CN113946751A - Data acquisition and interpretation method, device, equipment and medium based on artificial intelligence - Google Patents

Data acquisition and interpretation method, device, equipment and medium based on artificial intelligence Download PDF

Info

Publication number
CN113946751A
CN113946751A CN202111231734.7A CN202111231734A CN113946751A CN 113946751 A CN113946751 A CN 113946751A CN 202111231734 A CN202111231734 A CN 202111231734A CN 113946751 A CN113946751 A CN 113946751A
Authority
CN
China
Prior art keywords
behavior
server
interest
data
module
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111231734.7A
Other languages
Chinese (zh)
Inventor
余玉霞
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An International Smart City Technology Co Ltd
Original Assignee
Ping An International Smart City Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ping An International Smart City Technology Co Ltd filed Critical Ping An International Smart City Technology Co Ltd
Priority to CN202111231734.7A priority Critical patent/CN113946751A/en
Publication of CN113946751A publication Critical patent/CN113946751A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9535Search customisation based on user profiles and personalisation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/906Clustering; Classification

Abstract

The application relates to the field of artificial intelligence and discloses a data acquisition and interpretation method based on artificial intelligence.A network data acquisition unit acquires operation records of a user side on a service platform to generate interest information and sends the interest information to a first server; the first server divides the interest information according to the behavior attribute and the generation time to generate a multi-level behavior table and sends the multi-level behavior table to a second server; the second server reads the data meaning of the multi-level behavior table, obtains a characteristic calculation table reflecting the attention degree of the user end to the network product on the service platform, and sends the characteristic calculation table to a third server; and the third server performs clustering operation on the data in the characteristic calculation table to obtain a characteristic label which is most matched with the characteristic calculation table, and sends the interest information, the characteristic calculation table and the characteristic label to a data server for storage. The data information describing the times of the interest information is realized, the behavior interest of the user is accurately identified, and the actual intention of the user is further accurately identified.

Description

Data acquisition and interpretation method, device, equipment and medium based on artificial intelligence
Technical Field
The application belongs to the technical field of artificial intelligence, and particularly relates to a data acquisition reading method, device, equipment and storage medium based on artificial intelligence.
Background
In practical application, the user interest model is more and more widely applied, the preference of a user is described, and accurate operation and personalized recommendation are performed. The accurate and high-reality behavior portrait is the important factor for establishing the user interest model. At present, a clustering algorithm, a classifier and other computer models are generally adopted to classify information to obtain a useful classification result for a user, original data are generally directly converted into feature vectors, and the feature vectors are analyzed through the models to classify the information; however, the classification method often has low judgment accuracy due to the fact that data in the information is too discrete.
Disclosure of Invention
In view of the above, the present invention provides a data acquisition and interpretation method, device, equipment and storage medium based on artificial intelligence, and aims to solve the technical problem of low data classification accuracy in the prior art.
In order to achieve the above object, the present invention provides a data acquisition and interpretation method based on artificial intelligence, which comprises:
the network data acquisition unit acquires operation records of a user side on a service platform to generate interest information and sends the interest information to the first server;
the first server divides the interest information according to the behavior attribute and the generation time to generate a multi-level behavior table and sends the multi-level behavior table to a second server;
the second server reads the data meaning of the multi-level behavior table, obtains a characteristic calculation table reflecting the attention degree of the user end to the network product on the service platform, and sends the characteristic calculation table to a third server;
and the third server performs clustering operation on the data in the characteristic calculation table to obtain a characteristic label which is most matched with the characteristic calculation table, and sends the interest information, the characteristic calculation table and the characteristic label to a data server for storage.
Preferably, the acquiring, by the network data collector, operation records of the user side on the service platform are collected to generate interest information, and the interest information is sent to the first server, including:
the method comprises the steps that an acquisition module of a network data acquisition unit acquires a unique identifier of a user side and acquires an operation record on a service platform according to the unique identifier;
and an information module of the network data acquisition unit collects the unique identifier and the operation record to form interest information.
Preferably, the dividing, by the first server, the interest information according to the behavior attribute and the generation time to generate a multi-level behavior table includes:
the decomposition module of the first server divides the interest information according to behavior attributes to form behavior decomposition information, and sends the behavior decomposition information to the long-term data module of the first server;
the long-term data module acquires the total amount of the behavior decomposition information and sets the total amount as an interest accumulated value, subtracts the current time from a preset long-period value to obtain a long-term time point, acquires the amount of the behavior decomposition information between the current time and the long-term time point, and sets the amount as a long-term interest value;
the short-term data module of the first server subtracts the current time from a preset first short-term value to obtain a first short-term time point, obtains the quantity of behavior decomposition information between the current time and the first short-term time point, and sets the quantity as a first short-term interest value; subtracting a preset second short period value from the current time to obtain a second short period time point, acquiring the quantity of behavior decomposition information between the current time and the second short period time point, and setting the quantity as a second short period interest value;
and the summarizing module of the first server takes the interest accumulated value, the long-term interest value, the first short-period interest value and the second short-period interest value as interest characteristic data, and summarizes the interest characteristic data of each behavior attribute to obtain a multi-level behavior table.
Preferably, the summarization module comprises a data summarization unit and a data cleaning unit.
Preferably, the second server interprets the data meaning of the multilevel behavior table to obtain a feature calculation table reflecting the degree of interest of the user end to the network product on the service platform, and the feature calculation table includes:
a first calculation module of the second server writes a mean value formula, and the first calculation module calculates a mean value of interest characteristic data of each behavior attribute in the multi-level behavior table through the mean value formula to obtain an index mean value of each behavior attribute and writes the index mean value into the multi-level behavior table to form a first behavior calculation table;
a second calculation module of the second server writes a variance formula, and the second calculation module calculates the index average index and the interest characteristic data of each behavior attribute in the first behavior calculation table through the variance formula to obtain the index variance of each behavior attribute and writes the index variance into the first behavior calculation table to form a second behavior calculation table;
a third calculation module of the second server adds the index variances of the behavior contents in a second behavior calculation table to obtain a behavior total variance;
and a fourth calculation module of the second server is provided with a proportion formula, calculates the proportion of the index variance of each behavior content in the second behavior calculation table in the behavior total variance through the proportion formula, sets the proportions as the index weights of each behavior content respectively, and writes the index weights and the total variance into the second behavior calculation table to form a feature calculation table.
Preferably, the second server interprets the data meaning of the multilevel behavior table to obtain a feature calculation table reflecting the degree of interest of the user end to the network product on the service platform, and further includes:
the forgetting module of the second server is provided with a forgetting function, calculates the time length from the time point of the first appearance of the behavior attribute to the current time point in the operation record, and sets the time length as the total period value of the behavior attribute;
and the forgetting module adjusts the interest characteristic data in the characteristic calculation table through an accumulated adjustment value, a long-period adjustment value, a first short-period adjustment value and a second short-period adjustment value so as to enable the interest characteristic data to accord with a forgetting rule of a real user.
Preferably, the clustering operation performed by the third server on the data in the feature calculation table to obtain the feature tag that is the closest match to the feature calculation table includes:
a preprocessing module of the third-party server prepares a feature calculation table into a feature calculation matrix, wherein each element in the feature calculation matrix corresponds to a value in the feature calculation table one by one;
and the clustering module of the third server is provided with a mature clustering analysis model, and the clustering module records the characteristic calculation matrix into the clustering analysis model to classify the clustering analysis model so as to obtain the characteristic label which is most matched with the characteristic calculation table.
In order to achieve the above object, the present invention further provides an artificial intelligence based data acquisition and interpretation device, which comprises:
the acquisition module is used for acquiring operation records of a user side on a service platform by a network data acquisition device to generate interest information and sending the interest information to the first server;
the dividing module is used for dividing the interest information by the first server according to the behavior attribute and the generation time to generate a multi-level behavior table and sending the multi-level behavior table to the second server;
the interpretation module is used for the second server to interpret the data meaning of the multi-level behavior table, obtain a characteristic calculation table reflecting the attention degree of the user end to the network product on the service platform, and send the characteristic calculation table to the third server;
and the clustering module is used for carrying out clustering operation on the data in the characteristic calculation table by the third server to obtain a characteristic label which is most matched with the characteristic calculation table, and sending the interest information, the characteristic calculation table and the characteristic label to the data server for storage.
In order to achieve the above object, the present invention also provides an electronic device, including:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores a program executable by the at least one processor to enable the at least one processor to perform the artificial intelligence based data collection interpretation method.
In order to achieve the above object, the present invention further provides a computer readable storage medium, wherein the computer readable storage medium stores an artificial intelligence based data acquisition and interpretation program, and when the artificial intelligence based data acquisition and interpretation program is executed by a processor, the steps of the artificial intelligence based data acquisition and interpretation method are implemented.
According to the method and the device, the interest information is divided according to the behavior attributes and the generation time to obtain the interest characteristic data, and the interest characteristic data of each behavior attribute is summarized to form a multi-level behavior table, so that the technical effect of describing the data information of the interest information times is achieved. The behavior interest of the user is accurately identified, the true intention of the user is accurately identified, and the situation that the true intention of the user is judged by mistake due to the fact that the user behavior is identified through simple weighted summation at present is avoided.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a diagram of an electronic device according to a preferred embodiment of the present invention;
FIG. 2 is a block diagram of the preferred embodiment of the artificial intelligence based data acquisition and interpretation apparatus of FIG. 1;
FIG. 3 is a flow chart of a preferred embodiment of the data acquisition and interpretation method based on artificial intelligence of the present invention;
the implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a schematic diagram of an electronic device 1 according to a preferred embodiment of the invention.
The electronic device 1 includes but is not limited to: memory 11, processor 12, display 13, and network interface 14. The electronic device 1 is connected to a network through a network interface 14 to obtain raw data. The network may be a wireless or wired network such as an Intranet (Internet), the Internet (Internet), a Global System for mobile communications (GSM), Wideband Code Division Multiple Access (WCDMA), a 4G network, a 5G network, Bluetooth (Bluetooth), or a Wi-Fi communication network.
The memory 11 includes at least one type of readable storage medium including a flash memory, a hard disk, a multimedia card, a card type memory (e.g., SD or DX memory, etc.), a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a Read Only Memory (ROM), an Electrically Erasable Programmable Read Only Memory (EEPROM), a Programmable Read Only Memory (PROM), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the storage 11 may be an internal storage unit of the electronic device 1, such as a hard disk or a memory of the electronic device 1. In other embodiments, the memory 11 may also be an external storage device of the electronic device 1, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like equipped with the electronic device 1.
Of course, the memory 11 may also comprise both an internal memory unit and an external memory device of the electronic device 1. In this embodiment, the memory 11 is generally used for storing an operating system installed in the electronic device 1 and various application software, such as a program code of the artificial intelligence based data acquisition and interpretation program 10. Further, the memory 11 may also be used to temporarily store various types of data that have been output or are to be output.
Processor 12 may be a Central Processing Unit (CPU), controller, microcontroller, microprocessor, or other data Processing chip in some embodiments. The processor 12 is typically used for controlling the overall operation of the electronic device 1, such as performing data interaction or communication related control and processing. In this embodiment, the processor 12 is configured to run the program code stored in the memory 11 or process data, for example, run the program code of the artificial intelligence based data acquisition and interpretation program 10.
The display 13 may be referred to as a display screen or display unit. In some embodiments, the display 13 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an Organic Light-emitting diode (OLED) touch screen, or the like. The display 13 is used for displaying information processed in the electronic device 1 and for displaying a visual work interface, e.g. displaying the results of data statistics.
The network interface 14 may optionally comprise a standard wired interface, a wireless interface (e.g. WI-FI interface), the network interface 14 typically being used for establishing a communication connection between the electronic device 1 and other electronic devices.
Fig. 1 shows only the electronic device 1 with the components 11-14 and the artificial intelligence based data acquisition interpretation program 10, but it is to be understood that not all of the shown components are required and that more or fewer components may be implemented instead.
Optionally, the electronic device 1 may further comprise a target user interface, the target user interface may comprise a Display (Display), an input unit such as a Keyboard (Keyboard), and the optional target user interface may further comprise a standard wired interface and a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an Organic Light-Emitting Diode (OLED) touch screen, or the like. The display, which may also be referred to as a display screen or display unit, is suitable for displaying information processed in the electronic device 1 and for displaying a visualized target user interface, among other things.
The electronic device 1 may further include a Radio Frequency (RF) circuit, a sensor, an audio circuit, and the like, which are not described in detail herein.
In the above embodiment, the processor 12, when executing the artificial intelligence based data collection interpretation program 10 stored in the memory 11, may implement the following steps:
the network data acquisition unit acquires operation records of a user side on a service platform to generate interest information and sends the interest information to the first server;
the first server divides the interest information according to the behavior attribute and the generation time to generate a multi-level behavior table and sends the multi-level behavior table to a second server;
the second server reads the data meaning of the multi-level behavior table, obtains a characteristic calculation table reflecting the attention degree of the user end to the network product on the service platform, and sends the characteristic calculation table to a third server;
and the third server performs clustering operation on the data in the characteristic calculation table to obtain a characteristic label which is most matched with the characteristic calculation table, and sends the interest information, the characteristic calculation table and the characteristic label to a data server for storage.
For the detailed description of the above steps, please refer to the following description of fig. 2 regarding a functional block diagram of an embodiment of the artificial intelligence based data collecting and interpreting apparatus 100 and fig. 3 regarding a flowchart of an embodiment of an artificial intelligence based data collecting and interpreting method.
Referring to fig. 2, a functional block diagram of the data acquisition and interpretation device 100 based on artificial intelligence is shown.
The artificial intelligence-based data acquisition and interpretation device 100 can be installed in electronic equipment. According to the implemented functions, the artificial intelligence based data collection and interpretation device 100 may include: an acquisition module 110, a division module 120, an interpretation module 130, and a clustering module 140. The module of the present invention, which may also be referred to as a unit, refers to a series of computer program segments that can be executed by a processor of an electronic device and that can perform a fixed function, and that are stored in a memory of the electronic device.
In the present embodiment, the functions regarding the respective modules/units are as follows:
the acquisition module 110 is used for acquiring operation records of a user side on a service platform by a network data acquisition device to generate interest information and sending the interest information to a first server;
a dividing module 120, configured to divide the interest information by the first server according to the behavior attribute and the generation time to generate a multi-level behavior table, and send the multi-level behavior table to the second server;
an interpretation module 130, configured to interpret the data meaning of the multilevel behavior table by the second server, obtain a feature calculation table reflecting the degree of interest of the user end to the network product on the service platform, and send the feature calculation table to a third server;
and the clustering module 140 is configured to perform clustering operation on the data in the feature calculation table by the third server to obtain a feature tag that is most matched with the feature calculation table, and send the interest information, the feature calculation table, and the feature tag to the data server for storage.
In addition, the invention also provides a data acquisition and interpretation method based on artificial intelligence. Fig. 3 is a schematic method flow chart of an embodiment of the artificial intelligence-based data acquisition and interpretation method according to the present invention. When the processor 12 of the electronic device 1 executes the artificial intelligence based data acquisition and interpretation program 10 stored in the memory 11, the artificial intelligence based data acquisition and interpretation method is implemented, which includes steps S101-S104. The respective steps will be specifically described below.
S101, a network data collector collects operation records of a user side on a service platform to generate interest information and sends the interest information to a first server.
Specifically, the acquiring, by the network data collector, an operation record of a user on a service platform is collected to generate interest information, and the interest information is sent to the first server, including:
a1, the acquisition module of the network data acquisition unit acquires the unique identifier of the user terminal and acquires the operation record on the service platform according to the unique identifier.
Illustratively, the acquisition module accesses a user database of a service platform through an identification unit, and acquires a unique identification (e.g., an MEI code, or a network card address, or a bluetooth address, or a device ID, etc.) of a user terminal registered, or logged in, or temporarily accessed at the service platform from the user database; for example: obtaining a unique identifier: 868034031518269.
the acquisition module crawls the operation record of the unique identifier on a service platform by using a general crawler through a crawler unit so as to obtain the operation record of the user side corresponding to the unique identifier; the operation record at least comprises a behavior attribute and generation time, wherein the behavior attribute is information describing behaviors of the operation record, such as collecting behaviors, commenting behaviors, praise behaviors and click behaviors; the generation time is information describing behavior occurrence time of the operation record, and in the application, the generation time is a timestamp generated when the user operates a collection behavior, a comment behavior or a click behavior. For example, the acquired operation record includes "2021-09-10, favorite behavior", "2021-07-10, comment behavior", "2021-05-10, click behavior", "2021-05-10, like behavior".
And A2, an information module of the network data collector collects the unique identification and the operation record to form interest information.
For example: and summarizing the unique identifier and the operation record in the same table. The following were used:
868034031518269, operation record
Figure BDA0003316155920000071
S102, the first server divides the interest information according to the behavior attributes and the generation time to generate a multi-level behavior table, and sends the multi-level behavior table to the second server.
Specifically, the dividing, by the first server, the interest information according to the behavior attribute and the generation time to generate a multi-level behavior table includes:
and B1, the decomposition module of the first server divides the interest information according to the behavior attributes to form behavior decomposition information, and sends the behavior decomposition information to the long-term data module of the first server.
In the step, the decomposition module sequentially identifies each behavior attribute in the interest information, and summarizes the interest information with the same behavior attribute to form behavior decomposition information; for example: behavior decomposition information with behavior attribute of 'collecting behavior' is obtained as follows:
Figure BDA0003316155920000072
Figure BDA0003316155920000081
and B2, the long-term data module obtains the total amount of the behavior decomposition information and sets the total amount as an interest accumulated value, subtracts the current time from a preset long period value to obtain a long-term time point, obtains the amount of the behavior decomposition information between the current time and the long-term time point, and sets the amount as a long-term interest value.
Wherein the long period value is time data describing a period span, which can be set by a user as desired.
For example: assume the current time is 2021-12-31 and the long period value is 1 year;
based on the above example, the long-term interest information is shown in the following table:
behavior decomposition information Behavioral decomposition information between current time and long-term time points
2019-01-10, Collection behavior 2021-12-10, collecting behavior
2019-03-10, Collection behavior 2021-07-10, Collection behavior
2020-09-10, Collection behavior 2021-10-10, Collection behavior
2020-07-10, Collection behavior 2021-05-10, Collection behavior
2021-12-10, collecting behavior
2021-07-10, Collection behavior
2021-10-10, Collection behavior
2021-05-10, Collection behavior
Thus, the cumulative interest value and long-term interest value are shown in the following table:
accumulated value of interest Long term interest value
8 4
B3, the short-term data module of the first server subtracts the current time from a preset first short-term value to obtain a first short-term time point, obtains the amount of behavior decomposition information between the current time and the first short-term time point, and sets the amount as a first short-term interest value; and subtracting the preset second short period value from the current time to obtain a second short period time point, acquiring the quantity of the behavior decomposition information between the current time and the second short period time point, and setting the quantity as a second short period interest value.
The first short period value and the second short period value are time data describing a period span, which can be set by a user according to needs.
For example, the first short period value is 3 months and the second short period value is 1 month;
Figure BDA0003316155920000082
thus, the first short cycle interest value and the second short cycle interest value are shown in the following table:
first short period interest value Second short period interest value
2 1
And B4, the summarizing module of the first server takes the interest accumulated value, the long-term interest value, the first short-period interest value and the second short-period interest value as interest characteristic data, and the summarizing module summarizes the interest characteristic data of each behavior attribute to obtain a multi-level behavior table.
For example: based on the above example, the interest feature data is shown in the following table:
behavior attributes Collecting behavior
Accumulated value of interest 8
Long term interest value 4
First short period interest value 2
Second short period interest value 1
It should be noted that the decomposition module, the long-term data module and the short-term data module in the first server in the present application are spark sql modules established under a spark framework based on a hadoop platform, and the modules are configured to calculate the quantity of behavior decomposition information, the quantity of behavior decomposition information between the current time and the long-term time point, the quantity of behavior decomposition information between the current time and the first short-term time point, and the quantity of behavior decomposition information between the current time and the second short-term time point, so as to obtain an interest cumulative value, a long-term interest value, a first short-term interest value, and a second short-term interest value, and store the interest cumulative value, the long-term interest value, the first short-term interest value, and the second short-term interest value in a summary module with a HIVE data table, so as to obtain interest feature data.
The hadoop platform is a system platform for realizing a distributed file system, has the characteristic of high fault tolerance, and can provide high throughput to access data of an application program, so that a user can develop the distributed program without knowing details of a distributed bottom layer, and the distributed program can be operated and stored at a high speed by fully utilizing the power of a cluster.
The Spark framework Spark is a fast, general-purpose computing engine designed specifically for large-scale data processing. The sparkSQL module is a module for processing structured data, and can count the number of behavior decomposition information, the number of behavior decomposition information between the current time and the long-term time point, the number of behavior decomposition information between the current time and the first short-term time point, and the number of behavior decomposition information between the current time and the second short-term time point through a count function to obtain an interest cumulative value, a long-term interest value, a first short-term interest value, and a second short-term interest value.
The HIVE data table is a data warehouse processing tool with Hadoop packaged at the bottom layer, data query is realized by using SQL-like HiveQL language, and all HIVE data are stored in a Hadoop compatible file system.
In this embodiment, the summarization module includes a data summarization unit and a data cleansing unit.
The data summarization unit summarizes the interest characteristic data of each behavior attribute to form a primary behavior table.
It should be noted that the preliminary behavior table is formed by summarizing the HIVE data tables of the behavior attributes by the data summarization unit.
And the data cleaning unit is used for cleaning the data of the behavior decomposition information in the primary behavior table to remove invalid data in the behavior index table, so that the behavior index table is converted into a multi-level behavior table.
It should be noted that the data in the primary behavior table is flushed by using python as a data flushing unit to remove test data and abnormal data, and null values and garbage data are flushed. The technical problem solved by the present application is how to obtain a multi-level behavior table capable of describing interesting information features, rather than how to clean data, and meanwhile, a person skilled in the art can clean data by using python through common general knowledge, so that the process of cleaning data by python is not described in detail in the present application.
Based on the above example, the multi-level behavior table is shown as the following table:
behavior attributes Collecting behavior Commenting behaviors Like behavior
Accumulated value of interest 8 9 2
Long term interest value 4 3 2
First short period interest value 2 2 2
Second short period interest value 1 2 2
S103, the second server interprets the data meaning of the multi-level behavior table, obtains a characteristic calculation table reflecting the attention degree of the user end to the network product on the service platform, and sends the characteristic calculation table to a third server.
Specifically, the second server interprets the data meaning of the multilevel behavior table to obtain a feature calculation table reflecting the attention degree of the user end to the network product on the service platform, and the feature calculation table comprises:
and C1, writing a mean value formula in a first calculation module of the second server, and calculating the mean value of the interest feature data of each behavior attribute in the multi-level behavior table by the first calculation module through the mean value formula to obtain an index mean value of each behavior attribute and write the index mean value into the multi-level behavior table to form a first behavior calculation table.
In this step, the average formula is as follows:
Figure BDA0003316155920000101
in this embodiment, the number i of the interest cumulative value is 1, the number i of the long-term interest value is 2, the number i of the first short-period interest value is 3, and the number i of the second short-period interest value is 4; n is the number of the interest feature data, and in the embodiment, the number n of the interest feature data is 4; z refers to the feature data of interest, so ZijThe value of the interest characteristic data with the number of i under the behavior attribute with the number of j;
Figure BDA0003316155920000102
means the mean value of the index.
For example, the first behavior calculation table for obtaining the behavior attribute of "favorite behavior" based on the above example is as follows:
behavior attributes Collecting behavior
Accumulated value of interest 8
Long term interest value 4
First short period interest value 2
Second short period interest value 1
Mean value of index 3.75
And C2, writing a variance formula in a second calculation module of the second server, and calculating the index average index and the interest feature data of each behavior attribute in the first behavior calculation table by the second calculation module through the variance formula to obtain the index variance of each behavior attribute and write the index variance into the first behavior calculation table to form a second behavior calculation table.
In this step, the variance formula is as follows:
Figure BDA0003316155920000111
wherein Z isijRefers to the value of the interest feature data with the number i under the behavior attribute with the number j,
Figure BDA0003316155920000112
means the index mean, σ, of the behavior attribute numbered jjRefers to the index variance of the behavior attribute numbered j.
For example, the second behavior calculation table in which the behavior attribute is "favorite behavior" is obtained based on the above example is shown in the following table:
behavior attributes Collecting behavior
Accumulated value of interest 8
Long term interest value 4
First short period interest value 2
Second short period interest value 1
Mean value of index 3.75
Index variance 4.918
By calculating the variance of the interest accumulation value, the long-term interest value, the first short-period interest value and the second short-period interest value, if the variance value is larger, the behavior that a user always collects, reviews or approves the network product from history to the present is expressed, so that the user always keeps strong operation interest on the network product; if the variance value is smaller, expressing that the user operates the network product for the first time or operates the network product historically without operating the network product in the next time, so that the interest level of the user in operating the network product is not high; in a word, the behavioral interest of the user is effectively identified by analyzing the interest characteristic data through the variance algorithm model.
And C3, the third calculation module of the second server adds the index variances of the behavior contents in the second behavior calculation table to obtain a total variance of the behavior.
For example, a second behavior calculation table having all behavior attributes obtained based on the above example is as follows:
behavior attributes Collecting behavior Commenting behaviors Like behavior
Accumulated value of interest 8 9 2
Long term interest value 4 3 2
First short period interest value 2 2 2
Second short period interest value 1 2 2
Mean value of index 3.75 4 2
Index variance 4.918 5.831 0
The total variance was obtained as 4.918+5.831+ 0-10.749 based on the above table.
Therefore, based on the behavioral interest identification of the user, all the behavioral interests of the user are quantitatively summarized through the index variance, and the total variance is obtained, so that the service system can be helped to effectively identify whether the user has interest in the network product or not based on the total variance.
And C4, the fourth calculation module of the second server has a proportion formula, the fourth calculation module calculates the proportion of the index variance of each behavior content in the second behavior calculation table in the total behavior variance through the proportion formula, sets the proportion as the index weight of each behavior content, and writes the index weight and the total variance into the second behavior calculation table to form a feature calculation table.
The step calculates the proportion of various behaviors of the user on the network products in the total behaviors of the user through a proportion formula so as to identify the behavior preference of the user on the network products which are interested.
The proportion formula is as follows:
Figure BDA0003316155920000121
wherein, WjMeans the ratio of the behavior numbered j to the total behavior, σjRefers to the index variance of the behavior attribute with the number j, and ρ refers to the number of behavior attributes in the primary attention feature.
For example, a feature calculation table obtained based on the above example is as follows:
Figure BDA0003316155920000122
therefore, the occupation ratio of each behavior attribute in all behaviors of the user is quantitatively expressed through the index weight, and the occupation ratio of various behaviors of the user on the network product in the total behaviors of the user is quantized.
Specifically, the second server interprets the data meaning of the multilevel behavior table to obtain a feature calculation table reflecting the attention degree of the user end to the network product on the service platform, and further includes:
and C5, the forgetting module of the second server has a forgetting function, the forgetting module calculates the duration between the time point of the first appearance of the behavior attribute in the operation record and the current time point, and sets the duration as the total period value of the behavior attribute.
The forgetting module calculates the total period value, the long period value, the first short period value and the second short period value through a forgetting function so as to obtain an accumulated adjustment value, a long period adjustment value, a first short period adjustment value and a second short period adjustment value; wherein, the forgetting function is a function formula conforming to the forgetting characteristic.
In this step, the forgetting function is as follows:
Figure BDA0003316155920000123
where f is the adjustment value, t is the forgetting coefficient, and in this embodiment, t is the adjustment value0Assuming that the cumulative adjustment value f1 is 0.4, the long period adjustment value f2 is 0.8, the first short period adjustment value f3 is 0.9, and the second short period adjustment value f4 is 1, according to the above settings, 0.0025 and c is 0.025; the forgetting function accords with the forgetting characteristic, namely the larger the period is, the smaller the weight is according to the forgetting rule of people.
And C6, the forgetting module adjusts the interest characteristic data in the characteristic calculation table through the accumulated adjustment value, the long period adjustment value, the first short period adjustment value and the second short period adjustment value, so that the interest characteristic data conform to the forgetting rule of the real user.
For example: based on the above example: obtaining a characteristic calculation table:
Figure BDA0003316155920000131
by adopting the forgetting function which accords with the forgetting characteristic, the accumulated adjustment value, the long-period adjustment value, the first short-period adjustment value and the second short-period adjustment value are obtained, so that the obtained characteristic calculation table is more accordant with the forgetting rule of a person, the characteristic calculation table is more real, and the accuracy of the characteristic calculation table is further ensured.
And S104, the third server performs clustering operation on the data in the feature calculation table to obtain a feature tag which is most matched with the feature calculation table, and sends the interest information, the feature calculation table and the feature tag to a data server for storage.
Specifically, the clustering operation performed by the third server on the data in the feature calculation table to obtain the feature tag that is the closest match to the feature calculation table includes:
d1, the preprocessing module of the third-party server makes the feature calculation table into a feature calculation matrix, wherein, the values of each element in the feature calculation matrix are in one-to-one correspondence with the values in the feature calculation table.
Exemplarily, the following steps are carried out: obtaining a characteristic calculation table:
Figure BDA0003316155920000132
deleting data belonging to the calculation process in the characteristic calculation table, and making the characteristic calculation table into a characteristic calculation matrix X as follows:
Figure BDA0003316155920000141
wherein, the data belonging to the calculation process comprises index mean value and index weight.
D2, the clustering module of the third server is provided with a mature cluster analysis model, and the clustering module inputs a feature calculation matrix into the cluster analysis model to classify the feature calculation matrix so as to obtain a feature label which is most matched with a feature calculation table.
In an exemplary embodiment, the mature cluster analysis model is obtained by:
acquiring a historical operation record with a characteristic label from a historical database;
processing the historical operation records according to the steps of S101-S103 to obtain a historical characteristic calculation table, and processing the historical characteristic calculation table according to the step of D1 to obtain a historical characteristic calculation matrix;
recording a plurality of historical characteristic calculation matrixes into an initial clustering algorithm, wherein the clustering algorithm divides a characteristic calculation table corresponding to each characteristic calculation matrix into a plurality of stable clusters according to the historical characteristic calculation matrixes so as to obtain a mature clustering analysis model; wherein each cluster has a center point.
In the step, the characteristic calculation matrix can be regarded as one point in a clustering algorithm, the clustering algorithm judges that the adjustment attention matrix is suitable to be divided into a plurality of clusters according to attraction information and attribution information among the points, and the adjustment attention matrix or the adjustment attention matrices are suitable to be the central point of the clusters.
It should be noted that the clustering algorithm used in the present application is an AP clustering algorithm, which is a computer algorithm that performs iterative computation on data points based on attraction information and attribution information between the data points to finally obtain clusters and cluster center points; a person skilled in the art can use the algorithm to calculate any data to obtain classification information, however, the technical problem solved by the present application is how to make information classification more accurate, and therefore, the specific working principle of the clustering algorithm is not described in detail in the present application.
Further, the step of the clustering module inputting the feature calculation matrix into the cluster analysis model to classify the cluster analysis model so as to obtain the feature label which is most matched with the feature calculation table comprises:
the clustering module records a characteristic calculation matrix into the clustering analysis model and calculates Euclidean distances between the characteristic calculation matrix and the central points of clusters in the clustering analysis model;
setting the cluster where the central point with the shortest Euclidean distance is located as a target cluster to realize the classification of the feature calculation matrix;
and extracting feature tags of all historical operation records in the target cluster, and taking the feature tag with the largest proportion as the feature tag which is most matched with the feature calculation table.
In an exemplary embodiment, after the clustering operation is performed on the data in the feature calculation table by the third server in S104 to obtain the feature tag that best matches the feature calculation table, the method may further include:
d3, the image module of the third server multiplies the interest characteristic data of each behavior attribute in the characteristic calculation table by the index weight to respectively obtain the preference value of each behavior attribute, and the preference value is loaded into the characteristic calculation table to form the image attention characteristic.
For example: based on the above example:
preference for collection behavior (3.2+3.2+1.8+1) × 0.47 ═ 4.324;
preference value of comment behavior ═ (3.6+2.4+1.8+2) × 0.53 ═ 5.194;
the preference value for the praise behavior is (0.8+1.6+1.8+2) × 0 is 0.
Thus, the image interest characteristics obtained are shown in the following table:
Figure BDA0003316155920000151
a quantitative data basis is provided for completing the behavior portraits of the users by calculating the preference values, and the intuitiveness and the reliability of the data are guaranteed.
D4, the representation module creates a behavioral representation describing the user interest feature from the representation attention feature.
Further, the portrait module in D4 creating a behavioral portrait for describing the user interest feature from the portrait focus feature comprises:
d4-1: extracting the total variance of the image attention features, and comparing the total variance with a preset interest threshold; if the total variance is smaller than the interest threshold, judging that the user has no interest in the network product and ending; if the total variance is larger than an interest threshold, setting the network product as a preference product, extracting a preference value of the portrait attention feature, acquiring a behavior attribute corresponding to the maximum preference value, and setting the behavior attribute as a first preference attribute;
d4-2: comparing the preference value with a preset behavior attribute threshold value, and setting the behavior attribute corresponding to the preference value larger than the behavior attribute threshold value as an attention preference attribute;
d4-3: and summarizing the product name, the first preference attribute and the attention preference attribute of the preference product to obtain a behavior portrait for describing the user interest characteristics.
Therefore, the behavior portrait expresses the network products preferred by the user, and the user can intuitively obtain the behavior preference of the user by observing the preference attribute in the behavior portrait through which operation describes the preference of the network products.
For example, assuming that the interest threshold is 5, based on the above example, since the total variance is 10.749, the network product is set as a preferred product, so that the preference value of the portrait attention feature is extracted, and the behavior attribute corresponding to the maximum preference value is obtained, that is: commenting on the behavior, so setting the commenting behavior as a first preference attribute; assume that the behavior attribute threshold includes: presetting a collection threshold, a comment threshold and a like threshold; suppose that the favorites threshold, the comment threshold, and the like are: 2. 3, 4; based on the above example, since the preference value of the comment behavior and the preference value of the favorite behavior are respectively greater than the favorite threshold and the comment threshold, the comment behavior and the favorite behavior are set as the attention preference attribute; extracting the product name "A" of the preference product, summarizing the product name, the first preference attribute and the attention preference attribute, obtaining a behavior portrait and showing the behavior portrait in the following table:
preference product A
First preference attribute Commenting behaviors
Attention preference attribute Comment behavior, collection behavior
According to the method and the device, the interest information is divided according to the behavior attributes and the generation time to obtain the interest characteristic data, and the interest characteristic data of each behavior attribute is summarized to form a multi-level behavior table, so that the technical effect of describing the data information of the interest information times is achieved.
And calculating the variance of each interest feature data in the multi-level behavior table through a variance algorithm model, analyzing the variance to obtain a primary concern feature for describing the concern degree of the user on the network product, and accurately identifying the behavior interest of the user by adopting a variance screening mode to further accurately identify the real intention of the user, so that the condition that the real intention of the user is judged by mistake due to the fact that the user behavior is identified through simple weighted summation at present is avoided.
Adjusting the primary attention feature according to the forgetting characteristic through a forgetting algorithm model to obtain an adjusted attention feature, and creating a behavior portrait for describing the user interest feature according to the adjusted attention feature; because the attention adjusting characteristic is characteristic data which reflects the real attention degree of the user to the network product based on the forgetting characteristic, the attention adjusting characteristic better accords with the forgetting rule of the user, so that the behavior portrait is more real, and the accuracy and the authenticity of the behavior portrait are ensured.
Furthermore, the embodiment of the present invention also provides a computer-readable storage medium, which may be any one or any combination of a hard disk, a multimedia card, an SD card, a flash memory card, an SMC, a Read Only Memory (ROM), an Erasable Programmable Read Only Memory (EPROM), a portable compact disc read only memory (CD-ROM), a USB memory, and the like. The computer readable storage medium comprises a storage data area and a storage program area, the storage data area stores data created according to the use of the block chain nodes, the storage program area stores an artificial intelligence based data acquisition and interpretation program 10, and when the artificial intelligence based data acquisition and interpretation program 10 is executed by a processor, the operation of the artificial intelligence based data acquisition and interpretation method is realized.
In another embodiment, in order to further ensure the privacy and security of all the data, all the data may be stored in a node of a block chain.
It should be noted that the blockchain in the present invention is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, and encryption algorithm. A block chain (Blockchain), which is essentially a decentralized database, is a series of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, so as to verify the validity (anti-counterfeiting) of the information and generate a next block. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.
The specific implementation of the computer-readable storage medium of the present invention is substantially the same as the above-mentioned data acquisition and interpretation method based on artificial intelligence, and will not be described herein again.
It should be noted that the above-mentioned numbers of the embodiments of the present invention are merely for description, and do not represent the merits of the embodiments. And the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, apparatus, article, or method that includes the element.
It should be noted that, the above embodiments of the present invention may acquire and process related data based on an artificial intelligence technique. Among them, Artificial Intelligence (AI) is a theory, method, technique and application system that simulates, extends and expands human Intelligence using a digital computer or a machine controlled by a digital computer, senses the environment, acquires knowledge and uses the knowledge to obtain the best result.
The artificial intelligence infrastructure generally includes technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) as described above, and includes instructions for enabling an electronic device (such as a mobile phone, a computer, an electronic apparatus, or a network device) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.

Claims (10)

1. A method for data acquisition and interpretation based on artificial intelligence, the method comprising:
the network data acquisition unit acquires operation records of a user side on a service platform to generate interest information and sends the interest information to the first server;
the first server divides the interest information according to the behavior attribute and the generation time to generate a multi-level behavior table and sends the multi-level behavior table to a second server;
the second server reads the data meaning of the multi-level behavior table, obtains a characteristic calculation table reflecting the attention degree of the user end to the network product on the service platform, and sends the characteristic calculation table to a third server;
and the third server performs clustering operation on the data in the characteristic calculation table to obtain a characteristic label which is most matched with the characteristic calculation table, and sends the interest information, the characteristic calculation table and the characteristic label to a data server for storage.
2. The artificial intelligence based data collection and interpretation method of claim 1, wherein the network data collector collects operation records of a user terminal on a service platform to generate interest information and sends the interest information to a first server, comprising:
the method comprises the steps that an acquisition module of a network data acquisition unit acquires a unique identifier of a user side and acquires an operation record on a service platform according to the unique identifier;
and an information module of the network data acquisition unit collects the unique identifier and the operation record to form interest information.
3. The artificial intelligence based data collection interpretation method of claim 1, wherein the first server partitioning the interest information by behavior attributes and generation time to generate a multi-level behavior table, comprises:
the decomposition module of the first server divides the interest information according to behavior attributes to form behavior decomposition information, and sends the behavior decomposition information to the long-term data module of the first server;
the long-term data module acquires the total amount of the behavior decomposition information and sets the total amount as an interest accumulated value, subtracts the current time from a preset long-period value to obtain a long-term time point, acquires the amount of the behavior decomposition information between the current time and the long-term time point, and sets the amount as a long-term interest value;
the short-term data module of the first server subtracts the current time from a preset first short-term value to obtain a first short-term time point, obtains the quantity of behavior decomposition information between the current time and the first short-term time point, and sets the quantity as a first short-term interest value; subtracting a preset second short period value from the current time to obtain a second short period time point, acquiring the quantity of behavior decomposition information between the current time and the second short period time point, and setting the quantity as a second short period interest value;
and the summarizing module of the first server takes the interest accumulated value, the long-term interest value, the first short-period interest value and the second short-period interest value as interest characteristic data, and summarizes the interest characteristic data of each behavior attribute to obtain a multi-level behavior table.
4. The artificial intelligence based data collection and interpretation method of claim 3, wherein the summarization module comprises a data summarization unit and a data cleansing unit.
5. The artificial intelligence-based data acquisition and interpretation method of claim 1, wherein the second server interprets the data meaning of the multilevel behavior table to obtain a feature calculation table reflecting the attention degree of the user to the network product on the service platform, comprising:
a first calculation module of the second server writes a mean value formula, and the first calculation module calculates a mean value of interest characteristic data of each behavior attribute in the multi-level behavior table through the mean value formula to obtain an index mean value of each behavior attribute and writes the index mean value into the multi-level behavior table to form a first behavior calculation table;
a second calculation module of the second server writes a variance formula, and the second calculation module calculates the index average index and the interest characteristic data of each behavior attribute in the first behavior calculation table through the variance formula to obtain the index variance of each behavior attribute and writes the index variance into the first behavior calculation table to form a second behavior calculation table;
a third calculation module of the second server adds the index variances of the behavior contents in a second behavior calculation table to obtain a behavior total variance;
and a fourth calculation module of the second server is provided with a proportion formula, calculates the proportion of the index variance of each behavior content in the second behavior calculation table in the behavior total variance through the proportion formula, sets the proportions as the index weights of each behavior content respectively, and writes the index weights and the total variance into the second behavior calculation table to form a feature calculation table.
6. The artificial intelligence based data acquisition and interpretation method of claim 1, wherein the second server interprets the data meaning of the multilevel behavior table to obtain a feature calculation table reflecting the degree of interest of the user to the network product on the service platform, further comprising:
the forgetting module of the second server is provided with a forgetting function, calculates the time length from the time point of the first appearance of the behavior attribute to the current time point in the operation record, and sets the time length as the total period value of the behavior attribute;
and the forgetting module adjusts the interest characteristic data in the characteristic calculation table through an accumulated adjustment value, a long-period adjustment value, a first short-period adjustment value and a second short-period adjustment value so as to enable the interest characteristic data to accord with a forgetting rule of a real user.
7. The artificial intelligence based data collection interpretation method of claim 1, wherein the third server performs a clustering operation on the data in the feature calculation table to obtain a feature label that best matches the feature calculation table, comprising:
a preprocessing module of the third-party server prepares a feature calculation table into a feature calculation matrix, wherein each element in the feature calculation matrix corresponds to a value in the feature calculation table one by one;
and the clustering module of the third server is provided with a mature clustering analysis model, and the clustering module records the characteristic calculation matrix into the clustering analysis model to classify the clustering analysis model so as to obtain the characteristic label which is most matched with the characteristic calculation table.
8. An artificial intelligence-based data acquisition and interpretation device, the device comprising:
the acquisition module is used for acquiring operation records of a user side on a service platform by a network data acquisition device to generate interest information and sending the interest information to the first server;
the dividing module is used for dividing the interest information by the first server according to the behavior attribute and the generation time to generate a multi-level behavior table and sending the multi-level behavior table to the second server;
the interpretation module is used for the second server to interpret the data meaning of the multi-level behavior table, obtain a characteristic calculation table reflecting the attention degree of the user end to the network product on the service platform, and send the characteristic calculation table to the third server;
and the clustering module is used for carrying out clustering operation on the data in the characteristic calculation table by the third server to obtain a characteristic label which is most matched with the characteristic calculation table, and sending the interest information, the characteristic calculation table and the characteristic label to the data server for storage.
9. An electronic device, characterized in that the electronic device comprises:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores a program executable by the at least one processor to enable the at least one processor to perform the artificial intelligence based data collection interpretation method as recited in any one of claims 1 to 7.
10. A computer-readable storage medium storing an artificial intelligence based data acquisition and interpretation program, wherein the artificial intelligence based data acquisition and interpretation program, when executed by a processor, implements the steps of the artificial intelligence based data acquisition and interpretation method according to any one of claims 1 to 7.
CN202111231734.7A 2021-10-22 2021-10-22 Data acquisition and interpretation method, device, equipment and medium based on artificial intelligence Pending CN113946751A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111231734.7A CN113946751A (en) 2021-10-22 2021-10-22 Data acquisition and interpretation method, device, equipment and medium based on artificial intelligence

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111231734.7A CN113946751A (en) 2021-10-22 2021-10-22 Data acquisition and interpretation method, device, equipment and medium based on artificial intelligence

Publications (1)

Publication Number Publication Date
CN113946751A true CN113946751A (en) 2022-01-18

Family

ID=79332035

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111231734.7A Pending CN113946751A (en) 2021-10-22 2021-10-22 Data acquisition and interpretation method, device, equipment and medium based on artificial intelligence

Country Status (1)

Country Link
CN (1) CN113946751A (en)

Similar Documents

Publication Publication Date Title
CN108416198B (en) Device and method for establishing human-machine recognition model and computer readable storage medium
CN107818344B (en) Method and system for classifying and predicting user behaviors
CN110866181B (en) Resource recommendation method, device and storage medium
CN108021651B (en) Network public opinion risk assessment method and device
CN110336838B (en) Account abnormity detection method, device, terminal and storage medium
CN111831636A (en) Data processing method, device, computer system and readable storage medium
CN112329816A (en) Data classification method and device, electronic equipment and readable storage medium
CN111310052A (en) User portrait construction method and device and computer readable storage medium
CN111612610A (en) Risk early warning method and system, electronic equipment and storage medium
CN110647995A (en) Rule training method, device, equipment and storage medium
CN106294406B (en) Method and equipment for processing application access data
CN116034379A (en) Activity level measurement using deep learning and machine learning
CN112069269B (en) Big data and multidimensional feature-based data tracing method and big data cloud server
CN117312657A (en) Abnormal function positioning method and device for financial application, computer equipment and medium
CN114495137B (en) Bill abnormity detection model generation method and bill abnormity detection method
CN113722437B (en) User tag identification method, device, equipment and medium based on artificial intelligence
CN115860835A (en) Advertisement recommendation method, device and equipment based on artificial intelligence and storage medium
CN113946751A (en) Data acquisition and interpretation method, device, equipment and medium based on artificial intelligence
CN115186188A (en) Product recommendation method, device and equipment based on behavior analysis and storage medium
CN104376021A (en) File recommending system and method
CN115051863A (en) Abnormal flow detection method and device, electronic equipment and readable storage medium
CN115062200A (en) User behavior mining method and system based on artificial intelligence
CN110674020B (en) APP intelligent recommendation method and device and computer readable storage medium
CN113420018A (en) User behavior data analysis method, device, equipment and storage medium
CN115617790A (en) Data warehouse creation method, electronic device and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination