CN113326441A

CN113326441A - Information pushing method and system based on big data service positioning

Info

Publication number: CN113326441A
Application number: CN202110519359.XA
Authority: CN
Inventors: 崔海燕
Original assignee: Individual
Current assignee: Individual
Priority date: 2020-11-17
Filing date: 2020-11-17
Publication date: 2021-08-31
Also published as: CN112464105B; CN113326442A; CN112464105A

Abstract

The embodiment of the application provides an information pushing method and system based on big data service positioning, after recommendation weight parameters between a selected recommendation page and a target data source recommendation page are generated according to the theme similarity of the selected recommendation page and the target data source recommendation page under at least one data source recommendation page, the recommendation weight parameters between the selected recommendation page and the target data source recommendation page under the subscription labels of various users are fused, a large number of reference bases based on the recommendation pages can be utilized, so that more target data source recommendation pages are obtained, the accuracy of subsequent material matching is improved, and the condition that errors occur when an index is carried out on a recommendation object to be indexed can be avoided when the recommendation pages are taken as the minimum basic unit, so the accuracy of material matching is improved.

Description

Information pushing method and system based on big data service positioning

The application is a divisional application of Chinese application with the name of 'internet platform information pushing method based on big data positioning and cloud computing center' invented and created by application number 202011286424.0 and application date of 17.11.2020.

Technical Field

The application relates to the technical field of big data positioning and artificial intelligence, in particular to an information pushing method and system based on big data service positioning.

Background

The current internet platform information push algorithm is a matching algorithm based on topic similarity to determine whether two recommended page nodes are matched. For example, the existing topic similarity algorithm uses either the recommendation page node as the minimum unit or the recommendation page as the minimum unit. The subject similarity algorithm with the recommendation page node as the minimum basic unit can avoid the error when indexing the recommendation object to be indexed, and can better solve the problem of wrong recommendation page nodes, but the minimum unit in the process of actually pushing information is the recommendation page, and the method is difficult to utilize a large amount of reference bases based on the recommendation page. In addition, the topic similarity algorithm using the recommendation page as the minimum basic unit is often prone to error when indexing is performed on the recommendation object to be indexed, and therefore the accuracy of matching of the pushed materials is low.

Disclosure of Invention

In order to overcome at least the above-mentioned deficiencies in the prior art, the present application aims to provide an information pushing method and system based on big data service positioning, after generating a recommendation weight parameter between the selected recommendation page and the target data source recommendation page according to the subject similarity of the selected recommendation page and the target data source recommendation page under at least one data source recommendation page, by fusing the recommendation weight parameters between the selected recommendation page and the target data source recommendation page under each user subscription label, a large number of reference bases based on the recommendation pages can be utilized, so that more recommendation pages of the target data source are obtained, the accuracy of matching of subsequent materials is improved, and, when the recommendation page is used as the minimum basic unit, the situation of error in indexing the recommendation object to be indexed can be avoided, and therefore the accuracy of material matching is improved.

In a first aspect, the application provides an internet platform information pushing method based on big data positioning, which is applied to a cloud computing center, wherein the cloud computing center is in communication connection with a plurality of business service terminals, and the method includes:

acquiring service pushing configuration optimization information of a service link in big data service positioning data of the service terminal;

acquiring a recommendation object to be indexed matching a plurality of page nodes to be recommended and a target index data source corresponding to the recommendation object to be indexed based on the service push configuration optimization information, wherein the target index data source is an index data source of a service tag to which the content of the recommendation object to be indexed belongs, and the target index data source comprises at least one data source recommendation page;

combining a plurality of page nodes to be recommended to obtain a selected recommended page which is subject-related to at least one data source recommended page, and generating a recommendation weight parameter between the selected recommended page and a target data source recommended page according to subject similarity of the selected recommended page and the at least one data source recommended page under a target user subscription tag;

and fusing recommendation weight parameters between the selected recommendation page and the target data source recommendation page under the subscription label of each user, selecting a target recommendation material matched with the recommendation object to be indexed from a preset target recommendation material library according to a fusion result, and pushing the target recommendation material to the business service terminal.

In a possible implementation manner of the first aspect, the generating, according to the topic similarity of the selected recommendation page and at least one data source recommendation page under a target user subscription tag, a recommendation weight parameter between the selected recommendation page and a target data source recommendation page includes:

determining a target user subscription tag corresponding to each data source recommendation page according to the topic association relationship between the selected recommendation page and the data source recommendation page;

based on the determined target user subscription tag, calculating the topic similarity of the selected recommendation page and at least one data source recommendation page in the determined target user subscription tag, and determining the data source recommendation page with the maximum topic similarity as a target data source recommendation page;

and generating a recommendation weight parameter between the selected recommendation page and the target data source recommendation page according to the topic similarity of the selected recommendation page and the target data source recommendation page under at least one user subscription tag.

In a possible implementation manner of the first aspect, the calculating, based on the determined target user subscription tag, a topic similarity of the selected recommendation page and at least one data source recommendation page in the determined target user subscription tag, and determining the data source recommendation page with the largest topic similarity as the target data source recommendation page includes:

calculating the similarity of a first topic of the selected recommendation page and at least one data source recommendation page under the same topic subscription tag, and determining the data source recommendation page with the maximum similarity of the first topic as a first target data source recommendation page;

calculating second topic similarity of the selected recommendation page and at least one data source recommendation page under the hierarchical topic subscription tag, and determining the data source recommendation page with the maximum second topic similarity as a second target data source recommendation page;

calculating the third topic similarity of the selected recommendation page and at least one data source recommendation page under the weight topic subscription tag, and determining the data source recommendation page with the maximum third topic similarity as a third target data source recommendation page;

the calculating the first topic similarity of the selected recommendation page and at least one data source recommendation page under the same topic subscription tag, and determining the data source recommendation page with the maximum first topic similarity as a first target data source recommendation page includes:

selecting a same-subject list set from the target index data source, wherein the same-subject list set comprises a plurality of same-subject lists, and each same-subject list comprises at least two data source recommendation pages with the same subject meaning;

determining a same-subject list which is the same as the recommended word vector label of the selected recommended page to obtain a target same-subject list;

calculating a first topic similarity between the selected recommendation page and each data source recommendation page in the target same-topic list, and determining a first target data source recommendation page from the data source recommendation page with the maximum first topic similarity;

the generating of the recommendation weight parameter between the selected recommendation page and the target data source recommendation page according to the topic similarity of the selected recommendation page and the target data source recommendation page under at least one user subscription tag comprises:

generating a first recommendation weight parameter between the selected recommendation page and a first target data source recommendation page according to the first theme similarity between the selected recommendation page and the first target data source recommendation page;

or, the calculating a second topic similarity between the selected recommendation page and at least one data source recommendation page under the hierarchical topic subscription tag, and determining the data source recommendation page with the maximum second topic similarity as a second target data source recommendation page includes:

determining a hierarchical theme relationship between the selected recommended page and at least one data source recommended page according to the recommended word vector label of the selected recommended page and the recommended word vector label of each data source recommended page;

calculating second topic similarity between the selected recommended page and a corresponding upper data source recommended page based on the determined hierarchical topic relation, and determining the data source recommended page with the maximum second topic similarity as a second target data source recommended page;

generating a second recommendation weight parameter between the selected recommendation page and a second target data source recommendation page according to a second theme similarity between the selected recommendation page and the second target data source recommendation page;

or, the calculating a third topic similarity between the selected recommendation page and at least one data source recommendation page under the weighted topic subscription tag, and determining the data source recommendation page with the maximum third topic similarity as a third target data source recommendation page includes:

collecting a theme weight value pre-established for each data source recommendation page;

calculating the similarity between the selected recommendation page and each data source recommendation page, and determining the data source recommendation page with the similarity larger than a preset value as a hot recommendation page to be selected;

calculating the third topic similarity of the selected recommendation page and the hot recommendation page to be selected with the topic weight value larger than the preset weight, and determining the hot recommendation page to be selected with the maximum third topic similarity as a third target data source recommendation page;

and generating a third recommendation weight parameter between the selected recommendation page and a third target data source recommendation page according to the third subject similarity between the selected recommendation page and the third target data source recommendation page.

In a possible implementation manner of the first aspect, the fusing recommendation weight parameters between the selected recommendation page and the target data source recommendation page under each user subscription tag includes:

acquiring a preset weight coefficient corresponding to each user subscription label;

calculating the product of the obtained weight coefficient and a recommendation weight parameter between the selected recommendation page and the target data source recommendation page under the corresponding user subscription tag to obtain a weighted recommendation weight parameter corresponding to each user subscription tag;

and fusing the weighted recommendation weight parameters corresponding to the subscription labels of the users.

In a possible implementation manner of the first aspect, the combining the multiple page nodes to be recommended to obtain a selected recommendation page associated with a topic existing in at least one data source recommendation page includes:

identifying the page pushing category of each page node to be recommended;

and removing the page nodes to be recommended with the page pushing category being the blacklist recommendation category, and arranging and combining the reserved page nodes to be recommended to obtain the selected recommendation page with theme association with at least one data source recommendation page.

In a possible implementation manner of the first aspect, the obtaining of the service push configuration optimization information of the service link in the big data service positioning data of the service terminal includes:

acquiring big data service positioning data of the service terminal, and performing service flow label extraction processing on the big data service positioning data to obtain service flow label information of service links in the big data service positioning data, wherein the big data service positioning data is a data set formed by service data recorded by each service link and acquired based on a single subscription service;

mining user intention characteristics based on the service flow label information of the service link to obtain target user intention characteristics of the service link;

performing user portrait feature extraction processing on the big data service positioning data based on an artificial intelligence model to obtain user portrait feature information of the service link;

and extracting a business relation between the target user intention characteristic of a business service link in the big data business positioning data and the user portrait characteristic information of the business service link to obtain business relation extraction information of the business service link, and performing business pushing configuration updating on the subscription business based on the business relation extraction information of the business service link to obtain business pushing configuration optimization information of the business service link.

In a possible implementation manner of the first aspect, the performing service flow label extraction processing on the big data service positioning data to obtain service flow label information of a service link in the big data service positioning data includes:

acquiring a service flow updating data list recorded by a service flow recording process of each big data corpus unit in the big data service positioning data, wherein the service flow updating data list comprises service flow updating data taking each service flow updating time sequence range as a recording unit range, and the service flow updating data comprises corpus acquisition triggering information and corpus pointing information of the service flow updating time sequence range and updating corpus records in the service flow updating time sequence range;

updating a time sequence range for each service flow, updating each behavior corpus in a plurality of behavior updating corpora in an updating corpus record of the time sequence range for updating the service flow according to each big data corpus unit, determining whether each behavior corpus description information in the behavior updating corpus is a candidate target behavior corpus description information according to a description feature vector of the behavior updating corpus description information, determining each candidate corpus tagging corpus unit segment corresponding to the behavior updating corpus according to the number of the candidate target behavior corpus description information in the behavior updating corpus, dividing the candidate corpus tagging corpus unit segment into a plurality of sub corpus unit segments according to each candidate corpus tagging corpus unit segment, determining whether the candidate corpus tagging corpus unit segment is a target corpus tagging unit segment according to a description label of each row of the behavior corpus description information in each sub corpus unit segment and a preset label range, wherein each behavior corpus description information corresponds to each corpus unit update behavior;

acquiring service flow node information of a preset service flow rule matched with each behavior corpus description information in the target corpus labeling corpus unit segment, wherein the service flow node information comprises corpus unit calling information and corpus unit subscription information, and the preset service flow rule comprises matching modes corresponding to different corpus unit use types;

determining the subject feature of each corpus unit subject intention topological tree and the abstract feature of each corpus unit abstract intention topological tree according to the service flow node information of each update corpus record of each different service flow update time sequence range in the service flow update data list, marking the subject feature of each corpus unit subject intention topological tree and the abstract feature of each corpus unit abstract intention topological tree in a corpus unit section according to the target corpus, determining the corpus unit label object of each big data corpus unit in the service flow update time sequence range, and taking the feature in the corpus unit range of the corpus unit label object and the feature out of the corpus unit range of the corpus unit label object and related to the corpus unit label object as the corpus label feature of each big data corpus unit in the service flow update time sequence range After the data is obtained, the corpus tag characteristics of each big data corpus unit in the updating time sequence range of all the service flows are fused to obtain the service flow tag information of the service link in the big data service positioning data.

In a possible implementation manner of the first aspect, the step of performing user intention feature mining based on traffic flow label information of the traffic service link to obtain a target user intention feature of the traffic service link includes:

acquiring retrieval instruction request information of a user retrieval instruction set added to corpus description information of each corpus tag feature in business flow tag information of the business service link, and determining a first retrieval behavior list corresponding to the retrieval instruction request information, wherein the retrieval instruction request information comprises access intention object information of intention word retrieval information determined according to intention word input information and intention word output information of the user retrieval instruction set, and the first retrieval behavior list comprises a high-low sequence of a plurality of access intention levels of the access intention object information;

determining corpus description information of each corpus tag feature based on a first intention word vector of intention word input information and based on a second intention word vector of intention word output information;

determining decision tree mining parameters for performing decision tree mining on the first search behavior list according to the access intention level relation between the first intention word vector and the second intention word vector;

performing decision tree mining on the first retrieval behavior list based on the decision tree mining parameters to obtain a second retrieval behavior list;

clustering the second retrieval behavior list to obtain a plurality of retrieval category cluster, and performing feature extraction on each retrieval category cluster to obtain a cluster feature vector;

determining user intention characteristics of each corpus tag characteristic according to the user intention characteristics corresponding to the plurality of cluster characteristic vectors corresponding to the second retrieval behavior list;

and obtaining the target user intention characteristic of the business service link based on the user intention characteristic of each corpus tag characteristic.

For example, in a possible implementation manner of the first aspect, the step of performing user portrait feature extraction processing on the big data service positioning data based on an artificial intelligence model to obtain user portrait feature information of the service link includes:

inputting the big data service positioning data into a pre-trained artificial intelligence model, and obtaining a confidence coefficient that the big data service positioning data is matched with each preset user portrait, wherein the artificial intelligence model is obtained by training based on a training sample and training annotation information corresponding to the training sample, the training sample is a big data service positioning data sample, and the training annotation information is user portrait annotation information;

determining a target user portrait corresponding to the big data service positioning data according to the confidence coefficient that the big data service positioning data is matched with each preset user portrait;

and extracting user portrait characteristic information matched with each service link from the user portrait description information of the target user portrait corresponding to the big data service positioning data.

For example, in a possible implementation manner of the first aspect, the step of performing service relationship extraction on the target user intention feature of the service link and the user portrait feature information of the service link in the big data service positioning data to obtain service relationship extraction information of the service link includes:

adding the target user intention characteristics and the user portrait characteristic information of the business service link into a business relationship extraction space, and determining a business source tracing script of the target user intention characteristics and the user portrait characteristic information of the business service link corresponding to each business relationship extraction node from the business relationship extraction space;

clustering the service tracing scripts according to the relationship network labels of the service relationships among the service relationship extraction nodes in the different service tracing scripts to obtain at least one service tracing script set; the relation network label of the service relation extraction node in any two service tracing scripts in the same service tracing script set covers a preset label range;

aiming at each service tracing script set, based on each service tracing script in the service tracing script set, determining service tracing characteristics corresponding to the target user intention characteristics and the user portrait characteristic information of the service tracing script set aiming at the service link from the service relationship extraction space; the service tracing characteristics at least comprise characteristic information data of each service tracing list item of each service tracing script in a service tracing script set aiming at target user intention characteristics and user portrait characteristic information of a service link, and the service tracing characteristics are used for determining service relationship extracting nodes in each service tracing script in the service tracing script set corresponding to service link target user intention characteristics and user portrait characteristic information service relationship tracing results;

and determining a service relation traceability result corresponding to a service relation extraction node in each service traceability script in the service traceability script set based on the service traceability characteristics, classifying the service relation traceability result by adopting a preset service relation classification model to obtain a classification result, and obtaining service relation extraction information of the service link according to the classification result.

For example, in a possible implementation manner of the first aspect, the step of determining a service relationship tracing result corresponding to a service relationship extraction node in each service tracing script in the service tracing script set based on the service tracing characteristics includes:

determining tracing service track information and tracing service types of the service tracing features, determining a plurality of tracing service linkage combinations according to a plurality of service relationship indexes in historical service relationship vector results, performing similarity calculation according to service matching degrees of each service relationship index in each tracing service linkage combination and the tracing service type in the tracing service track information to obtain a plurality of similarity calculation results respectively corresponding to the plurality of tracing service linkage combinations, and calling a frequent weighted value of the service of the tracing service linkage combination corresponding to each similarity calculation result as a weighted value of each similarity calculation result;

respectively obtaining similar business relation tracks of business tracing scripts which are determined to correspond to the tracing business categories in the tracing business track information based on the plurality of similarity calculation results to obtain a plurality of similar business relation tracks, integrating the plurality of similar business relation tracks obtained based on the plurality of similarity calculation results according to the weight value of each similarity calculation result to obtain a first similar business relation track set, wherein business calling frequency weighted values between any two business relation indexes in each tracing business linkage combination are equal, the business calling frequency weighted values corresponding to different tracing business linkage combinations are different, and each similarity calculation result is used for determining business relations in the business tracing scripts corresponding to the tracing business categories in any set time sequence section to extract business coincidence data corresponding to the nodes, the similarity calculation result is obtained by performing similarity calculation according to the historical business relation vector result;

acquiring service relationship extraction identification results of the service relationship extraction nodes between the tracing service track information and the tracing service category according to the historical service relationship vector result and service coincidence data of the service relationship extraction nodes corresponding to the tracing service category, and taking corresponding target request messages of the service relationship extraction identification results corresponding to the service relationship extraction nodes as second similar service relationship tracks;

and calculating the service relation related parameters of the first similar service relation track set and the second similar service relation track, and determining the service relation traceability result corresponding to each service relation extraction node according to the service relation related parameters and the service coincidence data of each service relation extraction node.

In a possible implementation manner of the first aspect, determining a service relationship tracing result corresponding to each service relationship extraction node according to the service relationship related parameter and the service coincidence data of each service relationship extraction node includes:

determining knowledge graph information of a business relation circle network corresponding to each business relation extraction node based on the business relation related parameters, and determining business triggering index information of each business relation extraction node through a business coverage area of business coincidence data of each business relation extraction node in a corresponding business tracing script;

extracting a first user intention characteristic sequence corresponding to the knowledge graph information and a second user intention characteristic sequence corresponding to the service triggering index information, and determining a plurality of service relation meta-knowledge networks with different feature description values, wherein the service relation meta-knowledge networks are respectively included in the first user intention characteristic sequence and the second user intention characteristic sequence; the business relation meta-knowledge network is a meta-knowledge network formed by user intention characteristics with incidence relation in the first user intention characteristic sequence and the second user intention characteristic sequence;

on the premise that the service relationship between the entities of the first user intention feature sequence is the same as the service relationship between the entities of the second user intention feature sequence, obtaining basic relationship map information of the knowledge map information in any service relationship meta-knowledge network of the first user intention feature sequence, and determining the service relationship meta-knowledge network with the minimum feature description value in the second user intention feature sequence as a target service relationship meta-knowledge network in parallel;

extracting a recommendation page extraction strategy based on the service relationship of each service relationship extraction node, adding the basic relationship map information to the target service relationship meta-knowledge network, and determining service upstream and downstream relationship information corresponding to the basic relationship map information in the target service relationship meta-knowledge network;

generating a knowledge graph connecting edge between the knowledge graph information and the service triggering index information according to service difference information between the basic relationship graph information and the service upstream and downstream relationship information;

acquiring target entity member information in the target service relationship element knowledge network by taking the service upstream and downstream relationship information as target information, adding the target entity member information to the service relationship element knowledge network in which the basic relationship map information is located according to a plurality of information matching paths corresponding to the knowledge map connecting edges, so as to obtain a member matching result corresponding to the target entity member information in the service relationship element knowledge network in which the basic relationship map information is located, and determining the member matching result as a service traceability result according to the target information; the target entity member information is information of an entity member corresponding to the access action;

determining a corresponding business relation result when the basic relation map information is added to the target business relation meta-knowledge network;

according to the association degree between the member matching result and the service relationship characteristic labels corresponding to the service relationship tracing nodes in the service relationship result, sequentially acquiring the service tracing labels corresponding to the service tracing result in the second user intention feature sequence according to the sequence of the time sequence weight from big to small until the number of the intellectual property network nodes of the service relation element intellectual network where the service tracing labels are positioned is consistent with the number of the intellectual property network nodes of the service tracing result in the first user intention feature sequence, stopping acquiring the service tracing labels in the next service relation element intellectual network, and establishing traceability associated information between the service traceability result and the last obtained service traceability label, and determining the service relationship traceability result corresponding to each service relationship extraction node based on the traceability associated information.

For example, in a possible implementation manner of the first aspect, the classifying the service relationship tracing result by using a preset service relationship classification model to obtain a classification result, and obtaining the service relationship extraction information of the service link according to the classification result includes:

determining a business relation floating parameter of a sample sequence of a preset business relation classification model, updating the sample sequence according to a previous business relation classification result when the business relation floating parameter does not reach a set floating range, inputting a business relation traceability result into the preset business relation classification model when the business relation floating parameter reaches the set floating range, and judging whether a classification result output by the preset business relation classification model is received or not; wherein the previous time period is determined according to the set time sequence period;

when a classification result output by the preset business relation classification model is received, a model updating process is established according to the classification result;

when a classification result corresponding to a business tracing classification label is output by a preset business relation classification model, determining an update source parameter of the classification result corresponding to the business tracing classification label in the model update process;

when the update source parameter of the classification result corresponding to the service tracing classification label in the model update process is determined to be a dynamic update source parameter or a static update source parameter in a set category, directly loading the classification result corresponding to the service tracing classification label into a to-be-processed parameter set of the model update process, or after loading the classification result corresponding to the service tracing classification label into a pre-processing data set of a preset service relation classification model, conducting the classification result corresponding to the service tracing classification label to the to-be-processed parameter set of the model update process through the preset service relation classification model; running the model updating process to obtain depth feature distribution information output by the model updating process after performing depth feature extraction on the classification result;

determining depth feature distribution service nodes and depth feature labels of the depth feature distribution information;

under the condition that the depth feature distribution information contains the deviation feature distribution according to the depth feature distribution service node, determining relevant parameters between each depth feature label in the distribution area of the depth feature distribution information and each depth feature label in the deviation feature distribution of the depth feature distribution information according to the depth feature labels in the deviation feature distribution of the depth feature distribution information and the distribution mode information of the depth feature labels, and distributing the depth feature labels in the distribution area of the depth feature distribution information and the depth feature labels in the deviation feature distribution;

extracting a first feature vector corresponding to the depth feature label deviating from the feature distribution and a second feature vector corresponding to the depth feature label under the distribution area, determining business relation extraction queue features of the first feature vector and the second feature vector, inputting the business relation extraction queue features into the preset business relation classification model, and obtaining a classification result of the preset business relation classification model for classifying and outputting the business relation extraction queue features;

and obtaining the business relation extraction information of the business service link according to the classification result.

In a possible implementation manner of the first aspect, the performing, by the service relationship extraction information based on the service link, service push configuration update on the subscription service to obtain service push configuration optimization information of the service link includes:

acquiring service push configuration information of the service link under the subscription service;

acquiring service push configuration items under the service push configuration information and service relation configuration information corresponding to each service push configuration item;

and overlaying and configuring the business relation extraction information of the business service link under the business relation configuration information corresponding to each business pushing configuration project to obtain the business pushing configuration optimization information of the business service link.

In a second aspect, an embodiment of the present application further provides an internet platform information pushing apparatus based on big data positioning, which is applied to a cloud computing center, where the cloud computing center is in communication connection with a plurality of service terminals, and the apparatus includes:

the first acquisition module is used for acquiring service pushing configuration optimization information of a service link in big data service positioning data of the service terminal;

a second obtaining module, configured to obtain, based on the service push configuration optimization information, a recommended object to be indexed that matches multiple recommended page nodes and a target index data source corresponding to the recommended object to be indexed, where the target index data source is an index data source of a service tag to which content of the recommended object to be indexed belongs, and the target index data source includes at least one data source recommended page;

the generation module is used for combining a plurality of page nodes to be recommended to obtain a selected recommendation page which is subject-related to at least one data source recommendation page, and generating a recommendation weight parameter between the selected recommendation page and a target data source recommendation page according to subject similarity of the selected recommendation page and the at least one data source recommendation page under a target user subscription tag;

and the pushing module is used for fusing recommendation weight parameters between the selected recommendation page and the target data source recommendation page under each user subscription tag, selecting a target recommendation material matched with the recommendation object to be indexed from a preset target recommendation material library according to a fusion result, and pushing the target recommendation material to the business service terminal.

In a third aspect, an embodiment of the present application further provides an internet platform information pushing system based on big data positioning, where the internet platform information pushing system based on big data positioning includes a cloud computing center and a plurality of service terminals in communication connection with the cloud computing center;

the cloud computing center is used for:

In a fourth aspect, an embodiment of the present application further provides a cloud computing center, where the cloud computing center includes a processor, a machine-readable storage medium, and a network interface, where the machine-readable storage medium, the network interface, and the processor are connected through a bus system, the network interface is used for being communicatively connected to at least one business service terminal, the machine-readable storage medium is used for storing a program, an instruction, or a code, and the processor is used for executing the program, the instruction, or the code in the machine-readable storage medium to execute the internet platform information pushing method based on big data positioning in the first aspect or any one of possible implementation manners in the first aspect.

In a fifth aspect, an embodiment of the present application provides a computer-readable storage medium, where instructions are stored in the computer-readable storage medium, and when the instructions are executed, the computer is caused to execute the method for pushing internet platform information based on big data positioning in the first aspect or any one of the possible implementations of the first aspect.

Based on any one of the above aspects, after the recommendation weight parameters between the selected recommendation page and the target data source recommendation page are generated according to the topic similarity of the selected recommendation page and the target data source recommendation page under at least one data source recommendation page, by fusing the recommendation weight parameters between the selected recommendation page and the target data source recommendation page under the subscription tag of each user, a large number of reference bases based on the recommendation pages can be utilized, so that more target data source recommendation pages are obtained, the accuracy of subsequent material matching is improved, and the error condition when the recommendation pages are taken as the minimum basic unit can be avoided when the recommendation pages are indexed, so that the accuracy of material matching is improved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that need to be called in the embodiments are briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.

Fig. 1 is a schematic view of an application scenario of an internet platform information push system based on big data positioning according to an embodiment of the present application;

fig. 2 is a schematic flowchart of an internet platform information pushing method based on big data positioning according to an embodiment of the present application;

fig. 3 is a schematic functional module diagram of an internet platform information pushing apparatus based on big data positioning according to an embodiment of the present application;

fig. 4 is a schematic block diagram of structural components of a cloud computing center for implementing the above internet platform information push method based on big data positioning according to an embodiment of the present application.

Detailed Description

The present application will now be described in detail with reference to the drawings, and the specific operations in the method embodiments may also be applied to the apparatus embodiments or the system embodiments.

Fig. 1 is an interaction diagram of an internet platform information push system 10 based on big data positioning according to an embodiment of the present application. The internet platform information pushing system 10 based on big data positioning may include a cloud computing center 100 and a business service terminal 200 communicatively connected to the cloud computing center 100. The big data positioning based internet platform information push system 10 shown in fig. 1 is only one possible example, and in other possible embodiments, the big data positioning based internet platform information push system 10 may also include only one of the components shown in fig. 1 or may also include other components.

Based on the inventive concept of the technical solution provided by the present application, the cloud computing center 100 provided by the present application can be applied to scenes such as smart medical care, smart city management, smart industrial internet, general service monitoring management, and the like, in which a big data technology or a cloud computing technology can be applied, and for example, the cloud computing center can also be applied to scenes such as but not limited to new energy automobile system management, smart cloud office, cloud platform data processing, cloud game data processing, cloud live broadcast processing, cloud automobile management platform, block chain financial data service platform, and the like, but is not limited thereto.

In this embodiment, the cloud computing center 100 and the business service terminal 200 in the internet platform information pushing system 10 based on big data positioning may cooperatively execute the internet platform information pushing method based on big data positioning described in the following method embodiment, and the detailed description of the following method embodiment may be referred to for the specific execution steps of the cloud computing center 100 and the business service terminal 200.

In order to solve the technical problem in the foregoing background art, fig. 2 is a schematic flow chart of the internet platform information pushing method based on big data positioning according to the embodiment of the present application, and the internet platform information pushing method based on big data positioning according to the embodiment of the present application may be executed by the cloud computing center 100 shown in fig. 1, and the internet platform information pushing method based on big data positioning is described in detail below.

Step S110, obtaining service push configuration optimization information of a service link in big data service positioning data of a service terminal.

Step S120, acquiring recommendation objects to be indexed matching with the plurality of page nodes to be recommended and target index data sources corresponding to the recommendation objects to be indexed based on the service push configuration optimization information.

In this embodiment, the target index data source may be understood as an index data source of a service tag to which content of a recommended object to be indexed belongs, where the target index data source includes at least one data source recommendation page. In an alternative implementation manner, the service push configuration optimization information may include one or more to-be-recommended page nodes, where a to-be-recommended page node may be understood as a page content area included in one recommended page, a recommended page may be understood as a complete content service interface displayed when finally pushed by a user, and a page content area may refer to an area where information is displayed for content in a certain service push area in the complete content service interface.

In addition, the recommendation object to be indexed matching each page node to be recommended may be obtained from the current real-time index recommendation object library on the basis of the page node to be recommended, or obtained from a pre-configured index recommendation object library, which is not limited specifically. In addition, the target index data source corresponding to the recommended object to be indexed may be obtained based on a data source parameter (for example, an SDK parameter for invoking a data interface) associated in advance with each recommended object to be indexed.

Step S130, combining the multiple page nodes to be recommended to obtain a selected recommendation page which is subject-related to the at least one data source recommendation page, and generating a recommendation weight parameter between the selected recommendation page and the target data source recommendation page according to subject similarity of the selected recommendation page and the at least one data source recommendation page under a target user subscription tag.

In this embodiment, the topic similarity may be understood as a confidence that the selected recommendation page and at least one data source recommendation page belong to the same pushed topic under the target user subscription tag, and a specific calculation manner may refer to an existing common topic algorithm model. In addition, the recommendation weight parameter may be used to represent the relative weight of the selected recommendation page and the target data source recommendation page that may be recommended by the information.

Step S140, fusing recommendation weight parameters between the selected recommendation page and the target data source recommendation page under each user subscription label, selecting a target recommendation material matched with the object to be indexed from a preset target recommendation material library according to the fusion result, and pushing the target recommendation material to the business service terminal 200.

In this embodiment, the recommended material may be understood as specific material content that is finally pushed to the business service terminal, and may include, but is not limited to, current affair material content, e-commerce material content, short video material content, and the like.

For example, in the process of selecting a target recommendation material matched with a recommendation object to be indexed from a preset target recommendation material library according to the fusion result and pushing the target recommendation material to the business service terminal 200, a target index recommendation object matched with a recommendation page node included in a selected recommendation page with a recommendation weight parameter greater than the preset recommendation weight parameter can be determined, and then the target recommendation materials corresponding to the target index recommendation objects are obtained from the preset target recommendation material library and pushed.

Based on the above steps, in this embodiment, after the recommendation weight parameter between the selected recommendation page and the target data source recommendation page is generated according to the topic similarity of the selected recommendation page and the target data source recommendation page under at least one data source recommendation page, by fusing the recommendation weight parameters between the selected recommendation page and the target data source recommendation page under the subscription tag of each user, a large number of reference bases based on the recommendation pages can be used, so that more target data source recommendation pages are obtained, which is beneficial to improving the accuracy of subsequent material matching, and the situation of errors occurring when indexing is performed on the recommendation object to be indexed can be avoided when the recommendation page is taken as the minimum basic unit, so that the accuracy of material matching is improved.

In one possible implementation manner, for step S130, in the process of generating the recommendation weight parameter between the selected recommendation page and the target data source recommendation page according to the topic similarity between the selected recommendation page and the at least one data source recommendation page under the target user subscription tag, the following exemplary sub-steps may be implemented, which are described in detail below.

And a substep S131, determining a target user subscription tag corresponding to each data source recommendation page according to the topic incidence relation between the selected recommendation page and the data source recommendation page.

And a substep S132, based on the determined target user subscription tag, calculating the topic similarity of the selected recommendation page and at least one data source recommendation page in the determined target user subscription tag, and determining the data source recommendation page with the maximum topic similarity as the target data source recommendation page.

And a substep S133, generating a recommendation weight parameter between the selected recommendation page and the target data source recommendation page according to the topic similarity of the selected recommendation page and the target data source recommendation page under at least one user subscription label.

For example, in the sub-step S132, the following three implementation manners can be implemented.

And (I) calculating the first topic similarity of the selected recommendation page and at least one data source recommendation page under the same topic subscription tag, and determining the data source recommendation page with the maximum first topic similarity as a first target data source recommendation page.

And (II) calculating second topic similarity of the selected recommendation page and at least one data source recommendation page under the hierarchical topic subscription tag, and determining the data source recommendation page with the maximum second topic similarity as a second target data source recommendation page.

And (III) calculating the third topic similarity of the selected recommendation page and at least one data source recommendation page under the weight topic subscription tag, and determining the data source recommendation page with the maximum third topic similarity as a third target data source recommendation page.

In the step (one), a same-subject list set can be selected in the target index data source, and the same-subject list set can comprise a plurality of same-subject lists, and each same-subject list comprises at least two data source recommendation pages with the same subject meaning. Then, determining a same-subject list which is the same as the recommended word vector label of the selected recommended page, and obtaining a target same-subject list. On the basis, calculating the first topic similarity between the selected recommendation page and each data source recommendation page in the target same-topic list, and determining the data source recommendation page with the maximum first topic similarity as a first target data source recommendation page.

In this way, in sub-step S133, a first recommendation weight parameter between the selected recommendation page and the first target data source recommendation page may be generated according to the first topic similarity between the selected recommendation page and the first target data source recommendation page.

Or, in the second step, the hierarchical topic relationship between the selected recommended page and the at least one data source recommended page may be determined according to the recommended word vector tag of the selected recommended page and the recommended word vector tag of each data source recommended page. Then, based on the determined hierarchical topic relation, calculating a second topic similarity between the selected recommendation page and the corresponding upper data source recommendation page, and determining the data source recommendation page with the maximum second topic similarity as a second target data source recommendation page.

In this way, in sub-step S133, a second recommendation weight parameter between the selected recommendation page and the second target data source recommendation page may be generated according to a second topic similarity between the selected recommendation page and the second target data source recommendation page.

Or, in the third step, a theme weight value pre-established for each data source recommendation page may be collected, then the similarity between the selected recommendation page and each data source recommendation page is calculated, and the data source recommendation page with the similarity greater than a preset value is determined as the hot recommendation page to be selected. On the basis, the third topic similarity of the selected recommendation page and the hot recommendation page to be selected with the topic weight value larger than the preset weight can be calculated, and the hot recommendation page to be selected with the maximum topic similarity is determined as the third target data source recommendation page.

In this way, in sub-step S133, a third recommendation weight parameter between the selected recommendation page and the third target data source recommendation page may be generated according to a third topic similarity between the selected recommendation page and the third target data source recommendation page.

Further, in a possible implementation manner, for step S140, in the process of fusing the recommendation weight parameter between the selected recommendation page and the target data source recommendation page under each user subscription tag, the following exemplary embodiments may be implemented, and the following detailed description is given below.

And a substep S141, obtaining preset weight coefficients corresponding to the subscription tags of each user.

And step S142, calculating the product of the obtained weight coefficient and the recommendation weight parameter between the selected recommendation page and the target data source recommendation page under the corresponding user subscription label to obtain the weighted recommendation weight parameter corresponding to each user subscription label.

And a substep S143, fusing the weighted recommendation weight parameters corresponding to the subscription labels of the users.

Further, in a possible implementation manner, for step S140, in the process of combining a plurality of to-be-recommended page nodes to obtain a selected recommended page associated with the subject of the at least one data source recommended page, the following exemplary embodiments may be implemented, and the following detailed description is given below.

And a substep S131, identifying the page pushing category of each page node to be recommended.

And a substep S132, removing the page nodes to be recommended with the page pushing category being the blacklist recommendation category, and performing permutation and combination on the reserved page nodes to be recommended to obtain a selected recommendation page which is associated with the theme of the at least one data source recommendation page.

For example, the permutation and combination mode may refer to a rule or configuration information of a page layout for the current subscription service, or perform adaptive permutation and combination according to the size of each page node to be recommended, all of which are within the protection scope of the embodiment of the present application, so that the selected recommended page associated with the subject of the at least one data source recommended page may be obtained.

In a possible implementation manner, for step S110, in the process of obtaining the service push configuration optimization information of the service link in the big data service positioning data of the service terminal, the following exemplary sub-steps may be implemented, which are described in detail as follows

Step S111, acquiring the big data service positioning data of the service terminal 200, and performing service flow label extraction processing on the big data service positioning data to obtain service flow label information of the service link in the big data service positioning data.

And step S112, mining the user intention characteristics based on the service flow label information of the service link to obtain the target user intention characteristics of the service link.

And step S113, carrying out user portrait feature extraction processing on the big data service positioning data based on the artificial intelligence model to obtain user portrait feature information of the service link.

Step S114, extracting the service relationship between the target user intention characteristic of the service link in the big data service positioning data and the user portrait characteristic information of the service link to obtain the service relationship extraction information of the service link, and updating the service push configuration of the subscription service based on the service relationship extraction information of the service link to obtain the service push configuration optimization information of the service link.

In this embodiment, the big data service location data may be understood as a data set formed by service data recorded in each service link collected based on a single subscription service. Here, the subscription service may refer to a service item (e.g., a home service item, a learning education service item, etc.) subscribed by the service terminal 200 in using any internet service (e.g., a group purchase service, a forum service, an e-commerce service, etc.). The service link may be a link formed by service nodes formed under the subscription service, and the service node may be understood as a service function existing under the service item, and an association relationship generally exists between different service functions, so that each service node having an association relationship may be formed into a service link based on the association relationship. In addition, the traffic flow tag information may be used to characterize user preferences associated with traffic flow data.

In this embodiment, the user intention characteristics may be used to represent the attention or interest points of the user for the content services provided by the respective service function portions, and the user portrait characteristic information may be used to describe behavior rule conditions of the attention or interest points of the user.

In this embodiment, after obtaining the service push configuration optimization information of the service link, the subsequent service information push operation can be performed based on the service push configuration optimization information of the service link, so that in this embodiment, the target user intention characteristic of the service link and the user portrait characteristic information of the service link are extracted through the service relationship, the service flow label information of the service link and the user portrait characteristic information of the service link are integrated, the abundant service relationship characteristic information of the service link is extracted, and data support is provided for accurate information push based on big data positioning; in addition, through the intention characteristics of the target user of the business service link, business pushing configuration updating is carried out on the business service link, business pushing configuration optimization information of the business service link is obtained, and therefore the accurate information pushing process based on big data positioning is achieved.

In a possible implementation manner, for step S111, in the process of performing a service flow label extraction process on the big data service positioning data to obtain service flow label information of a service link in the big data service positioning data, the following exemplary sub-steps may be implemented, which are described in detail below.

And a substep S1111, acquiring a service flow update data list recorded by the service flow recording process of each big data corpus unit in the big data service positioning data.

In this embodiment, it is worth to be noted that the service flow update data list includes service flow update data in which each service flow update timing sequence range is taken as a recording unit range, and the service flow update data includes corpus collection trigger information and corpus pointing information of the service flow update timing sequence range, and an update corpus record in the service flow update timing sequence range. For example, the service flow update timing range may be used to represent a time recording interval related to a service flow update process, the corpus collection trigger information may be used to represent a trigger node during corpus collection (for example, a click or browsing operation of a user may be used as a trigger node), and the corpus pointing information may be used to represent service skip information indicated after corpus collection.

Substep S1112, updating the time sequence range for each service flow, updating the corpus for each behavior in the plurality of behavior updating corpus in the updating corpus record of the time sequence range for each big data corpus unit according to each behavior updating corpus, determining whether each behavior corpus description information in the behavior updating corpus is the candidate target behavior corpus description information according to the description feature vector of the behavior updating corpus description information, determining each candidate corpus tagging corpus unit segment corresponding to the behavior updating corpus according to the number of the candidate target behavior corpus description information in the behavior updating corpus, dividing the candidate corpus tagging unit segment into a plurality of sub corpus unit segments for each candidate corpus tagging unit segment, and dividing the description tag of the corpus description information and the preset tag range for each row in each sub corpus unit segment, and determining whether the candidate corpus tagging corpus unit segment is a target corpus tagging corpus unit segment, wherein each behavior corpus description information corresponds to each corpus unit updating behavior.

And a substep S1113, obtaining service flow node information of each behavior corpus description information in the preset service flow rule matching target corpus tagging corpus unit segment, wherein the service flow node information comprises corpus unit calling information and corpus unit subscription information, and the preset service flow rule comprises matching modes corresponding to different corpus unit use types.

Substep S1114, determining a corpus unit subject feature of each corpus unit subject intention topology tree and a corpus unit abstract feature of each corpus unit abstract intention topology tree according to the service flow node information of each update corpus record of each different service flow update time sequence range in the service flow update data list, labeling the corpus unit subject feature of each corpus unit subject intention topology tree and the corpus unit abstract feature of each corpus unit abstract intention topology tree in the corpus unit segment according to the target corpus, determining a corpus unit tag object of each big data corpus unit in the service flow update time sequence range, using the feature in the corpus unit range of the corpus unit tag object and the feature outside the corpus unit range of the corpus unit tag object and related to the corpus unit tag object as the corpus tag feature of each big data corpus unit in the service flow update time sequence range, and summarizing the corpus tag characteristics of each big data corpus unit in the updating time sequence range of all service flows to obtain the service flow tag information of the service link in the big data service positioning data.

In a possible implementation manner, for step S112, in the process of mining the user intention characteristic based on the traffic flow label information of the traffic service link to obtain the target user intention characteristic of the traffic service link, the following exemplary sub-steps may be implemented, which are described in detail below.

In the sub-step S1121, retrieving instruction request information of a user retrieving instruction set added to the corpus description information of each corpus tag feature in the traffic flow tag information of the traffic service link is obtained, and a first retrieving behavior list corresponding to the retrieving instruction request information is determined, where the retrieving instruction request information includes access intention object information of intention word retrieving information determined according to intention word input information and intention word output information of the user retrieving instruction set, and the first retrieving behavior list includes a high-low order of multiple access intention levels of the access intention object information.

In sub-step S1122, it is determined that the corpus descriptive information of each corpus tag feature is based on the first intention word vector of the intention word input information and the second intention word vector of the intention word output information.

And a substep S1123 of determining decision tree mining parameters for performing decision tree mining on the first search behavior list according to the access intention level relation of the first intention word vector and the second intention word vector.

And a substep S1124, performing decision tree mining on the first search behavior list based on the decision tree mining parameter to obtain a second search behavior list.

And S1125, clustering the second search behavior list to obtain a plurality of search category cluster clusters, and extracting the features of each search category cluster to obtain a cluster feature vector.

And a substep S1126, determining the user intention characteristic of each corpus tag characteristic according to the user intention characteristics corresponding to the plurality of cluster characteristic vectors corresponding to the second retrieval behavior list.

And a substep S1127 of obtaining a target user intention characteristic of the business service link based on the user intention characteristic of each corpus tag characteristic.

Further, in a possible implementation manner, for step S113, in the process of performing user portrait feature extraction processing on big data service positioning data based on an artificial intelligence model to obtain user portrait feature information of a service link, the following exemplary sub-steps may be implemented, which are described in detail below.

And a substep S1131, inputting the big data service positioning data into a pre-trained artificial intelligence model, and obtaining a confidence coefficient that the big data service positioning data is matched with each preset user portrait.

It is worth explaining that the artificial intelligence model is obtained by training based on training samples and training label information corresponding to the training samples, the training samples are big data service positioning data samples, and the training label information is user portrait label information. The specific training process may refer to a conventional training mode provided in the prior art, and the training process does not belong to the technical problem intended to be solved by the embodiment of the present application, and is not described in detail herein.

And a substep S1132 of determining a target user portrait corresponding to the big data service positioning data according to the confidence level that the big data service positioning data is matched with each preset user portrait.

For example, a predetermined user image with a confidence level greater than a predetermined confidence level threshold may be determined as the target user image corresponding to the big data service positioning data.

The substep S1133 is to extract the user profile feature information matched with each service link from the user profile description information of the target user profile corresponding to the big data service positioning data.

For example, in the extraction process, the feature information with the structured description information matching each service link in the user profile description information can be specifically extracted.

In a possible implementation manner, for step S114, in the process of performing business relationship extraction on the target user intention feature of the business service link in the big data business positioning data and the user portrait feature information of the business service link to obtain the business relationship extraction information of the business service link, the following exemplary sub-steps may be implemented, which are described in detail below.

And a substep S1141 of adding the target user intention characteristic and the user portrait characteristic information of the business service link into the business relationship extraction space, and determining a business traceability script of the target user intention characteristic and the user portrait characteristic information of the business service link corresponding to each business relationship extraction node from the business relationship extraction space.

And a substep S1142 of clustering the service tracing scripts according to the relationship network labels of the service relationships among the service relationship extraction nodes in the different service tracing scripts to obtain at least one service tracing script set.

In this embodiment, the relationship network tag of the service relationship extraction node in any two service tracing scripts in the same service tracing script set covers the preset tag range.

And a substep S1143 of determining a service tracing characteristic of the service tracing script set corresponding to the target user intention characteristic of the service link and the user portrait characteristic information from the service relationship extraction space based on each service tracing script in the service tracing script set aiming at each service tracing script set.

In this embodiment, the service tracing characteristics at least include characteristic information data of each service tracing entry of each service tracing script in the service tracing script set for a target user intention characteristic and user portrait characteristic information of a service link, and the service tracing characteristics are used to determine a service relationship extracting node in each service tracing script in the service tracing script set for a service relationship tracing result corresponding to the target user intention characteristic and the user portrait characteristic information of the service link.

And a substep S1144 of determining a service relationship tracing result corresponding to the service relationship extraction node in each service tracing script in the service tracing script set based on the service tracing characteristics, classifying the service relationship tracing result by adopting a preset service relationship classification model to obtain a classification result, and obtaining service relationship extraction information of the service link according to the classification result.

Illustratively, in a possible example, for the sub-step S1144, in the process of determining, based on the service tracing characteristics, the service relationship tracing result corresponding to the service relationship extracting node in each service tracing script in the service tracing script set, the following alternative embodiments may be implemented.

And a substep S11441 of determining traceability service track information and traceability service types of service traceability features, determining a plurality of traceability service linkage combinations according to a plurality of service relationship indexes in historical service relationship vector results, performing similarity calculation according to service matching degrees of the service relationship indexes in each traceability service linkage combination and the traceability service types in the traceability service track information to obtain a plurality of similarity calculation results respectively corresponding to the plurality of traceability service linkage combinations, and calling a frequent weighted value of the service of the traceability service linkage combination corresponding to each similarity calculation result as a weighted value of each similarity calculation result.

And a substep S11442 of respectively obtaining similar service relationship tracks of each service tracing script corresponding to the tracing service category determined in the tracing service track information based on the plurality of similarity calculation results to obtain a plurality of similar service relationship tracks, and integrating the plurality of similar service relationship tracks obtained based on the plurality of similarity calculation results according to the weight value of each similarity calculation result to obtain a first similar service relationship track set.

It is worth explaining that the service calling frequency weighted values between any two service relation indexes in each tracing service linkage combination are equal, the service calling frequency weighted values corresponding to different tracing service linkage combinations are different, each similarity calculation result is used for determining service coincidence data corresponding to a service relation extraction node in each service tracing script corresponding to a tracing service category in any set timing segment, and the similarity calculation result is obtained by performing similarity calculation according to a historical service relation vector result.

And a substep S11443 of obtaining a service relationship extraction identification result of the plurality of service relationship extraction nodes between the tracing service track information and the tracing service type according to the historical service relationship vector result and the service coincidence data of the plurality of service relationship extraction nodes corresponding to the tracing service type, and taking a target request message corresponding to the service relationship extraction identification result corresponding to the plurality of service relationship extraction nodes as a second similar service relationship track.

And a substep S11444, calculating service relation related parameters of the first similar service relation track set and the second similar service relation track, and determining a service relation tracing result corresponding to each service relation extraction node according to the service relation related parameters and the service coincidence data of each service relation extraction node.

For example, in an alternative embodiment, for sub-step S11444, the following embodiment may be implemented.

(1) And determining knowledge graph information of the service relationship circle network corresponding to each service relationship extraction node based on the service relationship related parameters, and determining service triggering index information of each service relationship extraction node according to the service coverage area of the service coincidence data of each service relationship extraction node in the corresponding service tracing script.

(2) Extracting a first user intention characteristic sequence corresponding to the knowledge graph information and a second user intention characteristic sequence corresponding to the service triggering index information, and determining a plurality of service relation meta-knowledge networks with different characteristic description values respectively included in the first user intention characteristic sequence and the second user intention characteristic sequence.

The business relation meta-knowledge network can be understood as a meta-knowledge network formed by user intention characteristics with incidence relation in the first user intention characteristic sequence and the second user intention characteristic sequence.

(3) On the premise that the service relationship between the entities of the first user intention characteristic sequence is the same as the service relationship between the entities of the second user intention characteristic sequence, obtaining basic relationship map information of the knowledge map information in any service relationship element knowledge network of the first user intention characteristic sequence, and determining the service relationship element knowledge network with the minimum feature description value in the second user intention characteristic sequence as a target service relationship element knowledge network in parallel.

(4) And adding the basic relationship map information to a target service relationship meta-knowledge network based on the extraction strategy of the service relationship extraction field of each service relationship extraction node, and determining service upstream and downstream relationship information corresponding to the basic relationship map information in the target service relationship meta-knowledge network.

(5) And generating a knowledge graph connecting edge between the knowledge graph information and the service triggering index information according to the service difference information between the basic relationship graph information and the service upstream and downstream relationship information.

(6) And obtaining target entity member information in a target service relationship element knowledge network by taking the service upstream and downstream relationship information as reference information, adding the target entity member information to the service relationship element knowledge network in which the basic relationship map information is positioned according to a plurality of information matching paths corresponding to the connected edges of the knowledge map, obtaining a member matching result corresponding to the target entity member information in the service relationship element knowledge network in which the basic relationship map information is positioned, and determining the member matching result as a service tracing result according to the reference information.

The target entity member information may be understood as information of an entity member corresponding to the access action.

(7) And determining a corresponding service relation result when the basic relation map information is added to the target service relation meta-knowledge network.

(8) According to the association degree between the member matching result and the service relation feature labels corresponding to the service relation traceability nodes in the service relation result, the service traceability labels corresponding to the service traceability result are sequentially obtained in the second user intention feature sequence according to the sequence of the time sequence weight from big to small, until the number of the intellectual property network nodes of the service relation element intellectual network where the obtained service traceability labels are located is consistent with the number of the intellectual property network nodes of the service traceability results in the first user intention feature sequence, the service traceability labels in the next service relation element intellectual network are stopped to be obtained, the traceability association information between the service traceability results and the last obtained service traceability labels is established, and the service relation traceability result corresponding to each service relation extraction node is determined based on the traceability association information.

Illustratively, in a possible example, for the sub-step S1144, in the process of classifying the service relationship tracing result by using the preset service relationship classification model to obtain the classification result, and obtaining the service relationship extraction information of the service link according to the classification result, the following alternative implementation may be implemented.

And a substep S11445 of determining a business relation floating parameter of a sample sequence of the preset business relation classification model, updating the sample sequence according to a previous business relation classification result when the business relation floating parameter does not reach a set floating range, inputting a business relation traceability result into the preset business relation classification model when the business relation floating parameter reaches the set floating range, and judging whether a classification result output by the preset business relation classification model is received. The previous time period may be determined according to the set time period, and is not limited specifically.

And a substep S11446, when receiving the classification result output by the preset business relation classification model, creating a model updating process according to the classification result.

In the substep S11447, when the preset business relationship classification model outputs the classification result corresponding to the business tracing classification label, determining an update source parameter of the classification result corresponding to the business tracing classification label in the model update process.

And in the substep S11448, when it is determined that the update source parameter of the classification result corresponding to the service tracing classification tag in the model update process is the dynamic update source parameter or the static update source parameter in the set category, directly loading the classification result corresponding to the service tracing classification tag into the to-be-processed parameter set of the model update process, or after loading the classification result corresponding to the service tracing classification tag into the pre-processing data set of the preset service relation classification model, conducting the classification result corresponding to the service tracing classification tag to the to-be-processed parameter set of the model update process through the preset service relation classification model. And running a model updating process, and acquiring depth feature distribution information output after the model updating process carries out depth feature extraction on the classification result.

And a substep S11449 of determining a depth feature distribution service node and each depth feature label of the depth feature distribution information.

In the sub-step S114491, when it is determined that the depth feature distribution information includes the deviating feature distribution according to the depth feature distribution service node, the related parameters between each depth feature label in the distribution region of the depth feature distribution information and each depth feature label in the deviating feature distribution of the depth feature distribution information are determined according to the depth feature label in the deviating feature distribution of the depth feature distribution information and the distribution mode information thereof, and the depth feature label in the distribution region of the depth feature distribution information related to the depth feature label in the deviating feature distribution is allocated to the deviating feature distribution.

And a substep S114492, extracting a first feature vector corresponding to the depth feature label deviating from the feature distribution and a second feature vector corresponding to the depth feature label in the distribution region, determining a business relation extraction queue feature of the first feature vector and the second feature vector, inputting the business relation extraction queue feature into a preset business relation classification model, and obtaining a classification result of the preset business relation classification model for classifying and outputting the business relation extraction queue feature.

And a substep S114493 of obtaining service relationship extraction information of the service link according to the classification result.

In a possible implementation manner, still referring to step S114, in the process of performing service push configuration update on the subscription service based on the service relationship extraction information of the service link to obtain the service push configuration optimization information of the service link, the following exemplary sub-steps may be implemented, which are described in detail below.

And a substep S1145 of obtaining service push configuration information of the service link under the subscription service.

And a substep S1146 of obtaining the service push configuration items under the service push configuration information and the service relationship configuration information corresponding to each service push configuration item.

And a substep S1147 of overlaying and configuring the service relationship extraction information of the service link under the service relationship configuration information corresponding to each service push configuration item to obtain service push configuration optimization information of the service link.

Fig. 3 is a schematic diagram of functional modules of the internet platform information pushing apparatus 300 based on big data positioning according to the embodiment of the present disclosure, and in this embodiment, the internet platform information pushing apparatus 300 based on big data positioning may be divided into the functional modules according to the method embodiment executed by the cloud computing center 100, that is, the following functional modules corresponding to the internet platform information pushing apparatus 300 based on big data positioning may be used to execute the method embodiments executed by the cloud computing center 100. The internet platform information pushing apparatus 300 based on big data positioning may include a first obtaining module 310, a second obtaining module 320, a generating module 330, and a pushing module 340, and the functions of the functional modules of the internet platform information pushing apparatus 300 based on big data positioning are described in detail below.

The first obtaining module 310 is configured to obtain service push configuration optimization information of a service link in big data service positioning data of a service terminal. The first obtaining module 310 may be configured to perform the step S110, and for a detailed implementation of the first obtaining module 310, reference may be made to the detailed description of the step S110.

The second obtaining module 320 is configured to obtain, based on the service push configuration optimization information, a recommended object to be indexed matching with the multiple recommended page nodes and a target index data source corresponding to the recommended object to be indexed, where the target index data source is an index data source of a service tag to which content of the recommended object to be indexed belongs, and the target index data source includes at least one data source recommended page. The second obtaining module 320 may be configured to perform the step S120, and for a detailed implementation of the second obtaining module 320, reference may be made to the detailed description of the step S120.

The generating module 330 is configured to combine multiple page nodes to be recommended to obtain a selected recommendation page having a theme association with at least one data source recommendation page, and generate a recommendation weight parameter between the selected recommendation page and a target data source recommendation page according to theme similarity between the selected recommendation page and the at least one data source recommendation page under a target user subscription tag. The generating module 330 may be configured to execute the step S130, and the detailed implementation of the generating module 330 may refer to the detailed description of the step S130.

And the pushing module 340 is configured to fuse recommendation weight parameters between the selected recommendation page and the target data source recommendation page under each user subscription tag, select a matched target recommendation material from a preset target recommendation material library according to a fusion result, and push the target recommendation material to the service terminal. The pushing module 340 may be configured to perform the step S140, and the detailed implementation manner of the pushing module 340 may refer to the detailed description of the step S140.

It should be noted that the division of the modules of the above apparatus is only a logical division, and the actual implementation may be wholly or partially integrated into one physical entity, or may be physically separated. And these modules may all be implemented in software invoked by a processing element. Or may be implemented entirely in hardware. And part of the modules can be realized in the form of calling software by the processing element, and part of the modules can be realized in the form of hardware. For example, the first obtaining module 310 may be a separate processing element, or may be integrated into a chip of the apparatus, or may be stored in a memory of the apparatus in the form of program code, and a processing element of the apparatus calls and executes the functions of the first obtaining module 310. Other modules are implemented similarly. In addition, all or part of the modules can be integrated together or can be independently realized. The processing element described herein may be an integrated circuit having signal processing capabilities. In implementation, each step of the above method or each module above may be implemented by an integrated logic circuit of hardware in a processor element or an instruction in the form of software.

Fig. 4 is a schematic diagram illustrating a hardware structure of a cloud computing center 100 for implementing the above-mentioned internet platform information pushing method based on big data positioning according to an embodiment of the present disclosure, and as shown in fig. 4, the cloud computing center 100 may include a processor 110, a machine-readable storage medium 120, a bus 130, and a transceiver 140.

In a specific implementation process, at least one processor 110 executes computer-executable instructions stored in a machine-readable storage medium 120 (for example, a first obtaining module 310, a second obtaining module 320, a generating module 330, and a pushing module 340 included in the internet platform information pushing apparatus 300 based on big data positioning shown in fig. 3), so that the processor 110 may execute the internet platform information pushing method based on big data positioning according to the above method embodiment, where the processor 110, the machine-readable storage medium 120, and the transceiver 140 are connected through the bus 130, and the processor 110 may be configured to control a transceiving action of the transceiver 140, so as to perform data transceiving with the business service terminal 200.

For a specific implementation process of the processor 110, reference may be made to the above-mentioned method embodiments executed by the cloud computing center 100, and implementation principles and technical effects are similar, which are not described herein again.

In the embodiment shown in fig. 4, it should be understood that the Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in connection with the present invention may be embodied directly in a hardware processor, or in a combination of the hardware and software modules within the processor.

The machine-readable storage medium 120 may comprise high-speed RAM memory and may also include non-volatile storage NVM, such as at least one disk memory.

The bus 130 may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended ISA (EISA) bus, or the like. The bus 130 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, the buses in the figures of the present application are not limited to only one bus or one type of bus.

In addition, a readable storage medium is provided, where the readable storage medium stores computer execution instructions, and when a processor executes the computer execution instructions, the internet platform information pushing method based on big data positioning is implemented as above.

The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.

Having thus described the basic concept, it will be apparent to those skilled in the art that the foregoing detailed disclosure is to be regarded as illustrative only and not as limiting the present specification. Various modifications, improvements and adaptations to the present description may occur to those skilled in the art, although not explicitly described herein. Such modifications, improvements and adaptations are proposed in the present specification and thus fall within the spirit and scope of the exemplary embodiments of the present specification.

Also, particular push elements are used in this description to describe embodiments of this description. Reference to "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic described in connection with at least one embodiment of the specification. Therefore, it is emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, some features, structures, or characteristics of one or more embodiments of the specification may be combined as appropriate.

Moreover, those skilled in the art will appreciate that aspects of the present description may be illustrated and described in terms of several patentable species or contexts, including any new and useful combination of processes, machines, manufacture, or materials, or any new and useful improvement thereof. Accordingly, aspects of this description may be performed entirely by hardware, entirely by software (including firmware, resident software, micro-code, etc.), or by a combination of hardware and software. The above hardware or software may be referred to as "data block," module, "" engine, "" unit, "" component, "or" system. Furthermore, aspects of the present description may be represented as a computer product, including computer readable program code, embodied in one or more computer readable media.

The computer storage medium may comprise a propagated data signal with the computer program code embodied therewith, for example, on baseband or as part of a carrier wave. The propagated signal may take any of a variety of forms, including electromagnetic, optical, etc., or any suitable combination. A computer storage medium may be any computer-readable medium that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, or device. Program code located on a computer storage medium may be propagated over any suitable medium, including radio, cable, fiber optic cable, RF, or the like, or any combination of the preceding.

Computer program code required for the operation of various portions of this specification may be written in any one or more programming languages, including an object oriented programming language such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C + +, C #, VB.NET, Python, and the like, a conventional programming language such as C, VisualBasic, Fortran2003, Perl, COBOL2002, PHP, ABAP, a passive programming language such as Python, Ruby, and Groovy, or other programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any network format, such as a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet), or in a cloud computing environment, or as a service, such as a software as a service (SaaS).

Additionally, the order in which the elements and sequences are processed, the use of alphanumeric characters, or the use of other designations in this specification is not intended to limit the order of the processes and methods in this specification, unless otherwise specified in the claims. While various presently contemplated embodiments of the invention have been discussed in the foregoing disclosure by way of example, it is to be understood that such detail is solely for that purpose and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements that are within the spirit and scope of the embodiments herein. For example, although the system components described above may be implemented by hardware devices, they may also be implemented by software-only solutions, such as installing the described system on an existing server or mobile device.

Similarly, it should be noted that in the preceding description of embodiments of the present specification, various features are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure aiding in the understanding of one or more of the embodiments. This method of disclosure, however, is not intended to imply that more features than are expressly recited in a claim. Indeed, the embodiments may be characterized as having less than all of the features of a single embodiment disclosed above.

Finally, it should be understood that the examples in this specification are only intended to illustrate the principles of the examples in this specification. Other variations are also possible within the scope of this description. Thus, by way of example, and not limitation, alternative configurations of the embodiments of the specification can be considered consistent with the teachings of the specification. Accordingly, the embodiments of the present description are not limited to only those embodiments explicitly described and depicted herein.

Claims

1. An information pushing method based on big data service positioning is applied to a cloud computing center, the cloud computing center is in communication connection with a plurality of service terminals, and the method comprises the following steps:

identifying the page pushing category of each page node to be recommended, removing the page nodes to be recommended with the page pushing category being a blacklist recommending category, arranging and combining the reserved page nodes to be recommended to obtain a selected recommended page which is subject-related to at least one data source recommended page, and determining a target user subscription tag corresponding to each data source recommended page according to the subject-related relationship between the selected recommended page and the data source recommended page;

generating a recommendation weight parameter between the selected recommendation page and the target data source recommendation page according to the topic similarity of the selected recommendation page and the target data source recommendation page under at least one user subscription tag;

and fusing recommendation weight parameters between the selected recommendation page and the target data source recommendation page under each user subscription tag, determining a target index recommendation object matched with a recommendation page node included in the selected recommendation page with the recommendation weight parameter larger than a preset recommendation weight parameter, and pushing after obtaining a target recommendation material corresponding to the target index recommendation object from a preset target recommendation material library.

2. The big data service positioning-based information pushing method according to claim 1, wherein the step of calculating topic similarity of the selected recommendation page and at least one data source recommendation page at the determined target user subscription tag based on the determined target user subscription tag, and determining the data source recommendation page with the largest topic similarity as the target data source recommendation page comprises:

and calculating the third topic similarity of the selected recommendation page and at least one data source recommendation page under the weight topic subscription tag, and determining the data source recommendation page with the maximum third topic similarity as a third target data source recommendation page.

3. The big data service positioning-based information pushing method according to claim 2, wherein the calculating a first topic similarity between the selected recommendation page and at least one data source recommendation page under a same topic subscription tag, and determining the data source recommendation page with the largest first topic similarity as a first target data source recommendation page includes:

and calculating the first topic similarity between the selected recommendation page and each data source recommendation page in the target same-topic list, and determining the data source recommendation page with the maximum first topic similarity as the first target data source recommendation page.

4. The big data service positioning-based information pushing method according to claim 3, wherein the generating of the recommendation weight parameter between the selected recommendation page and the target data source recommendation page according to the topic similarity of the selected recommendation page and the target data source recommendation page under at least one user subscription tag comprises:

and generating a first recommendation weight parameter between the selected recommendation page and the first target data source recommendation page according to the first theme similarity between the selected recommendation page and the first target data source recommendation page.

5. The big data service positioning-based information pushing method according to claim 2, wherein the calculating a second topic similarity between the selected recommendation page and at least one data source recommendation page under a hierarchical topic subscription tag, and determining the data source recommendation page with the largest second topic similarity as a second target data source recommendation page comprises:

and calculating second topic similarity between the selected recommended page and the corresponding upper data source recommended page based on the determined hierarchical topic relation, and determining the data source recommended page with the maximum second topic similarity as a second target data source recommended page.

6. The big data service positioning-based information pushing method according to claim 5, wherein the generating of the recommendation weight parameter between the selected recommendation page and the target data source recommendation page according to the topic similarity of the selected recommendation page and the target data source recommendation page under at least one user subscription tag comprises:

and generating a second recommendation weight parameter between the selected recommendation page and a second target data source recommendation page according to the second theme similarity between the selected recommendation page and the second target data source recommendation page.

7. The big data service positioning-based information pushing method according to claim 2, wherein the calculating a third topic similarity between the selected recommendation page and at least one data source recommendation page under the weighted topic subscription tag, and determining the data source recommendation page with the maximum topic similarity as a third target data source recommendation page includes:

and calculating the third topic similarity of the selected recommendation page and the hot recommendation page to be selected with the topic weight value larger than the preset weight, and determining the hot recommendation page to be selected with the maximum third topic similarity as a third target data source recommendation page.

8. The big data service positioning-based information pushing method according to claim 7, wherein the generating of the recommendation weight parameter between the selected recommendation page and the target data source recommendation page according to the topic similarity of the selected recommendation page and the target data source recommendation page under at least one user subscription tag comprises:

9. The big data service positioning-based information push method according to any one of claims 1 to 8, wherein the fusing recommendation weight parameters between the selected recommendation page and the target data source recommendation page under each user subscription label includes:

10. The internet platform information pushing system based on big data positioning is characterized by comprising a cloud computing center and a plurality of business service terminals in communication connection with the cloud computing center;

the cloud computing center is used for: