Text Tag Generation via Multi-Strategy Extraction and Reduplication Removal
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing text tag extraction methods rely solely on word frequency, resulting in low accuracy and ineffective content representation.
Innovation Solution
A method and device for generating text tags that perform keyword extraction using strategies corresponding to tag types, followed by reduplication removal between different tag types to validate candidate tags, improving the accuracy of tag extraction by filtering and sorting based on semantic importance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If tag extraction is based solely on word frequency, then the extraction process is simple, but the accuracy of extracted tags is low
Solution Approach 1:
The patent segments the tag extraction process into multiple distinct modules: a candidate tag generation module that extracts potential tags based on word frequency and semantic analysis, a tag type classification module that categorizes candidates into entity words, segment texts, and topics, and a final selection module that determines optimal tags. This segmentation allows each module to specialize in specific tasks, improving overall accuracy while maintaining manageable complexity through modular design.
Solution Approach 2:
The patent changes multiple parameters in the extraction process: it considers not only word frequency but also semantic importance, tag type classifications, and redundancy metrics. By introducing these additional parameters and weighting them appropriately, the system achieves higher accuracy in tag extraction while the structured parameter management prevents excessive complexity.
2Measurement precision
If multiple tag types are extracted without reduplication removal, then more candidate tags are obtained, but redundancy increases and accuracy decreases
Solution Approach 1:
The patent implements a feedback mechanism through the reduplication removal module that analyzes the relationships between candidate tags from different types. The system evaluates semantic redundancy, hierarchical relationships, and overlap between entity words, segment texts, and topics, then provides feedback to filter out duplicate or overly similar tags. This feedback loop ensures that only the most informative and non-redundant tags are selected, improving accuracy while preserving essential information.
3Measurement precision
If keyword extraction uses multiple strategies for different tag types, then tag accuracy improves, but processing complexity increases
Solution Approach 1:
The patent divides the text processing into distinct segmentation stages: entity word extraction focuses on named entities and proper nouns, segment text extraction identifies phrase-level meaningful units, and topic extraction captures broader subject matter. Each segmentation strategy is optimized for its specific tag type, improving accuracy while the parallel processing capability maintains processing efficiency by avoiding sequential bottlenecks.
Solution Approach 2:
The patent designs a universal extraction framework that handles multiple tag types (entity words, segment texts, topics) through a common architecture. The candidate tag generation module serves all three tag types, and the reduplication removal module processes all candidates uniformly. This multi-functional design improves accuracy through comprehensive coverage while maintaining productivity by avoiding the need for separate processing pipelines for each tag type.
Data Source
AI summary
The present disclosure provides a method and a device for generating a text tag. The method includes: performing keyword extraction using strategies corresponding to respective tag types on a target text, to obtain one or more candidate tags of the respective tag types for the target text, wherein the tag type includes at least one of an entity word, a segment text and a topic; performing reduplication removing between different tag types on the one or more candidate tags of the respective tag types to obtain one or more validated candidate tags; and determining one or more target tags of the target text based on the one or more validated candidate tags.