Keyword Generation Model for Skill Phrase Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for keyword extraction from natural language text, such as direct extraction and neural network-based approaches, suffer from low accuracy in identifying relevant keywords for applications like talent management and job matching.
Innovation Solution
A keyword generating method that utilizes a text processing model to encode words into semantic representation vectors, predicts whether each vector can trigger a keyword, and outputs keywords based on these predictions, improving accuracy by distinguishing between consecutive and non-consecutive keywords.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If direct extraction or neural network models are used to generate keywords, then the process is simple, but the accuracy is low
Solution Approach 1:
The patent segments the keyword generation process into multiple distinct modules: text processing model for semantic representation, prediction model for keyword triggering probability, and keyword generation model for final output. Each module handles a specific aspect of the task, improving overall accuracy while maintaining manageable complexity through functional decomposition
Solution Approach 2:
The patent introduces semantic representation vectors as an intermediary between the input text and the prediction model. These vectors capture the semantic meaning of text sequences, enabling the prediction model to make more accurate predictions about keyword triggering probabilities without requiring direct complex text analysis
2Measurement precision
If only consecutive keywords are extracted, then the extraction process is simple, but non-consecutive keywords are missed
Solution Approach 1:
The patent employs a dynamic keyword generation process that adapts to the input text characteristics. The model dynamically determines whether to extract consecutive or non-consecutive keywords based on the semantic representation and prediction results, allowing flexibility to capture both types of keywords appropriately rather than being constrained to a fixed extraction pattern
Solution Approach 2:
The patent transitions from traditional single-dimension keyword extraction (consecutive words only) to a multi-dimensional approach by considering semantic relationships across different positions in the text. The semantic representation vectors enable the model to identify keywords based on semantic coherence rather than just positional adjacency, adding a semantic dimension to the extraction process
Data Source
AI summary
This application discloses a keyword generating method, an apparatus, a device and a storage medium, which relate to the field of natural language processing in the field of artificial intelligence. A specific implementation scheme includes: inputting a target text into a text processing model, obtaining a word sequence corresponding to the target text, and generating a semantic representation sequence corresponding to the word sequence; making prediction about each semantic representation vector in the semantic representation sequence respectively to obtain a prediction result; and if the prediction result indicates that a word corresponding to the semantic representation vector is capable of triggering a generation of a keyword, outputting the keyword based on the semantic representation vector and the prediction result. This method improves the accuracy of generating keywords.