AI Narrative Generation via Composable Communication Goals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional natural language generation (NLG) systems face limitations in communicating data-driven ideas per sentence, variability in word choice, and analyzing data sets to determine content for narrative stories, leading to constrained narrative generation capabilities.
Innovation Solution
The development of AI technology that uses composable communication goal statements and ontologies to facilitate the generation of narrative stories without requiring direct coding, allowing users to structure story outlines and edit narratives, with the ontology learning from user edits and adapting to data analysis goals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional template-based NLG systems are used, then narrative generation is simplified, but the ability to communicate multiple data-driven ideas per sentence and analyze data sets is limited
Solution Approach 1:
The system segments the narrative generation process into distinct modules: data analysis module, communication goal module, and natural language generation module. Each module handles specific functions independently, allowing the system to process complex data sets and generate varied narratives without requiring complete system redesign. The segmentation enables specialized analysis capabilities while maintaining overall system manageability.
Solution Approach 2:
The patent implements a universal NLG platform that can handle multiple communication goals (summarize, compare, explain, predict) and various narrative types through a single integrated system. The ontology-based approach allows the same system to adapt to different domains and communication objectives without requiring separate specialized systems for each function.
2Adaptability or versatility
If template-based approaches are used, then system simplicity is maintained, but variability in word choice and narrative diversity is constrained
Solution Approach 1:
The system dynamically generates narratives by selecting from multiple possible wordings and structures based on the analyzed data and communication goals. Rather than using fixed templates, the NLG module creates varied narratives adaptively, allowing the same data to produce different narrative versions. This dynamic generation maintains ease of operation through automated selection while significantly improving narrative diversity.
Solution Approach 2:
The system changes parameters such as word choice, sentence structure, and narrative focus based on the communication goals and data characteristics. By adjusting these parameters dynamically, the system can generate diverse narratives without requiring users to manually specify each detail, maintaining ease of operation while achieving high variability in output.
3Measurement precision
If conventional NLG systems are used, then basic narrative generation works, but the ability to analyze data sets and determine content for stories is limited
Solution Approach 1:
The system separates data analysis from narrative generation into distinct modules. The data analysis module independently processes data sets, identifies patterns, and determines content, while the NLG module focuses on translating this analysis into natural language. This segmentation allows sophisticated analysis capabilities without overwhelming the entire system, as each module can be optimized independently.
Solution Approach 2:
The patent introduces an intermediary ontology layer that mediates between raw data and narrative generation. The ontology structures and interprets data before it reaches the NLG module, enabling sophisticated data analysis and content determination. This intermediary layer handles the complexity of data interpretation while keeping the NLG module relatively simple and focused on language generation.
4Adaptability or versatility
If users need to directly code narrative generation systems, then customization is possible, but user accessibility is limited to programming experts
Solution Approach 1:
The system provides self-service capabilities by automatically generating narratives based on user-selected communication goals and provided data. Users can customize narratives through high-level parameters rather than coding, and the system automatically handles the complex tasks of data analysis, content determination, and language generation. This self-service approach maintains customization capability while making the system accessible to non-programming users.
Solution Approach 2:
The patent introduces an intermediary interface layer that translates user-friendly communication goal selections into the underlying system operations. This interface mediates between simple user inputs and complex system processing, allowing users to customize narratives through intuitive selections rather than coding. The intermediary layer handles the technical complexity while presenting a simplified, accessible interface to users.
Data Source
AI summary
Artificial intelligence (AI) technology can be used in combination with composable communication goal statements to facilitate a user's ability to quickly structure story outlines using “explanation” communication goals in a manner usable by an NLG narrative generation system without any need for the user to directly author computer code. This AI technology permits NLG systems to determine the appropriate content for inclusion in a narrative story about a data set in a manner that will satisfy a desired explanation communication goal such that the narratives will express various ideas that are deemed relevant to a given explanation communication goal.