Context-Based Personality Trait Inference in Dialogues

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for determining personality traits based on dialogues are inaccurate as they only consider the utterances of one participant, neglecting the influence of the other participant's language and context, which can alter the language used by the first participant.

Innovation Solution

A computer-implemented method and system that generate features based on both the first and second interlocutors' utterances in a dialog, considering the context of the conversation to determine the personality of the first interlocutor more accurately by using neural network-based models and psycholinguistic analytics.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If only the first interlocutor's utterance is considered for personality determination, then the analysis process is simple, but the accuracy of personality determination deteriorates

Engineering Contradiction:

Improveanalysis process complexityVSAvoidpersonality determination accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent merges the first interlocutor's utterance with the second interlocutor's utterance to form a combined input for personality determination. This combination allows the system to capture both the target person's language characteristics and the contextual influence from the conversation partner, thereby improving measurement precision without excessive complexity increase.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The second interlocutor's utterance serves as an intermediary element that provides contextual information about the conversation environment. This intermediary data helps distinguish between personality-driven language patterns and situation-driven language patterns, improving the accuracy of personality determination.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Ease of manufacture

If only the first interlocutor's utterance is used, then the data processing is straightforward, but the reflection of actual personality deteriorates

Engineering Contradiction:

Improvedata processing simplicityVSAvoidpersonality reflection accuracy

Core Design Contradiction:

Ease of manufactureVSReliability

Solution Approach 1:

The patent segments the language analysis into two distinct components: the first interlocutor's utterance (target language patterns) and the second interlocutor's utterance (contextual influence). This segmentation allows for systematic processing of each component's specific features while maintaining overall reliability in personality assessment.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent adds another dimension to the analysis by incorporating the second interlocutor's utterance as a separate input layer. This dimensional expansion transforms the analysis from a single-source evaluation to a multi-source evaluation, improving reliability by capturing both personality traits and situational context.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS10572585B2Context-based linguistic analytics in dialogues

Publication Date: 2020.02.25 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

This disclosure provides a computer-implemented method. The method may include extracting one or more features based on a first utterance from a first interlocutor in a dialog and a second utterance from a second interlocutor in the dialog. The method may further include inferring one or more personality traits of the first interlocutor based on the one or more extracted features from the dialog.