Query-Driven Summary Generation for Self-Media Articles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional summary generation methods fail to effectively address the demand for query-focused summaries, particularly in the context of self-media articles with incomplete narrative structures and unclear headlines, where users need direct answers to questions posed in the headlines.
Innovation Solution
A method and apparatus that acquire a target article, determine if a question is included in the headline, identify the relevant paragraph in the article body containing the answer, and generate a summary based on this information-satisfied-paragraph, including or excluding surrounding paragraphs as necessary to provide a coherent summary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional summary generation methods are used, then the summary covers key content of the article, but it does not directly answer users' questions posed in the headline
Solution Approach 1:
The patent segments the article into multiple paragraphs and identifies specific paragraphs that contain answers to questions in the headline. Instead of treating the article as a whole, the system divides it into manageable units and selectively processes those most relevant to user queries, thereby improving information satisfaction without requiring users to read the entire article.
Solution Approach 2:
The patent extracts the core question from the headline and identifies the specific paragraph containing the answer. By taking out only the relevant information-satisfied paragraph and generating the summary from that specific section, the system directly addresses user information demands while reducing unnecessary content in the summary.
2Stability of the object's composition
If the summary includes more content to ensure coherence, then the narrative structure is improved, but the summary length increases and may include irrelevant information
Solution Approach 1:
The patent dynamically adjusts the summary generation process based on the position of the information-satisfied paragraph within the article. When the answer paragraph is not at the beginning, the system adaptively includes preceding or following paragraphs to ensure narrative coherence, rather than using a fixed summary structure. This dynamic approach maintains readability while minimizing unnecessary content.
Solution Approach 2:
The patent applies different processing rules to different parts of the article based on their relationship to the information-satisfied paragraph. Specifically, it selectively includes only those surrounding paragraphs that are necessary for contextual coherence, rather than uniformly including all paragraphs. This local quality approach ensures narrative stability while controlling summary length.
3Loss of information
If the system processes the entire article to generate a comprehensive summary, then all key content is covered, but the processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary action by first identifying the question in the headline and locating the information-satisfied paragraph before generating the summary. This preliminary identification of the core answer paragraph allows the system to focus processing efforts only on relevant sections of the article, significantly reducing processing time while maintaining comprehensive coverage of key information.
Solution Approach 2:
The patent extracts and prioritizes the information-satisfied paragraph that contains the direct answer to the headline question. By taking out this core paragraph and using it as the foundation for summary generation, the system ensures comprehensive coverage of key content while avoiding the need to process and include all paragraphs, thereby reducing computational resources and processing time.
Data Source
AI summary
Embodiments of the application disclose a summary generation method and apparatus. A specific embodiment of the method comprises: acquiring a target article including a headline and a body of the article; determining whether a question is included in the headline; determining, in the body of the article, an information-satisfied-paragraph including an answer to the question, in response to determining that the question is included in the headline; and generating a summary of the target article based on the determined information-satisfied-paragraph. The above embodiment may generate a summary that directly satisfies the users' demand for information.