Summary Engine Template Instantiation for Narrative Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a need for an improved method to generate natural language narratives from event data, such as sports events, to make them more understandable and engaging for consumers, as existing resources like writers and time may be insufficient to create compelling stories.
Innovation Solution
A system comprising a summary engine that collects event data, selects templates with static text and dynamic slots, and uses conditions to choose words, creating the impression of a human-generated story, which can be published across various platforms to increase interest and traffic.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If manual writers create stories from event data, then the narrative quality and understandability are improved, but the time consumption and resource requirements increase significantly
Solution Approach 1:
The system enables automatic self-service generation of narratives by using templates that automatically populate event data into structured story formats, eliminating the need for manual writer intervention while maintaining narrative quality and understandability
Solution Approach 2:
The patent transforms the narrative generation process by changing parameters from manual text creation to automated template instantiation, where templates contain pre-defined narrative structures with placeholders that are automatically filled with event data parameters
2Productivity
If more writers are hired to create narratives, then the quantity of narratives increases, but the cost and resource requirements increase
Solution Approach 1:
The system uses template copying where a single template can be instantiated multiple times with different event data, allowing unlimited narrative generation from a finite set of templates, thereby increasing productivity without proportional resource increase
Solution Approach 2:
Templates serve multiple functions by being reusable across different events and narrative types, allowing one template to generate numerous narratives through parameter substitution, eliminating the need for separate resources for each narrative creation
3Loss of information
If detailed event data is presented, then the information completeness is improved, but the complexity and difficulty of understanding increases
Solution Approach 1:
The patent segments detailed event data into structured components that fit into predefined template sections, organizing complex information into manageable narrative units that are easier to understand while preserving completeness
Solution Approach 2:
Templates act as intermediaries between raw event data and human readers, transforming complex data structures into natural language narratives that maintain information completeness while reducing perceived complexity
Data Source
AI summary
Various embodiments for providing a natural language narrative are described. Embodiments may include a summary engine that collects data about an event, and selects a template to narrate a story about the event. A template may include static text, slots for data from the event, and slots for dynamic word choice. Templates may be selected according to a story type and story “flavor.” Templates may also include conditions that determine a choice of the use of template sub-sections. Various aspects of the templates may be selected randomly, to create the impression that the story is human-generated. The stories may be published to a variety of locations and formats. Other embodiments are described and claimed.