Chatbot Generation via Dialogue Tree Pruning

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Configuring chatbots for contact centers is labor-intensive and time-consuming, requiring extensive manual review and analysis of conversations to cover various topics and conversation paths, limiting their efficiency and topical coverage.

Innovation Solution

A method and system for automatically or semi-automatically generating topic-specific chatbots using sample dialogue data, which includes clustering interactions, extracting dialogue graphs, and pruning dialogue trees to create deterministic conversation paths, reducing the need for manual intervention and increasing the chatbot's ability to handle diverse topics and conversation paths.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If manual review and analysis of conversations is used to configure chatbots, then the chatbot can cover various topics and conversation paths, but the configuration process becomes labor-intensive and time-consuming

Engineering Contradiction:

Improvetopical coverageVSAvoidconfiguration time

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The system enables automatic chatbot generation by allowing the chatbot to self-configure through automated analysis of conversation transcripts. The generation module automatically extracts dialogue models and configures the chatbot without requiring manual intervention, thus reducing configuration time while maintaining topical coverage.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system changes the parameter of configuration automation from manual to automatic by implementing algorithms that automatically extract dialogue models from transcripts. This parameter change transforms the configuration process from labor-intensive manual review to automated generation, significantly reducing time while preserving comprehensive topic coverage.

Inventive Principle:

Principle #35Parameter changes

2Adaptability or versatility

If manual review and analysis of conversations is used to configure chatbots, then the chatbot can cover various topics and conversation paths, but the labor intensity increases

Engineering Contradiction:

Improvetopical coverageVSAvoidconfiguration effort

Core Design Contradiction:

Adaptability or versatilityVSEase of manufacture

Solution Approach 1:

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system replaces the mechanical manual review process with an automated computational system. The generation module uses algorithms to automatically analyze transcripts, extract dialogue models, and configure the chatbot, substituting human manual effort with automated mechanical processing.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If automated chatbot generation is used, then the configuration time and effort are reduced, but the ability to handle less frequent topics may be limited

Engineering Contradiction:

Improveconfiguration efficiencyVSAvoidcoverage of less frequent topics

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The system performs automated analysis on a subset of conversation transcripts that are representative of various topics including less frequent ones. By selecting and analyzing diverse samples, the system ensures that even less frequent topics are captured in the dialogue model extraction process, maintaining comprehensive coverage while achieving automation.

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The system incorporates feedback mechanisms where the generated chatbot is evaluated for its ability to handle various topics. This feedback loop allows the system to iteratively improve the dialogue model extraction process, ensuring that less frequent topics are adequately represented while maintaining high configuration efficiency.

Inventive Principle:

Principle #23Feedback

Data Source

PatentEP3724789B1Systems and methods for chatbot generation

Publication Date: 2023.12.06 GREENEDEN U S HLDG II LLC

Figure 1
Figure 2
Figure 3

AI summary

A method for configuring a topic-specific chatbot comprising: clustering a plurality of transcripts of interactions between customers and agents of a contact center of an enterprise to generate a plurality of clusters of interactions, each cluster of interactions corresponding to a topic, each of the interactions including agent phrases and customer phrases; for each cluster of the plurality of clusters of interactions: extracting a topic-specific dialogue tree for the cluster; pruning the topic-specific dialogue tree to generate a deterministic dialogue tree; and configuring a topic-specific chatbot in accordance with the deterministic dialogue tree; and outputting the one or more topic-specific chatbots, each of the topic-specific chatbots being configured to generate, automatically, responses to messages regarding the topic of the topic-specific chatbot from a customer in an interaction between the customer and the enterprise.