招聘行业文本召回方法及系统、设备与介质
By constructing a pre-defined knowledge graph and word representation model to process recruitment industry text, the problems of redundant recall results and slow calculation speed in existing technologies are solved, achieving faster and more accurate text recall in the recruitment industry.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANGHAI JIBEIKE IT CO LTD
- Filing Date
- 2022-11-04
- Publication Date
- 2026-07-17
AI Technical Summary
Existing text retrieval methods in the recruitment industry cannot accurately adapt to the special requirements of person-job matching. In particular, they are slow to compute when dealing with massive amounts of data. Furthermore, general text recommendation retrieval methods cannot be processed based on occupational and skill dimensions, resulting in redundant and inaccurate retrieval results.
By constructing a pre-defined knowledge graph and performing fuzzy matching, the correlation between skills and occupational dimensions is obtained. The text information is processed using a pre-defined ranking algorithm and word representation model to obtain vectors of main keywords and skill keywords. Vector concatenation and recall algorithm calculations are then performed to achieve accurate mapping of resume job feature vectors.
It enables faster and more accurate text retrieval in the recruitment industry, improving retrieval efficiency and quality, accurately matching the correlation between occupations and skills, and improving calculation speed and retrieval accuracy.
Smart Images
Figure CN115757694B_ABST