CN112464006A

CN112464006A - Data analysis method and system based on artificial intelligence and Internet

Info

Publication number: CN112464006A
Application number: CN202011498093.7A
Authority: CN
Inventors: 黄雨勤
Original assignee: Individual
Current assignee: Individual
Priority date: 2020-06-14
Filing date: 2020-06-14
Publication date: 2021-03-09
Also published as: CN112464007A; CN111723227B; CN111723227A

Abstract

The embodiment of the disclosure provides a data analysis method and system based on artificial intelligence and the Internet, by obtaining the graph classification grading bookmark corresponding to the target graph search object initiated by the Internet access equipment, so as to use the graph classification grading bookmark for representing the graph classification grading information of the target graph search object under the big data analysis bookmark, then, according to the graph classification grading bookmark, the big data analysis is carried out on each graph unit library in the target graph search object, therefore, information recommendation is carried out after portrait recognition, compared with the interactive process of a graphic unit library in a generalized range in the traditional scheme, big data analysis is carried out, the method can avoid the situation that the accuracy of portrait analysis is not high due to the introduction of excessive noise data, improve the pertinence of big data analysis, save cloud computing resources, improve the cloud computing performance and effectively reduce the computing amount.

Description

Data analysis method and system based on artificial intelligence and Internet

Technical Field

The disclosure relates to the technical field of artificial intelligence and the Internet, in particular to a data analysis method and system based on artificial intelligence and the Internet.

Background

With the development of big data and the internet, a user can search for a graphic search object through the internet, and in the process, big data analysis can be performed on each graphic unit library associated with the graphic search object of the user (for example, a graphic unit library accessed by the user and having continuous access relation with the graphic search object), so that an intention image of the user in the graphic search process can be effectively determined, and subsequent information recommendation is facilitated.

However, the inventor researches and discovers that the current scheme is generally to perform large data analysis on the interaction process of a graphic unit library in a broad range, and a huge calculation amount is generated due to the huge data amount of the graphic unit library, so that not only can the accuracy of portrait analysis caused by the introduction of noise data be low, but also cloud computing resources can be excessively consumed, and the cloud computing performance can be reduced.

Disclosure of Invention

In order to overcome at least the above disadvantages in the prior art, an object of the present disclosure is to provide a data analysis method and system based on artificial intelligence and the internet, which can avoid the situation of low accuracy of portrait analysis caused by introducing too much noise data, improve the pertinence of big data analysis, and simultaneously save cloud computing resources, improve cloud computing performance, and effectively reduce the amount of computation, compared with the case of performing big data analysis in the interactive process of a graphic unit library in a generalized range in the conventional scheme.

In a first aspect, the present disclosure provides a data analysis method based on artificial intelligence and the internet, which is applied to a cloud computing service platform, where the cloud computing service platform is in communication connection with a plurality of internet access devices, and the method includes:

acquiring a graph classification grading bookmark corresponding to a target graph search object initiated by the Internet access equipment, wherein the graph classification grading bookmark is used for representing graph classification grading information of the target graph search object under a big data analysis bookmark;

performing big data analysis on each graphic unit library in the target graphic search object according to the graphic classification grading bookmark to obtain graphic interaction behavior information of each graphic unit library in the target graphic search object and interaction portrait feature information corresponding to the graphic interaction behavior information;

classifying and identifying the interactive portrait feature information corresponding to the graphic interactive behavior information based on a pre-configured artificial intelligence recommendation model to obtain recommended content items matched with the interactive portrait feature information corresponding to the graphic interactive behavior information;

and pushing the hot spot recommendation information associated with the recommended content item to a graphical interaction interface of a corresponding graphical unit library.

In a possible implementation manner of the first aspect, the step of obtaining a graph classification ranking bookmark corresponding to a target graph search object initiated by the internet access device includes:

acquiring image classification labels of a target graph search object under the artificial intelligence recognition result of each artificial intelligence recognition model from the Internet access equipment, classifying the image classification labels under each artificial intelligence recognition result according to preset big data collection classification, and respectively generating an image classification label sequence of each big data collection classification;

determining a target graphic classification grade bookmark associated with each artificial intelligence identification result according to user search behavior information of the target graphic search object, and respectively determining grade label information of a first indexable grade bookmark of the target graphic classification grade bookmark in an image classification label sequence of corresponding big data collection classification aiming at the target graphic classification grade bookmark associated with each artificial intelligence identification result to obtain a first information management index sequence of the target graphic classification grade bookmark, wherein the target graphic classification grade bookmark is a graphic classification grade bookmark matched with the user search behavior information of the target graphic search object in advance;

determining the key graph classification grade bookmark associated with each artificial intelligence identification result according to the historical classification grade information of the target graph search object, respectively acquiring a second indexable classification bookmark of the key graph classification grade bookmark aiming at the key graph classification grade bookmark associated with each artificial intelligence identification result, and determining the hierarchical label information of the second indexable classification bookmark in the image classification label sequence of the corresponding big data collection classification to obtain a second information management index sequence of the key graph classification hierarchical bookmark, the key graphic classification grading bookmark is a graphic classification grading bookmark with the classification frequency greater than a set frequency threshold value in the historical classification grading information of the target graphic search object, the classification frequency is used for representing the search classification times of the graph classification grading bookmark in unit time;

and determining the graph classification grading bookmark corresponding to the target graph search object according to the matching relation between the first information management index sequence and the second information management index sequence.

In a possible implementation manner of the first aspect, the step of determining the graph classification hierarchical bookmark corresponding to the target graph search object according to a matching relationship between the first information management index sequence and the second information management index sequence includes:

matching the information management index sequence of each target graphic classification grading bookmark in the first information management index sequence with the information management index sequence of each matched key graphic classification grading bookmark in the second information management index sequence to obtain a plurality of matching degrees, wherein each matched key graphic classification grading bookmark in the second information management index sequence is matched with the arrangement sequence of the corresponding target graphic classification grading bookmark in the respective information management index sequence, and the matching degree is determined according to the contact ratio between the information management index sequence of the target graphic classification grading bookmark and the information management index sequence of the matched key graphic classification grading bookmark;

when the matching degree between the information management index sequence of any one target graphic classification grading bookmark and the information management index sequence of the matched key graphic classification grading bookmark is greater than the set matching degree, taking the target graphic classification grading bookmark and the key graphic classification grading bookmark as a big data analysis combined object;

when the matching degree between the information management index sequence of any one target graphic classification grading bookmark and the information management index sequence of the matched key graphic classification grading bookmark is not more than the set matching degree, the target graphic classification grading bookmark and the key graphic classification grading bookmark are independently used as a big data analysis independent object;

and determining the graph classification grading bookmark corresponding to the big data analysis combined object and the big data analysis independent object as the graph classification grading bookmark corresponding to the target graph search object.

In a possible implementation manner of the first aspect, the step of performing big data analysis on each graphic unit library in the target graphic search object according to the graphic classification hierarchical bookmark to obtain the graphic interaction behavior information of each graphic unit library in the target graphic search object and the interaction portrait feature information corresponding to the graphic interaction behavior information includes:

acquiring at least one candidate unit data area to be analyzed and an interaction behavior recording node corresponding to each candidate unit data area from each graphic unit library in the target graphic search object according to a preset big data analysis strategy of the graphic classification grading bookmark and an interaction history record of each graphic unit library in the target graphic search object;

grouping initial past interactive behavior nodes in a past interactive behavior node sequence corresponding to the preset big data analysis strategy to obtain at least one initial past interactive behavior node sequence, regarding any initial past interactive behavior node sequence in the at least one initial past interactive behavior node sequence, regarding a past interactive behavior node corresponding to a grouping cluster center of any initial past interactive behavior node sequence as any cluster center node, regarding any past interactive behavior node except the initial past interactive behavior node, adding any past interactive behavior node to the initial past interactive behavior node sequence corresponding to the cluster center node with the maximum similarity of any past interactive behavior node, and when no past interactive behavior node which is not added to the initial past interactive behavior node sequence exists, obtaining a past interactive behavior node sequence corresponding to each cluster core node;

for an interactive behavior recording node corresponding to any candidate unit data area, determining at least one target cluster core node corresponding to the interactive behavior recording node based on the similarity between the interactive behavior recording node and each cluster core node, and taking a union of past interactive behavior node sequences corresponding to each target cluster core node as a first candidate past interactive behavior node sequence corresponding to any candidate unit data area;

acquiring graph interaction behavior information of each candidate unit data area in at least one candidate unit data area as graph interaction behavior information of each graph unit library in the target graph search object based on a first candidate past interaction behavior node sequence corresponding to each candidate unit data area in the at least one candidate unit data area;

acquiring interactive service information in the interactive process of the graph interaction behavior information of each graph unit library in the target graph search object, and converting the interactive service information into corresponding structured data information;

determining a structural feature vector of at least one structural item corresponding to the structural data information through the coding node of the graph interaction behavior information;

determining interaction processing information of corresponding interaction nodes according to the structured feature vector of the at least one structured project through the decoding nodes of the graph interaction behavior information, and converting the interaction processing information of the corresponding interaction nodes into interaction feature vectors;

generating a portrait label corresponding to the structural feature vector of the structural project and a confidence of the portrait label according to the interactive feature vector and the corresponding fusion feature vector through a decoding node of the graph interactive behavior information;

selecting at least one portrait label to form interactive processing information corresponding to the structured data information according to the confidence of the interactive processing information;

and converting the interactive processing information into new interactive service information corresponding to the graphic interactive behavior information to obtain interactive portrait characteristic information corresponding to the graphic interactive behavior information.

In a possible implementation manner of the first aspect, the step of obtaining the graph interaction behavior information of each candidate unit data region in the at least one candidate unit data region based on the first candidate past interaction behavior node sequence corresponding to each candidate unit data region in the at least one candidate unit data region includes:

when the number of the at least one candidate unit data area is smaller than the set number, analyzing any candidate unit data area in a first candidate past interactive behavior node sequence corresponding to any candidate unit data area for any candidate unit data area to obtain the graph interactive behavior information of any candidate unit data area.

determining a whole target region sequence including a first target number of candidate unit data regions based on the at least one candidate unit data region when the number of the at least one candidate unit data region is not less than a set number;

acquiring the number of past interactive behavior nodes corresponding to any target region sequence in all the target region sequences;

dividing the at least one candidate unit data region into a candidate unit data region set with a second target number based on the number of past interactive behavior nodes corresponding to each target region sequence in all the target region sequences, wherein the second target number is the ratio of the number of the at least one candidate unit data region to the first target number;

for any candidate unit data region set, taking a union set of first candidate past interactive behavior node sequences corresponding to each candidate unit data region in the any candidate unit data region set as a second candidate past interactive behavior node sequence corresponding to the any candidate unit data region set;

analyzing each candidate unit data area in any candidate unit data area set in a second candidate past interaction behavior node sequence corresponding to any candidate unit data area set to obtain the graph interaction behavior information of each candidate unit data area in any candidate unit data area set.

In a possible implementation manner of the first aspect, the dividing, based on the number of past interaction behavior nodes corresponding to each target region sequence in all the target region sequences, the at least one candidate unit data region into a candidate unit data region set with a second target number includes:

determining the number of past interactive behavior nodes corresponding to each first interactive behavior node sequence based on the number of past interactive behavior nodes corresponding to each target region sequence in all the target region sequences, wherein any first interactive behavior node sequence comprises candidate unit data regions in two target region sequences meeting a first condition;

determining the number of past interactive behavior nodes corresponding to each second interactive behavior node sequence based on the number of past interactive behavior nodes corresponding to each first interactive behavior node sequence and the number of past interactive behavior nodes corresponding to each target region sequence, wherein any second interactive behavior node sequence comprises candidate unit data regions in three target region sequences meeting a second condition;

repeating the steps until the quantity of past interactive behavior nodes corresponding to the final interactive behavior node sequence comprising all the candidate unit data areas is determined;

performing reverse derivation based on the number of past interactive behavior nodes corresponding to the final interactive behavior node sequence, and determining a second target number of target region sequences corresponding to the number of past interactive behavior nodes corresponding to the final interactive behavior node sequence;

and dividing the at least one candidate unit data area into a second target number of candidate unit data area sets according to the second target number of target area sequences.

repeating the steps until the number of the past interactive behavior nodes corresponding to each intermediate interactive behavior node sequence is determined, wherein any intermediate interactive behavior node sequence comprises half the number of candidate unit data areas;

determining the number of past interactive behavior nodes corresponding to the final interactive behavior node sequence comprising all candidate unit data regions based on the number of past interactive behavior nodes corresponding to each intermediate interactive behavior node sequence;

In a possible implementation manner of the first aspect, the step of classifying and identifying the interaction image feature information corresponding to the graphical interaction behavior information based on a preconfigured artificial intelligence recommendation model to obtain a recommended content item matched with the interaction image feature information corresponding to the graphical interaction behavior information includes:

inputting the interaction portrait feature information corresponding to the graph interaction behavior information into the pre-configured artificial intelligence recommendation model, and extracting portrait recommendation feature vectors corresponding to the interaction portrait feature information corresponding to the graph interaction behavior information;

inputting the portrait recommendation feature vector into a classification layer of the pre-configured artificial intelligence recommendation model to classify recommended content items, so as to obtain a classification confidence value of the portrait recommendation feature vector in each recommended content item;

and determining the recommended content item with the classification confidence value larger than the set confidence value as the recommended content item matched with the interactive portrait feature information corresponding to the graphic interactive behavior information.

In a second aspect, an embodiment of the present disclosure further provides an artificial intelligence and internet-based data analysis apparatus, which is applied to a cloud computing service platform, where the cloud computing service platform is in communication connection with a plurality of internet access devices, and the apparatus includes:

the acquisition module is used for acquiring a graph classification grading bookmark corresponding to a target graph search object initiated by the Internet access equipment, wherein the graph classification grading bookmark is used for representing graph classification grading information of the target graph search object under a big data analysis bookmark;

the big data analysis module is used for carrying out big data analysis on each graphic unit library in the target graphic search object according to the graphic classification grading bookmark to obtain graphic interaction behavior information of each graphic unit library in the target graphic search object and interaction portrait feature information corresponding to the graphic interaction behavior information;

the classification identification module is used for classifying and identifying the interaction portrait characteristic information corresponding to the graph interaction behavior information based on a pre-configured artificial intelligence recommendation model to obtain a recommended content item matched with the interaction portrait characteristic information corresponding to the graph interaction behavior information;

and the pushing module is used for pushing the hot spot recommendation information associated with the recommended content item to a graphical interactive interface of a corresponding graphical unit library.

In a third aspect, an embodiment of the present disclosure further provides an artificial intelligence and internet-based data analysis system, where the artificial intelligence and internet-based data analysis system includes a cloud computing service platform and a plurality of internet access devices communicatively connected to the cloud computing service platform;

the cloud computing service platform is used for acquiring a graph classification grading bookmark corresponding to a target graph search object initiated by the Internet access equipment, wherein the graph classification grading bookmark is used for representing graph classification grading information of the target graph search object under a big data analysis bookmark;

the cloud computing service platform is used for carrying out big data analysis on each graphic unit library in the target graphic search object according to the graphic classification grading bookmark to obtain graphic interaction behavior information of each graphic unit library in the target graphic search object and interaction portrait feature information corresponding to the graphic interaction behavior information;

the cloud computing service platform is used for classifying and identifying the interaction portrait characteristic information corresponding to the graph interaction behavior information based on a pre-configured artificial intelligence recommendation model to obtain a recommended content item matched with the interaction portrait characteristic information corresponding to the graph interaction behavior information;

the cloud computing service platform is used for pushing the hot spot recommendation information associated with the recommended content item to a graphical interaction interface of a corresponding graphical unit library.

In a fourth aspect, the disclosed embodiment further provides a cloud computing service platform, where the cloud computing service platform 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 configured to be communicatively connected to at least one internet access device, the machine-readable storage medium is configured to store a program, an instruction, or a code, and the processor is configured to execute the program, the instruction, or the code in the machine-readable storage medium to perform the artificial intelligence and internet-based data analysis method in the first aspect or any one of the possible designs of the first aspect.

In a fifth aspect, an embodiment of the present disclosure provides a computer-readable storage medium, in which instructions are stored, and when executed, cause a computer to perform the method for data analysis based on artificial intelligence and the internet in the first aspect or any one of the possible designs of the first aspect.

Based on any one of the above aspects, the method includes the steps of obtaining a graphic classification grading bookmark corresponding to a target graphic search object initiated by an internet access device, using the graphic classification grading bookmark to represent graphic classification grading information of the target graphic search object under a big data analysis bookmark, and then performing big data analysis on each graphic unit library in the target graphic search object according to the graphic classification grading bookmark, so that information recommendation is performed after portrait recognition. And (4) punching time.

Drawings

To more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present disclosure and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings may be obtained from the drawings without inventive effort.

Fig. 1 is a schematic view of an application scenario of an artificial intelligence and internet-based data analysis system according to an embodiment of the present disclosure;

FIG. 2 is a schematic flow chart diagram of a data analysis method based on artificial intelligence and Internet according to an embodiment of the present disclosure;

FIG. 3 is a functional block diagram of an artificial intelligence and Internet-based data analysis device according to an embodiment of the disclosure;

fig. 4 is a block diagram illustrating a structure of a cloud computing service platform for implementing the artificial intelligence and internet-based data analysis method according to the embodiment of the present disclosure.

Detailed Description

The present disclosure is described in detail below with reference to the drawings, and the specific operation methods in the method embodiments can also be applied to the device embodiments or the system embodiments.

FIG. 1 is an interactive schematic diagram of an artificial intelligence and Internet-based data analysis system 10 provided by an embodiment of the present disclosure. The artificial intelligence and internet based data analysis system 10 may include a cloud computing service platform 100 and an internet access device 200 communicatively connected to the cloud computing service platform 100. The artificial intelligence and internet based data analysis system 10 shown in fig. 1 is only one possible example, and in other possible embodiments, the artificial intelligence and internet based data analysis system 10 may also include only some of the components shown in fig. 1 or may also include other components.

In this embodiment, the internet access device 200 may comprise a mobile device, a tablet computer, a laptop computer, etc., or any combination thereof. In some embodiments, the mobile device may include a smart home device, a wearable device, a smart mobile device, a virtual reality device, an augmented reality device, or the like, or any combination thereof. In some embodiments, the smart home devices may include control devices of smart electrical devices, smart monitoring devices, smart televisions, smart cameras, and the like, or any combination thereof. In some embodiments, the wearable device may include a smart bracelet, a smart lace, smart glass, a smart helmet, a smart watch, a smart garment, a smart backpack, a smart accessory, or the like, or any combination thereof. In some embodiments, the smart mobile device may include a smartphone, a personal digital assistant, a gaming device, and the like, or any combination thereof. In some embodiments, the virtual reality device and/or the augmented reality device may include a virtual reality helmet, virtual reality glass, a virtual reality patch, an augmented reality helmet, augmented reality glass, an augmented reality patch, or the like, or any combination thereof. For example, the virtual reality device and/or augmented reality device may include various virtual reality products and the like.

In this embodiment, the cloud computing service platform 100 and the internet access device 200 in the data analysis system 10 based on artificial intelligence and the internet can cooperatively perform the data analysis method based on artificial intelligence and the internet described in the following method embodiments, and the following detailed description of the method embodiments can be referred to for the specific steps performed by the cloud computing service platform 100 and the internet access device 200.

In this embodiment, the data analysis system 10 based on artificial intelligence and the internet can be implemented in various application scenarios, such as a blockchain application scenario, an intelligent home application scenario, and an intelligent control application scenario.

In order to solve the technical problem in the foregoing background, fig. 2 is a schematic flowchart of a data analysis method based on artificial intelligence and the internet according to an embodiment of the present disclosure, where the data analysis method based on artificial intelligence and the internet according to the embodiment of the present disclosure may be executed by the cloud computing service platform 100 shown in fig. 1, and it should be further understood that, in various embodiments of the present disclosure, the sequence number of each process does not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiment of the present disclosure. The data analysis method based on artificial intelligence and the internet will be described in detail below.

Step S110, a graph classification grading bookmark corresponding to the target graph search object initiated by the internet access device 200 is obtained.

And step S120, carrying out big data analysis on each graphic unit library in the target graphic search object according to the graphic classification grading bookmark to obtain the graphic interaction behavior information of each graphic unit library in the target graphic search object and the interaction portrait feature information corresponding to the graphic interaction behavior information.

And step S130, classifying and identifying the interaction portrait characteristic information corresponding to the graph interaction behavior information based on a pre-configured artificial intelligence recommendation model to obtain a recommended content item matched with the interaction portrait characteristic information corresponding to the graph interaction behavior information.

Step S140, hot spot recommendation information associated with the recommended content item is pushed to a graphical interactive interface of a corresponding graphical unit library.

In this embodiment, the graph classification grading bookmark may be used to represent graph classification grading information of the target graph search object under the big data analysis bookmark. For example, a big data analytics bookmark may refer to a preconfigured marker business location at the time of big data analytics.

In this embodiment, the target graphic search object may be a scan-and-photograph type graphic search object (for example, a user initiates a scan of some objects through a certain browser), an image upload type graphic search object (for example, a user uploads some objects through a certain recognition tool), and the like, which are not limited in detail herein.

In this embodiment, the preconfigured artificial intelligence recommendation model may be obtained by training based on a training sample (e.g., interactive image feature information) and a corresponding training label (e.g., recommended content item), and a specific training mode may adopt an existing general training scheme, which is not a key point of the embodiment of the present disclosure, and is not described herein again.

In this embodiment, the hot spot recommendation information associated with the recommended content item is pushed to the graphical interactive interface of the corresponding graphical unit library, and the detailed manner may be as follows: when a user accesses a graphical interaction interface of a graphical unit library, in order to provide better information service experience for the user, hotspot recommendation information associated with recommended content items can be displayed in a related recommendation page in the graphical interaction interface, so that the user can know the content of interest of the user, and the user experience is improved.

Based on the above steps, in this embodiment, the internet access device 200 is used to initiate the graph classification hierarchical bookmark corresponding to the target graph search object, so that the graph classification hierarchical bookmark is used to represent the graph classification hierarchical information of the target graph search object under the big data analysis bookmark, and then the big data analysis is performed on each graph unit library in the target graph search object according to the graph classification hierarchical bookmark, so as to perform information recommendation after image recognition.

In a possible implementation manner, as for step S110, the inventor of the present application has found that, in the conventional scheme, the difference between different big data collection classifications is not generally considered, so that a situation of multi-classification conflict is easily caused in the classification process, and in the classification process, a user may have a need to perform more accurate and faster big data processing on the image unit library based on some key graph classification hierarchical bookmarks based on the previous historical classification situation in the big data analysis, but the conventional scheme cannot meet the need, and further, for the user, a longer buffering time may be caused in the actual big data classification process.

To solve this problem, step S110 may be implemented by the following exemplary sub-steps, which are described in detail below.

And a substep S111 of obtaining the image classification label of the target graph search object under the artificial intelligence recognition result of each artificial intelligence recognition model from the internet access device 200, classifying the image classification labels under the artificial intelligence recognition results according to the preset big data collection classification, and respectively generating an image classification label sequence of each big data collection classification.

And a substep S112, determining a target graphic classification grade bookmark associated with each artificial intelligence identification result according to the user search behavior information of the target graphic search object, and respectively determining the grade label information of a first indexable grade bookmark of the target graphic classification grade bookmark in the corresponding image classification label sequence of big data collection classification aiming at the target graphic classification grade bookmark associated with each artificial intelligence identification result, so as to obtain a first information management index sequence of the target graphic classification grade bookmark.

And a substep S113, determining the key graph classification and classification bookmark associated with each artificial intelligence identification result according to the historical classification and classification information of the target graph search object, respectively obtaining the second indexable classification bookmark of the key graph classification and classification bookmark aiming at the key graph classification and classification bookmark associated with each artificial intelligence identification result, and determining the classification label information of the second indexable classification bookmark in the corresponding big data collection and classification image classification label sequence to obtain a second information management index sequence of the key graph classification and classification bookmark.

And a substep S114, determining the graph classification grading bookmark corresponding to the target graph search object according to the matching relation between the first information management index sequence and the second information management index sequence.

In this embodiment, the target graphic classification hierarchical bookmark is a graphic classification hierarchical bookmark that can be matched with user search behavior information of the target graphic search object in advance, and in detail, for different target graphic search objects (for example, a scan photographing type graphic search object, an image uploading type graphic search object, and the like), different corresponding graphic classification hierarchical bookmarks can be preset according to different service use requirements of the respective objects. For example, a graphical category rating bookmark may include a category label and a rating label to which the category label is located.

In this embodiment, the key graphic classification hierarchical bookmark may be a graphic classification hierarchical bookmark whose classification frequency in the history hierarchical classification information of the target graphic search object is greater than a set frequency threshold, and the classification frequency may be used to represent the number of search classifications of the graphic classification hierarchical bookmark in a unit time. The service usage requirement may be determined according to actual requirements, and may include information retrieval, information analysis, information filling, and the like, for example, and is not limited in detail herein.

In this embodiment, the artificial intelligence recognition model may be obtained by training samples of various pattern search objects in advance, and the artificial intelligence recognition model that recognizes different types of pattern objects may be trained, so that the image classification label under the artificial intelligence recognition result of the target pattern search object may be recognized.

Based on the above sub-steps, the present embodiment classifies the image classification tags under each artificial intelligence recognition result based on the predetermined big data collection classification, thereby taking into account the difference of different big data collection classifications, improving the situation of multi-classification conflict during the classification, and moreover, by combining the user search behavior information and the historical classification information of the target graph search object, after comparing the information management index sequences of the two graph classification bookmarks, respectively performing big data analysis on each graph cell library in the target graph search object based on each corresponding graph classification bookmark of the artificial intelligence recognition result, thereby facilitating the historical classification situation based on the previous big data analysis, and further performing more accurate and rapid big data processing on the graph cell library based on some key graph classification bookmark, and improving the big data analysis efficiency, reducing the buffering time.

In one possible implementation, for step S111, in order to improve the accuracy of the division and reduce redundant information to improve the classification accuracy, the following exemplary sub-steps may be further implemented, which are described in detail below.

And a substep S1111, acquiring a classification target corresponding to each preset big data collection classification, forming a classification target sequence of each preset big data collection classification, and acquiring associated classification target information of each target classification target of each artificial intelligence recognition result and the classification target of the classification target sequence.

And a substep S1112, calculating the density of the key classification targets of each target big data collection classification according to the related classification target information of the target classification targets and the classification targets of the classification target sequence, and selecting the classification targets from the classification target sequence according to the density of the key classification targets of each target big data collection classification to obtain the initial classification target arrangement distribution.

In the sub-step S1113, if the total classification target distribution density of the initial classification target arrangement distribution is greater than the maximum total classification target distribution density required by the total classification target distribution density, the first key classification target in the initial classification target arrangement distribution is dispersed to the first distribution density, and the second key classification target in the initial classification target arrangement distribution is aggregated to the first distribution density.

For example, in one possible example, the second key classification target may refer to a key classification target in which the label density (e.g., the number of labels in a unit area) of the label hierarchy in which the key classification target is located is less than a set level, and the first key classification target refers to a key classification target in which the label density of the label hierarchy in which the key classification target is located is not less than the set level.

And a substep S1114, calculating a total classification target distribution density of the updated initial classification target arrangement distribution.

And a substep S1115, if the total classification target distribution density of the initial classification target arrangement distribution after the updating is greater than the maximum total classification target distribution density, performing the above processing on the initial classification target arrangement distribution after the updating again.

In sub-step S1116, if the total classification target distribution density of the initial classification target arrangement distribution after the update is less than or equal to the maximum total classification target distribution density, the initial classification target arrangement distribution before the update is used as a first update arrangement distribution, and the target big data collection classifications are sorted according to the sequence from the low priority to the high priority of the big data collection classification, so as to obtain a target big data collection classification sequence.

And a substep S1117, classifying the image classification labels under each artificial intelligence recognition result according to the target big data collection classification sequence, and respectively generating an image classification label sequence of each big data collection classification.

In one possible implementation manner, for step S112, in order to accurately obtain the first information management index sequence of the target graphic classification hierarchical bookmark, the following exemplary implementation manner can be implemented, which is described in detail below.

And a substep S1121, for the target graphic classification grade bookmarks associated with each artificial intelligence recognition result, respectively obtaining index running scripts matched with the target graphic classification grade bookmarks, and obtaining a label classification object corresponding to the index running scripts when continuously indexing and searching for an object entity corresponding to one label classification object in the artificial intelligence recognition result in a preset time period as a target label classification object.

In the sub-step S1122, it is determined whether the classification index search feature of the target label classification object matches the classification index search feature of the index node of the information management index unit, and if the classification index search feature does not match, the classification index search feature of the target label classification object is adjusted to the label classification object matching the classification index search feature of the index node of the information management index unit and input to the information management index unit.

And a substep S1123 of calculating the input label classification object by using the information management index unit, acquiring hierarchical label information corresponding to the input label classification object, expanding each label marking information of the target graph classification hierarchical bookmark in the target label classification object, and acquiring label expansion information of each label marking information in the target label classification object.

In the substep S1124, the hierarchical label with the frequency of the label labeling information being greater than the preset frequency in the hierarchical label information corresponding to the input label classification object is determined as the first indexable classification bookmark, and the label feature vector of each label labeling information in the input label classification object is converted to obtain the label extension information of each label labeling information in the input label classification object.

In the sub-step S1125, a first tag expansion information sequence of the whole tag classification object is determined according to the tag expansion information of each tag label information in the target tag classification object, and a second tag expansion information sequence of the first indexable classification bookmark is determined according to the tag expansion information of each tag label information in the first indexable classification bookmark.

And a substep S1126 of determining a tag extension information sequence of the first indexable classification bookmark according to the first tag extension information sequence, the second tag extension information sequence and a preset proportion, and determining the hierarchical tag information of the first indexable classification bookmark of the target graph classification hierarchical bookmark in the corresponding image classification tag sequence of the big data collection classification according to the tag extension information and the tag extension information sequence of each tag marking information in the target tag classification object to obtain a first information management index sequence of the target graph classification hierarchical bookmark.

In one possible implementation, the substep S114 can be implemented in the following exemplary embodiments, which are described in detail below.

And a substep S1141 of matching the information management index sequence of each target graphic classification hierarchical bookmark in the first information management index sequence with the information management index sequence of each matched key graphic classification hierarchical bookmark in the second information management index sequence to obtain a plurality of matching degrees.

For example, each matched key graph classification hierarchical bookmark in the second information management index sequence is matched with the arrangement sequence of the corresponding target graph classification hierarchical bookmark in the respective information management index sequence, and the matching degree is determined according to the contact degree between the information management index sequence of the target graph classification hierarchical bookmark and the information management index sequence of the matched key graph classification hierarchical bookmark.

And a substep S1142, when the matching degree between the information management index sequence of any one target graphic classification grading bookmark and the information management index sequence of the matched key graphic classification grading bookmark is greater than the set matching degree, taking the target graphic classification grading bookmark and the key graphic classification grading bookmark as a big data analysis combined object.

And a substep S1143, when the matching degree between the information management index sequence of any one target graphic classification grading bookmark and the information management index sequence of the matched key graphic classification grading bookmark is not more than the set matching degree, independently using the target graphic classification grading bookmark and the key graphic classification grading bookmark as a big data analysis independent object.

And a substep S1144 of determining the graphic classification grading bookmark corresponding to the big data analysis combined object and the big data analysis independent object as the graphic classification grading bookmark corresponding to the target graphic search object.

Therefore, by combining the user searching behavior information and the historical hierarchical classification information of the target graphic searching object, after the information management index sequences of the two graphic classification hierarchical bookmarks are compared, each corresponding graphic classification hierarchical bookmark based on the artificial intelligence recognition result respectively carries out big data analysis on each graphic unit library in the target graphic searching object, the historical classification condition during the analysis based on the previous big data can be conveniently carried out, the more accurate and rapid big data processing can be carried out on the graphic unit libraries based on some key graphic classification hierarchical bookmarks, the big data analysis efficiency is improved, and the buffering time is shortened.

It should be particularly noted that after determining the key graph classification hierarchical bookmark associated with each artificial intelligence recognition result, the present embodiment may further obtain the second information management index sequence of the key graph classification hierarchical bookmark according to a similar operation manner of obtaining the first information management index sequence of the target graph classification hierarchical bookmark in the foregoing embodiment, which has been described in detail before, and thus is not described herein again.

In one possible implementation, for the sub-step S120, in order to further reduce the calculation amount of the analysis process, the following exemplary embodiments may be implemented, and the following detailed description is given.

And a substep S121, obtaining at least one candidate unit data area to be analyzed and an interaction behavior recording node corresponding to each candidate unit data area from each graphic unit library in the target graphic search object according to a preset big data analysis strategy of the graphic classification grading bookmark and the interaction history of each graphic unit library in the target graphic search object.

In this embodiment, for example, a target record object of an interaction history record of each graphic unit library in the target graphic search object may be obtained according to a preset big data analysis policy of the graphic classification hierarchical bookmark, then a data area matched with the target record object is obtained from each graphic unit library in the target graphic search object to serve as at least one candidate unit data area to be analyzed, and then an interaction behavior record node corresponding to each candidate unit data area is continuously obtained.

In this embodiment, the interaction behavior recording node may be understood as an interaction process recording unit in each interaction behavior.

And a substep S122, in the past interactive behavior node sequence corresponding to the preset big data analysis strategy, grouping the initial past interactive behavior nodes to obtain at least one initial past interactive behavior node sequence, regarding any initial past interactive behavior node sequence in at least one initial past interactive behavior node sequence, taking a past interactive behavior node corresponding to a grouping cluster center of any initial past interactive behavior node sequence as any cluster center node, for any past interactive behavior node except the initial past interactive behavior node, adding any past interactive behavior node into the initial past interactive behavior node sequence corresponding to the cluster core node with the maximum similarity of any past interactive behavior node, and when the past interactive behavior nodes which are not added to the initial past interactive behavior node sequence do not exist, obtaining the past interactive behavior node sequence corresponding to each cluster core node.

In this embodiment, the interaction behavior node may be understood as a specific interaction behavior in an interaction process record unit at each interaction behavior.

And a substep S123 of determining at least one target cluster core node corresponding to the interactive behavior recording node based on the similarity between the interactive behavior recording node and each cluster core node for the interactive behavior recording node corresponding to any candidate unit data region, and taking the union of past interactive behavior node sequences corresponding to each target cluster core node as a first candidate past interactive behavior node sequence corresponding to any candidate unit data region.

In this embodiment, the similarity between the interactive behavior recording node and each cluster core node may be determined based on the coincidence degree of the services between the interactive behavior recording node and each cluster core node, which is not limited herein in detail.

And a substep S124, obtaining the graph interaction behavior information of each candidate unit data region in the at least one candidate unit data region based on the first candidate past interaction behavior node sequence corresponding to each candidate unit data region in the at least one candidate unit data region, and using the graph interaction behavior information as the graph interaction behavior information of each graph unit library in the target graph search object.

In this embodiment, the graphic interaction behavior information may refer to graphic interaction change information under the interaction behavior.

Illustratively, when the number of the at least one candidate unit data region is smaller than the set number, for any candidate unit data region, analyzing any candidate unit data region in a first candidate past interaction behavior node sequence corresponding to any candidate unit data region to obtain the graph interaction behavior information of any candidate unit data region.

For another example, when the number of the at least one candidate unit data regions is not less than the set number, the entire target region sequence including the first target number of candidate unit data regions may be determined based on the at least one candidate unit data region. Then, the number of past interactive behavior nodes corresponding to any one of all the target region sequences is obtained, and based on the number of past interactive behavior nodes corresponding to each of all the target region sequences, at least one candidate unit data region is divided into a candidate unit data region set with a second target number, wherein the second target number is the ratio of the number of at least one candidate unit data region to the first target number.

On this basis, for any candidate unit data region set, a union of first candidate past interactive behavior node sequences corresponding to each candidate unit data region in any candidate unit data region set may be used as a second candidate past interactive behavior node sequence corresponding to any candidate unit data region set. And analyzing each candidate unit data area in any candidate unit data area set in a second candidate past interactive behavior node sequence corresponding to any candidate unit data area set to obtain the graph interactive behavior information of each candidate unit data area in any candidate unit data area set.

And a substep S125, obtaining interactive service information in the interactive process of the graphic interaction behavior information of each graphic unit library in the target graphic search object, and converting the interactive service information into corresponding structured data information.

And a substep S126, determining a structural feature vector of at least one structural item corresponding to the structural data information through the coding node of the graph interaction behavior information.

And a substep S127 of determining the interactive processing information of the corresponding interactive node according to the structural feature vector of the at least one structural project through the decoding node of the graph interactive behavior information, and converting the interactive processing information of the corresponding interactive node into an interactive feature vector.

And a substep S128 of generating a portrait label corresponding to the structural feature vector of the structural item and a confidence of the portrait label according to the interactive feature vector and the corresponding fusion feature vector through a decoding node of the graph interactive behavior information.

And a substep S129 of selecting at least one portrait label to form interactive processing information corresponding to the structured data information according to the confidence of the interactive processing information.

And a substep S1295, converting the interactive processing information into new interactive service information corresponding to the graphic interactive behavior information to obtain the interactive portrait feature information corresponding to the graphic interactive behavior information.

In a possible example, in the process of dividing at least one candidate unit data region into candidate unit data region sets of the second target number based on the number of past interaction behavior nodes corresponding to each target region sequence in all target region sequences, the present embodiment may determine the number of past interaction behavior nodes corresponding to each first interaction behavior node sequence based on the number of past interaction behavior nodes corresponding to each target region sequence in all target region sequences, where any first interaction behavior node sequence includes candidate unit data regions in two target region sequences that satisfy the first condition.

And then, determining the number of past interactive behavior nodes corresponding to each second interactive behavior node sequence based on the number of past interactive behavior nodes corresponding to each first interactive behavior node sequence and the number of past interactive behavior nodes corresponding to each target region sequence, wherein any second interactive behavior node sequence comprises candidate unit data regions in three target region sequences meeting a second condition.

And repeating the steps until the quantity of the past interactive behavior nodes corresponding to the final interactive behavior node sequence comprising all the candidate unit data areas is determined.

Therefore, reverse derivation can be performed based on the number of past interactive behavior nodes corresponding to the final interactive behavior node sequence, a second target number of target region sequences corresponding to the number of past interactive behavior nodes corresponding to the final interactive behavior node sequence are determined, and then at least one candidate unit data region is divided into a second target number of candidate unit data region sets according to the second target number of target region sequences.

In another possible example, in the process of dividing at least one candidate unit data region into candidate unit data region sets of the second target number based on the number of past interaction behavior nodes corresponding to each target region sequence in all target region sequences, the present embodiment may determine the number of past interaction behavior nodes corresponding to each first interaction behavior node sequence based on the number of past interaction behavior nodes corresponding to each target region sequence in all target region sequences, where any first interaction behavior node sequence includes candidate unit data regions in two target region sequences that satisfy the first condition.

And repeating the steps until the quantity of the past interactive behavior nodes corresponding to each intermediate interactive behavior node sequence is determined, wherein any intermediate interactive behavior node sequence comprises half the quantity of the candidate unit data areas.

Therefore, the number of past interactive behavior nodes corresponding to the final interactive behavior node sequence comprising all the candidate unit data regions can be determined based on the number of past interactive behavior nodes corresponding to each intermediate interactive behavior node sequence. And then, carrying out reverse derivation based on the number of past interactive behavior nodes corresponding to the final interactive behavior node sequence, and determining a second target number of target region sequences corresponding to the number of past interactive behavior nodes corresponding to the final interactive behavior node sequence. Thus, the at least one candidate unit data region may be divided into a second target number of candidate unit data region sets according to a second target number of target region sequences.

In one possible implementation, the substep S130 can be implemented in the following exemplary embodiments, which are described in detail below.

And a substep S131, inputting the interactive portrait feature information corresponding to the graphic interactive behavior information into a preset artificial intelligence recommendation model, and extracting the portrait recommendation feature vector corresponding to the interactive portrait feature information corresponding to the graphic interactive behavior information.

And a substep S132, inputting the portrait recommendation feature vector into a classification layer of a pre-configured artificial intelligence recommendation model to classify recommended content items, and obtaining a classification confidence value of the portrait recommendation feature vector in each recommended content item.

In the substep S133, the recommended content item whose classification confidence value is greater than the set confidence value is determined as a recommended content item whose interaction portrait feature information corresponding to the graphical interaction behavior information matches.

It is understood that the set confidence value may be configured according to actual requirements, for example, may be configured as 80, that is, for the recommended content item a, the recommended content item B, the recommended content item C, the recommended content item D, and the recommended content item E, assuming that the confidence values corresponding to the recommended content item a, the recommended content item B, the recommended content item C, the recommended content item D, and the recommended content item E are 60, 83, 77, 92, and 85, the matched recommended content item may be configured as the recommended content item B, the recommended content item D, and the recommended content item E.

It is to be noted that, in some possible embodiments, the recommended content item may be a multi-level menu, that is, further classification confidence determination may be performed on the recommended content items B, D, and the recommended content sub-items under the recommended content item E (for example, the recommended content sub-item B1, the recommended content sub-item B2, the recommended content sub-item D1, the recommended content sub-item E1, and the like), so as to determine a more accurate recommended content sub-item for subsequent information recommendation, and further improve user experience.

Fig. 3 is a schematic diagram of functional modules of a data analysis apparatus 300 based on artificial intelligence and the internet according to an embodiment of the present disclosure, in this embodiment, the data analysis apparatus 300 based on artificial intelligence and the internet may be divided into the functional modules according to a method embodiment executed by the cloud computing service platform 100, that is, the following functional modules corresponding to the data analysis apparatus 300 based on artificial intelligence and the internet may be used to execute each method embodiment executed by the cloud computing service platform 100. The artificial intelligence and internet based data analysis apparatus 300 may include an obtaining module 310, a big data analysis module 320, a classification and identification module 330, and a pushing module 340, wherein the functions of the functional modules of the artificial intelligence and internet based data analysis apparatus 300 are described in detail below.

An obtaining module 310, configured to obtain a graph classification grading bookmark corresponding to a target graph search object initiated by the internet access device, where the graph classification grading bookmark is used to represent graph classification grading information of the target graph search object under a big data analysis bookmark. The obtaining module 310 may be configured to perform the step S110, and the detailed implementation of the obtaining module 310 may refer to the detailed description of the step S110.

And a big data analysis module 320, configured to perform big data analysis on each graphic unit library in the target graphic search object according to the graphic classification hierarchical bookmark, so as to obtain graphic interaction behavior information of each graphic unit library in the target graphic search object and interaction portrait feature information corresponding to the graphic interaction behavior information. The big data analysis module 320 may be configured to perform the step S120, and the detailed implementation of the big data analysis module 320 may refer to the detailed description of the step S120.

And the classification identification module 330 is configured to perform classification identification on the interaction portrait feature information corresponding to the graphical interaction behavior information based on a preconfigured artificial intelligence recommendation model, so as to obtain a recommended content item matched with the interaction portrait feature information corresponding to the graphical interaction behavior information. The classification identifying module 330 may be configured to perform the step S130, and the detailed implementation of the classification identifying module 330 may refer to the detailed description of the step S130.

The pushing module 340 is configured to push the hot spot recommendation information associated with the recommended content item to a graphical interaction interface of a corresponding graphical unit library. 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 obtaining module 310 may be a processing element separately set up, or may be implemented by being integrated into a chip of the apparatus, or may be stored in a memory of the apparatus in the form of program code, and the processing element of the apparatus calls and executes the functions of the 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.

For example, the above modules may be one or more integrated circuits configured to implement the above methods, such as: one or more Application Specific Integrated Circuits (ASICs), or one or more microprocessors (DSPs), or one or more Field Programmable Gate Arrays (FPGAs), among others. For another example, when some of the above modules are implemented in the form of a processing element scheduler code, the processing element may be a general-purpose processor, such as a Central Processing Unit (CPU) or other processor that can call program code. As another example, these modules may be integrated together, implemented in the form of a system-on-a-chip (SOC).

Fig. 4 illustrates a hardware structure diagram of the cloud computing service platform 100 for implementing the artificial intelligence and internet-based data analysis method, where as shown in fig. 4, the cloud computing service platform 100 may include a processor 110, a machine-readable storage medium 120, a bus 130, and a transceiver 140.

In a specific implementation process, the at least one processor 110 executes computer-executable instructions stored in the machine-readable storage medium 120 (for example, the obtaining module 310, the big data analysis module 320, the classification identifying module 330, and the pushing module 340 included in the artificial intelligence and internet-based data analysis apparatus 300 shown in fig. 3), so that the processor 110 may execute the artificial intelligence and internet-based data analysis method according to the above method embodiment, where the processor 110, the machine-readable storage medium 120, and the transceiver 140 are connected via the bus 130, and the processor 110 may be configured to control the transceiving action of the transceiver 140, so as to perform data transceiving with the aforementioned internet access device 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 service platform 100, and implementation principles and technical effects thereof are similar, and details of this embodiment 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 disclosure are not limited to only one bus or one type of bus.

In addition, the embodiment of the disclosure also provides a readable storage medium, in which computer execution instructions are stored, and when a processor executes the computer execution instructions, the data analysis method based on artificial intelligence and the internet is implemented.

The readable storage medium described above may be either volatile memory or nonvolatile memory, or may include both volatile and nonvolatile memory. The non-volatile memory may be a Read-only memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically Erasable PROM (EEPROM), or a flash memory. Volatile Memory can be Random Access Memory (RAM), which acts as external cache Memory. By way of example, but not limitation, many forms of RAM are available, such as Static random access memory (Static RAM, SRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic random access memory (Synchronous DRAM, SDRAM), Double Data rate Synchronous Dynamic random access memory (DDR SDRAM), Enhanced Synchronous SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), and direct memory bus RAM (DR RAM). It should be noted that the memory of the systems and methods described herein is intended to comprise, without being limited to, these and any other suitable types of memory.

It is also to be understood that the terminology used in the embodiments of the disclosure and the appended claims is for the purpose of describing particular embodiments only, and is not intended to be limiting of the embodiments of the disclosure. For example, as used in the disclosed embodiments and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.

Finally, it should be noted that: the above embodiments are only used for illustrating the technical solutions of the present disclosure, and not for limiting the same; while the present disclosure has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art will understand that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present disclosure.

Claims

1. A data analysis method based on artificial intelligence and the Internet is applied to a cloud computing service platform, the cloud computing service platform is in communication connection with a plurality of Internet access devices, and the method comprises the following steps:

pushing hotspot recommendation information associated with the recommended content item to a graphical interaction interface of a corresponding graphical unit library;

the target graphic search object comprises a scanning photographing type graphic search object and an image uploading type graphic search object;

the pre-configured artificial intelligence recommendation model is obtained by training based on a training sample and a corresponding training label, wherein the training sample comprises interactive portrait feature information, and the training label comprises a recommended content item;

pushing hotspot recommendation information associated with the recommended content item to a graphical interaction interface of a corresponding graphical unit library, specifically comprising: and when the user accesses the graphical interactive interface of the graphical unit library, displaying the hot spot recommendation information associated with the recommended content item to a related recommendation page in the graphical interactive interface.

2. The method for analyzing data based on artificial intelligence and internet as claimed in claim 1, wherein the step of obtaining the graphic classification hierarchical bookmark corresponding to the target graphic search object initiated by the internet access device comprises:

3. The method for analyzing data based on artificial intelligence and internet as claimed in claim 1, wherein the step of determining the graphic classification hierarchical bookmark corresponding to the target graphic search object according to the matching relationship between the first information management index sequence and the second information management index sequence comprises:

4. The method of claim 1, wherein the step of analyzing big data of each graphic unit library in the target graphic search object according to the graphic classification hierarchical bookmark to obtain the graphic interaction behavior information of each graphic unit library in the target graphic search object and the interaction image feature information corresponding to the graphic interaction behavior information comprises:

5. The method of claim 4, wherein the step of obtaining the graph interaction behavior information of each candidate unit data area in the at least one candidate unit data area based on the first candidate past interaction behavior node sequence corresponding to each candidate unit data area in the at least one candidate unit data area comprises:

6. The artificial intelligence and internet-based data analysis method according to claim 4 or 5, wherein the step of obtaining the graph interaction behavior information of each candidate unit data area in the at least one candidate unit data area based on the first candidate past interaction behavior node sequence corresponding to each candidate unit data area in the at least one candidate unit data area comprises:

7. The artificial intelligence and internet-based data analysis method according to claim 6, wherein the step of dividing the at least one candidate unit data area into a second target number of candidate unit data area sets based on the number of past interactive behavior nodes corresponding to each target area sequence of the entire target area sequences comprises:

8. The artificial intelligence and internet-based data analysis method according to claim 6, wherein the step of dividing the at least one candidate unit data area into a second target number of candidate unit data area sets based on the number of past interactive behavior nodes corresponding to each target area sequence of the entire target area sequences comprises:

9. The artificial intelligence and internet-based data analysis method according to any one of claims 1 to 8, wherein the step of classifying and identifying the interaction image feature information corresponding to the graphical interaction behavior information based on a preconfigured artificial intelligence recommendation model to obtain the recommended content item matched with the interaction image feature information corresponding to the graphical interaction behavior information comprises:

10. An artificial intelligence and internet based data analysis system, comprising a cloud computing service platform and a plurality of internet access devices in communication connection with the cloud computing service platform;

the cloud computing service platform is used for pushing hotspot recommendation information associated with the recommended content items to a graphical interaction interface of a corresponding graphical unit library;