Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Commodity parallel dynamic pushing method in e-commerce platform

An e-commerce platform and commodity technology, which is applied in the fields of parallel dynamic push of commodities, multi-threaded computing and e-commerce website push, and parallel clustering, which can solve the problems of time-consuming, inability to push commodities to customers, and difficulty in maintenance.

Active Publication Date: 2020-03-31
HARBIN UNIV OF SCI & TECH
View PDF10 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, there is a disadvantage in the push method described above: when screening and pushing among a large number of customers and commodities, it will be time-consuming and cannot push commodities to customers in time
At present, there is already a Hadoop platform for product push, but the Hadoop platform requires multiple data nodes, a name node and a secondary name node, and each node requires an independent computer, which is difficult to maintain. Moreover, in Hadoop The MapReduce operation also needs to access the disk, which will inevitably cause I / O overhead

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
  • Commodity parallel dynamic pushing method in e-commerce platform

Examples

Experimental program
Comparison scheme
Effect test

example 1

[0108] Example 1: Build a matrix.

[0109] Taking the construction of scoring matrix as an example, set the master process ID number master_id=0, and the slave process ID number slaver_id=1;

[0110] The main process master constructs the scoring matrix, the function is MasterConstructMatrix(type), as follows:

[0111] S_MasterConstructMatrix_1, initialize the MPI environment, and obtain server parameters;

[0112] S_MasterConstructMatrix_2, start all slave processes in the distributed cluster to construct block matrix;

[0113] S_MasterConstructMatrix_3, use the MPI_Recv(0, 1, matrix, &status) function to receive the block matrix constructed from the process, and assemble it into a complete matrix;

[0114] S_MasterConstructMatrix_4, the main process construction matrix ends;

[0115] Construct the block matrix in parallel from the process slaver, the function is SlaverConstructMatrix(type), as follows:

[0116] S_SlaverConstructMatrix_1. Construct the block matrix matrix...

example 2

[0119] Example 2: Parallel clustering of customers.

[0120] Let the number of clusters generated be k 1 =2,k 2 =3,k 3 =4, master process ID number master_id=0, assign three slave processes, the ID numbers of the slave processes are 1, 2, 3 in turn; w=1, c 1 =c 2 =2,z 1 =z 2 =0.5, r=1;

[0121] Let each row vector in the customer matrix be:

[0122] u 1 =(4,0,0,5,1,0,0,3); U 2 =(5,5,4,0,0,2,0,0);

[0123] u 3 =(0, 0, 0, 2, 5, 4, 3, 3); U 4 =(0,3,0,2,0,0,3,0);

[0124] u 5 =(3,2,0,2,0,0,4,0); U 6 =(4,0,0,0,1,0,0,2);

[0125] u 7 =(5,0,4,0,0,3,0,0); U 8 =(0,0,0,2,4,0,3,3);

[0126] u 9 =(0,2,0,0,0,0,3,0); U 10 =(3,2,0,0,0,0,4,0);

[0127] The target customer is U 3 , the slave process takes slaver_id=1 as an example, and the number of threads allocated from the process is 4;

[0128] The main process master executes the particle swarm optimization algorithm to optimize the number k of clusters. The function is MasterAirosol(), as follows:

[0129] S_Maste...

example 3

[0197] Example 3: Parallel calculation of product similarity.

[0198] Let each customer's rating vector for the product in the rating matrix be:

[0199] R 1 =(-0.33, 1.17, 0.83, 0);

[0200] R 2 =(0, -0.25, 0, 0.25);

[0201] R 3 =(-1.25,0,1.25,0);

[0202] R 4 =(0.5,0,0,-0.5);

[0203] Each commodity vector in the commodity attribute matrix is:

[0204] I 1 =(0,0,1,0);

[0205] I 2 =(1,0,1,1);

[0206] I 3 =(1,1,1,0);

[0207] I 4 =(0,0,0,1);

[0208] The number of allocated threads is 2, the similarity weight is w=0.4, and the target product is I 4 ;

[0209] The function for parallel calculation of product similarity is ParallelSim(), as follows:

[0210] S_ParallelSim_1, allocate 2 threads and add them to the thread queue Queue, that is, Queue=(thread_1, thread_2);

[0211] S_ParallelSim_2, set similarity weight w=0.4;

[0212] S_ParallelSim_3, thread thread_1, thread_2 dequeue, Queue=(empty);

[0213] S_ParallelSim_4, the commodity I 1 , I 2 Assig...

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 provides a commodity parallel dynamic pushing method in an e-commerce platform. The method comprises the following steps that: firstly, a server constructs a client matrix, a commodity attribute matrix and a scoring matrix by adopting a distributed parallel mode of a message passing interface (MPI): each slave process constructs a client block matrix in a distributed database in parallel, and a master process merges the block matrixes constructed by each slave process into a complete matrix; secondly, the clients are clustered in a parallel mode based on MPI; the master process executes a particle swarm optimization algorithm to obtain the optimal cluster number required by customer clustering, the slave processes execute a k-means clustering algorithm in parallel, and when the k-means clustering algorithm is executed, each slave process allocates a plurality of threads to calculate the Euclidean distance between a customer vector and a mean vector and updates the mean vector; secondly, the similarity between the target commodity and the remaining commodities is calculated in parallel in a multi-thread mode; and finally, dynamic pushing is carried out to generate a pushing list.

Description

technical field [0001] The invention relates to a method for parallel clustering, multi-thread calculation and e-commerce website push, in particular to a parallel dynamic push method for commodities in an e-commerce platform, which belongs to the field of parallel computing. Background technique [0002] The rapid development of the Internet and Web technology has promoted the growing prosperity of the e-commerce market based on the application of network information technology, and the traditional production and lifestyle of human beings have undergone earth-shaking changes. The convenient and fast shopping experience has made many consumers increasingly favor online shopping. Online shopping has become an indispensable and important part of people's life. However, with the increasing variety and quantity of products on e-commerce websites, the problem of information overload caused when online shopping provides people with rich and colorful product information makes custo...

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): G06F16/9536G06Q30/02G06Q30/06G06K9/62G06N3/00
CPCG06F16/9536G06Q30/0271G06Q30/0631G06N3/006G06F18/23213Y02D10/00
Inventor 刘嘉辉朱宝森
Owner HARBIN UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com