Computational Model for Detecting Non-Narrative Text Segments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems struggle to accurately provide relevant and narrative-focused information about media content items, often including non-narrative segments such as advertisements or promotional content in text descriptions.
Innovation Solution
The use of a trained computational model to identify and score segments in text associated with media content items, allowing for the removal of non-narrative segments and the generation of clean text that is topical and relevant to the narrative.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If text descriptions are provided from producers or authors, then information availability is improved, but information accuracy regarding narrative content deteriorates due to inclusion of non-narrative segments
Solution Approach 1:
The text description is divided into multiple segments or sentences, and each segment is independently evaluated by the computational model to determine whether it belongs to the narrative content or non-narrative content, enabling precise identification and separation of relevant information
Solution Approach 2:
The computational model extracts and identifies non-narrative segments from the text description, separating them from the narrative content. This extraction process removes advertisements, promotional content, and other non-narrative elements while preserving the core narrative information
2Measurement precision
If computational models are applied to identify non-narrative segments, then information accuracy is improved, but processing complexity increases
Solution Approach 1:
The computational model is pre-trained on large datasets of text descriptions with labeled narrative and non-narrative segments. This preliminary training equips the model with the knowledge needed to accurately identify non-narrative content without requiring complex real-time analysis, reducing processing complexity during actual use
Solution Approach 2:
The computational model serves as an intermediary between the raw text description and the final cleaned output. It automatically performs the complex task of identifying and flagging non-narrative segments, simplifying the overall processing pipeline and reducing the need for manual intervention
Data Source
AI summary
A method includes retrieving a text from a database. The text corresponds to audio from a media content item that is provided by a media providing service, and the text includes a plurality of segments. The method also includes assigning a score for each segment in the text by applying the text to a trained computational model. The score corresponds to a predicted relevance of the respective segment to a narrative of the media content item. The method further includes identifying a non-narrative segment within the text using the assigned scores.