A method and device for commodity entity matching based on set similarity
A technology that integrates similarity and entities, applied in the field of artificial intelligence, can solve the problems of relying on domain information, lack of domain information in the field of material and trade, and high requirements for the professionalism and accuracy of labelers, so as to narrow the matching range, reduce manual intervention, Solve the effect of fusion difficulties
Active Publication Date: 2022-03-08
鲁班(北京)电子商务科技有限公司
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] Entity matching methods based on supervised learning and semi-supervised learning rely on manual labeling of a large number of entity pairs as prior knowledge, which requires high professionalism and accuracy of labelers, and cannot perform well for multi-source heterogeneous data fusion. Effect
However, the entity matching method based on unsupervised learning takes a long time to accurately cluster and is not suitable for application scenarios. Although the entity matching method based on the probability model can directly use the jaccard similarity to judge the entity pair matching, However, this method relies on field information when used, and lacks effective field information for the material trade field targeted by the present invention, so it is not applicable
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0092] In order to make the technical problems, technical solutions, and advantages to the present invention, the technical solutions and advantages will be described in detail below with reference to the accompanying drawings and specific embodiments.
[0093] Such as figure 1 As shown, the embodiment of the present invention provides a merchandise matching method based on a set similarity, which is implemented by an electronic device. Such as figure 1 The flowchart of a commodity entity matching method based on a set similarity, which can include the following steps:
[0094] S11, get the platform knowledge base and the knowledge base to be matched.
[0095] S12, input the platform knowledge base and the entries to the real body matching model.
[0096] S13, based on platform knowledge base, to match the knowledge base and entity matching model, output entity match collection.
[0097] Alternatively, the entity matching model includes a knowledge base division module, a data pre...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a commodity entity matching method and device based on set similarity, and relates to the technical field of artificial intelligence. Including: acquiring the platform knowledge base and the knowledge base to be matched; inputting the platform knowledge base and the knowledge base to be matched into the entity matching model; outputting the entity matching set based on the platform knowledge base, the knowledge base to be matched and the entity matching model. The invention screens entities based on domain knowledge to narrow the matching range, uses optimized set similarity to calculate entity pair similarity, and uses domain rules to adjust entity pair sorting, which can effectively improve the accuracy of entity alignment in multi-source heterogeneous data, effectively It solves the problem of difficult data fusion at the bottom of traditional intelligent e-commerce platforms, greatly reduces manual intervention, and provides new ideas for the sustainable development of e-commerce in traditional industries.
Description
Technical field [0001] The present invention relates to the field of artificial intelligence technology, in particular to a commodity entity matching method and apparatus based on aggregate similarity. Background technique [0002] In recent years, knowledge maps have better organizational, manage data, can store different types of data and complex entity relationships and have good data flow efficiency, which is widely used in scenarios that require a lot of knowledge, such as with Q & A In the e-commerce industry system for new business operations such as search, recommendations such as demand. In the field of e-commerce, the business scale is expanded, and more complex data application scenarios, unstructured large amounts of data are dispersed in various sources and basically in non-structured text, and the demand for data interconnection is more Strong, and the deep cognition requirements for user needs are also improved. In this context, many types of e-commerce domain know...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/36G06F40/289G06F40/242G06N5/02
CPCG06F16/367G06F40/289G06F40/242G06N5/022G06N5/027
Inventor 张磊王文文任毅肖明明陈富强寇嘉敏
Owner 鲁班(北京)电子商务科技有限公司