comprises one or more textual objects selected from a group consisting of a phoneme, a syllable, a letter, a word, a sentence, and a paragraph.

### Claim 6 (depends on 1)

The system of claim 1, wherein the dialogue comprises at least one of a sequence of word entities, a sequence of textual objects, a textual dialogue, a textual conversation, or a corpus of conversation log data.

### Claim 7 (depends on 1)

The system of claim 1, wherein the computer executable components further comprise:
an interface component that presents at least one of a word entity variation or a word entity disambiguation of the word entity and receives feedback data corresponding to at least one of the word entity variation or the word entity disambiguation.

### Claim 8 (depends on 1)

The system of claim 1, wherein the computer executable components further comprise:
a learner component that learns one or more word entity variations of the word entity based on feedback data corresponding to at least one of a word entity variation or a word entity disambiguation of the word entity, thereby facilitating at least one of improved accuracy, efficiency, or performance of at least one of the dialogue management component or the processor.

### Claim 9

A computer-implemented method, comprising:
employing, by a system operatively coupled to a processor, an artificial intelligence model to generate a profile vector of an entity based on one or more binary values representing profile data of the entity and a word vector of a word entity in a dialogue based on one or more second word entities adjacent to the word entity in the dialogue; and
disambiguating, by the system, the word entity based on the profile vector and the word vector.

### Claim 10 (depends on 9)

The computer-implemented method of claim 9, further comprising:
collecting, by the system, at least one of the profile data of the entity or one or more conversation logs comprising at least one of the dialogue, the word entity, or the one or more second word entities.

### Claim 11 (depends on 9)

The computer-implemented method of claim 9, wherein the disambiguating comprises:
disambiguating, by the system, the word entity based on at least one of a distance between the profile vector and the word vector or distances between the profile vector and multiple word vectors.

### Claim 12 (depends on 9)

The computer-implemented method of claim 9, wherein the employing comprises:
employing, by the system, at least one of a language model, a document to vector model, a word to vector model, a natural language model, a term vector model, a vector space model, a natural language processing model, a statistical language model, or a feature learning model to generate the profile vector of the entity based on the one or more binary values representing the profile data of the entity and the word vector of the word entity in the dialogue based on the one or more second word entities adjacent to the word entity in the dialogue.

### Claim 13 (depends on 9)

The computer-implemented method of claim 9, further comprising:
presenting, by the system, at least one of a word entity variation or a word entity disambiguation of the word entity; and
receiving, by the system, feedback data corresponding to at least one of the word entity variation or the word entity disambiguation.

### Claim 14 (depends on 9)

The computer-implemented method of claim 9, further comprising:
learning, by the system, one or more word entity variations of the word entity based on feedback data corresponding to at least one of a word entity variation or a word entity disambiguation of the word entity, thereby facilitating at least one of improved accuracy, efficiency, or performance of at least one of the system or the processor.

### Claim 15

A computer program product facilitating word entity disambiguation, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to:
employ, by the processor, an artificial intelligence model to generate a profile vector of an entity based on one or more binary values representing profile data of the entity and a word vector of a word entity in a dialogue based on one or more second word entities adjacent to the word entity in the dialogue; and
disambiguate, by the processor, the word entity based on the profile vector and the word vector.

### Claim 16 (depends on 15)

The computer program product of claim 15, wherein the program instructions are further executable by the processor to cause the processor to:
collect, by the processor, at least one of the profile data of the entity or one or more conversation logs comprising at least one of the dialogue, the word entity, or the one or more second word entities.

### Claim 17 (depends on 15)

The computer program product of claim 15, wherein the program instructions are further executable by the processor to cause the processor to:
disambiguate, by the processor, the word entity based on at least one of a distance between the profile vector and the word vector or distances between the profile vector and multiple word vectors.

### Claim 18 (depends on 15)

The computer program product of claim 15, wherein the program instructions are further executable by the processor to cause the processor to:
employ, by the processor, at least one of a language model, a document to vector model, a word to vector model, a natural language model, a term vector model, a vector space model, a natural language processing model, a statistical language model, or a feature learning model to generate the profile vector of the entity based on the one or more binary values representing the profile data of the entity and the word vector of the word entity in the dialogue based on the one or more second word entities adjacent to the word entity in the dialogue.

### Claim 19 (depends on 15)

The computer program product of claim 15, wherein the program instructions are further executable by the processor to cause the processor to:
present, by the processor, at least one of a word entity variation or a word entity disambiguation of the word entity; and
receive, by the processor, feedback data corresponding to at least one of the word entity variation or the word entity disambiguation.

### Claim 20 (depends on 15)

The computer program product of claim 15, wherein the program instructions are further executable by the processor to cause the processor to:
learn, by the processor, one or more word entity variations of the word entity based on feedback data corresponding to at least one of a word entity variation or a word entity disambiguation of the word entity.
