Check patentability & draft patents in minutes with Patsnap Eureka AI!

Similarity-based address matching method and device

An address matching and similarity technology, which is applied in the field of similarity-based address matching methods and devices, can solve the problems of inability to adapt to logistics needs, high cost of classification models, and split delivery range, etc., to achieve convenient address matching, low cost, The effect of the exact address matching effect

Pending Publication Date: 2021-12-24
深圳云路信息科技有限责任公司
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the existing address matching technology often uses a classification model based on natural language, but the classification model based on natural language processing is extremely expensive, and the last-level network in the actual business often undergoes business changes and splits the delivery range. The class classification model needs to be retrained in a short cycle to learn business changes, so it cannot adapt to existing logistics needs

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Similarity-based address matching method and device
  • Similarity-based address matching method and device
  • Similarity-based address matching method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0101] see figure 1 , figure 1 It is a schematic flowchart of a similarity-based address matching method disclosed in an embodiment of the present invention. in, figure 1 The described method can be applied to a corresponding model training terminal, model training device or model training server, and the server can be a local server or a cloud server. Such as figure 1 As shown, the similarity-based address matching method may include the following operations:

[0102] 101. Obtain an address to be matched and multiple historical address vectors.

[0103] Optionally, the address to be matched may be a receiving address or a sending address, and the historical address vector may be a historical receiving address or a historical sending address, which is not limited in the present invention. Optionally, the address to be matched may be the receiving address of the distribution network to be confirmed, and the historical address vector may be a vector corresponding to an addr...

Embodiment approach

[0109] As an optional implementation, the method also includes:

[0110] Determine the target longitude and latitude information corresponding to the target matching address vector;

[0111] According to the latitude and longitude information of the target and the geographic fence information of multiple matching outlets, determine the outlet information corresponding to the latitude and longitude information of the target.

[0112] Optionally, the geographic fence information is used to indicate the latitude and longitude range of delivery or the latitude and longitude range of the matching network point.

[0113]Optionally, the target latitude and longitude information can be matched with the geo-fence information of each matching network point, and if it is found that the target longitude and latitude information is within the range of the geo-fence information of any matching network point, then the matching network point is determined as the target longitude and latitude ...

Embodiment 2

[0194] see figure 2 , figure 2 It is a structural schematic diagram of a similarity-based address matching device disclosed in an embodiment of the present invention. in, figure 2 The described apparatus can be applied to a corresponding model training terminal, model training device or model training server, and the server can be a local server or a cloud server, which is not limited in the embodiment of the present invention. Such as figure 2 As shown, the device may include:

[0195] The acquiring module 201 is configured to acquire an address to be matched and a plurality of historical address vectors.

[0196] Optionally, the address to be matched may be a receiving address or a sending address, and the historical address vector may be a historical receiving address or a historical sending address, which is not limited in the present invention. Optionally, the address to be matched may be the receiving address of the distribution network to be confirmed, and the ...

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

No PUM Login to View More

Abstract

The invention discloses a similarity-based address matching method and device. The method comprises the following steps: collecting a to-be-matched address and a plurality of historical address vectors; determining a to-be-matched address vector corresponding to the to-be-matched address according to a trained vector conversion model and a conversion weight table; and based on an HNSW algorithm, determining a historical address vector most similar to the to-be-matched address vector from the plurality of historical address vectors, and obtaining a target matched address vector. Therefore, according to the method, the to-be-matched address can be converted into the vector based on the trained vector conversion model and the weight table, and then the most similar historical address vector is matched through the algorithm, so that the address matching with lower cost and more convenience is realized, the cost of the algorithm is reduced, the expandability of the algorithm is improved. The technical scene of repeated transformation of the existing logistics network can be handled, and a more accurate address matching effect can be achieved.

Description

technical field [0001] The invention relates to the technical field of intelligent logistics, in particular to an address matching method and device based on similarity. Background technique [0002] In the field of intelligent logistics, address matching is an important part. It can be used to match the address of the distribution network or the address of the self-pickup network corresponding to the receiving address, or match the pickup address corresponding to the sending address. Therefore, to a large extent , affects the overall transportation efficiency of the entire logistics transportation system. However, the existing address matching technology often uses a classification model based on natural language, but the classification model based on natural language processing is extremely expensive, and the last-level network in the actual business often undergoes business changes and splits the delivery range. Class classification models need to be retrained to learn b...

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
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06Q10/08G06K9/62G06F40/289G06F40/216G06N3/08
CPCG06Q10/083G06F40/216G06F40/289G06N3/088G06F18/22
Inventor 阚向阳姜东晓崔德冠简雄孙海林杨经海
Owner 深圳云路信息科技有限责任公司
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More