CN115934675B - Dynamic tag output method and device, storage medium and electronic equipment - Google Patents
Dynamic tag output method and device, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN115934675B CN115934675B CN202211536454.1A CN202211536454A CN115934675B CN 115934675 B CN115934675 B CN 115934675B CN 202211536454 A CN202211536454 A CN 202211536454A CN 115934675 B CN115934675 B CN 115934675B
- Authority
- CN
- China
- Prior art keywords
- data
- label
- service
- sample data
- processor
- 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.)
- Active
Links
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method and a device for outputting a dynamic label, a storage medium and electronic equipment, wherein the method comprises the following steps: reading raw sample data from a customer relationship management CRM system; performing data cleaning and feature extraction on the original sample data, and outputting intermediate sample data meeting preset requirements; inputting the intermediate sample data into a pre-trained tag prediction model, outputting target tag data on the CRM system; and responding to a label calling request of a service end, and returning the target label data to the service end. The method and the device solve the technical problem that the related technology cannot generate the dynamic label in real time, and meet the real-time requirement of the service end on the label data.
Description
Technical Field
The present invention relates to the field of computer technologies, and in particular, to a method and apparatus for outputting a dynamic tag, a storage medium, and an electronic device.
Background
In the related art, the digital marketing of the merchant needs to perform data modeling on historical purchase data of consumers on an e-commerce platform, and due to the privacy protection of users, the main stream e-commerce platform has many restrictions on the output of consumer sensitive data, and the merchant cannot derive transaction data from a customer relationship management (Customer Relationship Management, CRM) system of the e-commerce platform in real time to perform data modeling.
In the related art, the prior art needs to apply for data domain-out pre-application to an e-commerce platform, and after a plurality of workdays pass the audit, the sensitive data needed by modeling can be obtained to the local and then the modeling prediction is performed. The conventional electronic commerce platform customer relationship management system does not provide or only provides basic label capability, and cannot provide dynamic labels generated through modeling prediction.
In view of the above problems in the related art, no effective solution has been found yet.
Disclosure of Invention
The embodiment of the application provides a dynamic tag output method and device, a storage medium and electronic equipment.
According to an aspect of an embodiment of the present application, there is provided a method for outputting a dynamic tag, including: reading raw sample data from a customer relationship management CRM system; performing data cleaning and feature extraction on the original sample data, and outputting intermediate sample data meeting preset requirements; inputting the intermediate sample data into a pre-trained tag prediction model, outputting target tag data on the CRM system; and responding to a label calling request of a service end, and returning the target label data to the service end.
Further, performing data cleaning and feature extraction on the original sample data, and outputting intermediate sample data meeting preset requirements includes: performing deduplication cleaning and format content cleaning on the original sample data to obtain first sample data; selecting and extracting a sample attribution identifier and a sample characteristic from the first sample data; and constructing a feature table by taking the sample attribution identifier as a key and the corresponding sample feature as a value, and determining the feature table as intermediate sample data.
Further, inputting the intermediate sample data into a pre-trained tag prediction model, outputting target tag data on the CRM system comprises: determining the type of the label to be output; if the tag type is a probability tag, selecting a probability prediction model from a model library, taking the intermediate sample data as input data of the probability prediction model, and outputting first target tag data from the probability prediction model locally by the CRM system; and if the label type is a preference label, selecting a commodity recommendation model from a model library, taking the intermediate sample data as input data of the commodity recommendation model, and outputting second target label data from the commodity recommendation model locally by the CRM system, wherein the model library comprises the probability prediction model and the commodity recommendation model.
Further, responding to a label calling request of a service end, and returning the target label data to the service end comprises one of the following steps: responding to a first label calling request of a service end, performing variable length character coding on target label data to generate json data, and returning the json data to the service end by adopting an Application Programming Interface (API) interface; and responding to a second tag call request of the service end, performing variable length character coding on the target tag data to generate file data, and returning the file data to the service end by adopting a file call interface.
Further, before reading the raw sample data from the customer relationship management CRM system, the method further comprises: initializing service processor classes of a plurality of scheduling containers, wherein the scheduling containers comprise a plurality of container spaces, each container space corresponds to one service processor class, each service processor class corresponds to one tag type, and each service processor class comprises a plurality of code data connected by a responsibility chain; after initialization is completed, acquiring service demand information, and analyzing a processor list and code environment parameters of the scheduling container according to the service demand information; if the processor list and the code environment parameters are successfully analyzed, analyzing class parameters of the processor list; and after the class parameters are successfully parsed, starting the scheduling container.
Further, initializing a traffic processor class of the scheduling container includes: comparing the scheduling minimized actions, and extracting an abstract template of the service processor according to a comparison result to obtain a virtual class of the service processor; generating a service processor class by adopting the service processor virtual class; and calling a main program of the scheduling container to initialize service processor classes of all service processors.
Further, before initializing the service processor class of the number of scheduling containers, the method further comprises: determining a target scheduling container currently executed; parsing the container token of the target scheduling container; and positioning the calling time sequence of the target scheduling container in a first-in first-out FIFO pipeline by adopting the container token.
According to another aspect of the embodiment of the present application, there is also provided an output device for a dynamic tag, including: the acquisition module is used for reading the original sample data from the Customer Relationship Management (CRM) system; the processing module is used for carrying out data cleaning and feature extraction on the original sample data and outputting intermediate sample data meeting preset requirements; the prediction module is used for inputting the intermediate sample data into a pre-trained label prediction model and outputting target label data on the CRM system; and the return module is used for responding to the label calling request of the service end and returning the target label data to the service end.
Further, the processing module includes: the processing unit is used for carrying out deduplication cleaning and format content cleaning on the original sample data to obtain first sample data; an extraction unit, configured to select and extract a sample home identifier and a sample feature from the first sample data; and the storage unit is used for constructing a feature table by taking the sample attribution identifier as a key and the corresponding sample feature as a value, and determining the feature table as intermediate sample data.
Further, the prediction module includes: a determining unit for determining a type of tag to be output; the prediction unit is used for selecting a probability prediction model from a model library if the label type is a probability label, taking the intermediate sample data as input data of the probability prediction model, and outputting first target label data from the probability prediction model locally in the CRM system; and if the label type is a preference label, selecting a commodity recommendation model from a model library, taking the intermediate sample data as input data of the commodity recommendation model, and outputting second target label data from the commodity recommendation model locally by the CRM system, wherein the model library comprises the probability prediction model and the commodity recommendation model.
Further, the return module includes one of: the first return unit is used for responding to a first tag call request of a service end, carrying out variable-length character coding on the target tag data, generating json data, and returning the json data to the service end by adopting an Application Programming Interface (API) interface; and the second return unit is used for responding to a second tag call request of the service end, carrying out variable-length character coding on the target tag data, generating file data, and returning the file data to the service end by adopting a file call interface.
Further, the apparatus further comprises: the system comprises an acquisition module, an initialization module and a service processing module, wherein the acquisition module is used for acquiring original sample data from a customer relationship management CRM system, and the service processing module is used for initializing service processor classes of a plurality of scheduling containers, wherein the scheduling containers comprise a plurality of container spaces, each container space corresponds to one service processor class, each service processor class corresponds to one label type, and each service processor class comprises a plurality of code data connected by a responsibility chain; the first analysis module is used for acquiring service demand information after initialization is completed and analyzing a processor list and code environment parameters of the scheduling container according to the service demand information; the second analysis module is used for analyzing the class parameters of the processor list if the processor list and the code environment parameters are successfully analyzed; and the starting module is used for starting the scheduling container after the class parameter analysis is successful.
Further, the initialization module includes: the extraction unit is used for comparing the scheduling minimization actions and extracting an abstract template of the service processor according to a comparison result to obtain a virtual class of the service processor; the generating unit is used for generating service processor classes by adopting the service processor virtual classes; and the initialization unit is used for calling the main program of the scheduling container to initialize the service processor classes of all the service processors.
Further, the apparatus further comprises: the determining module is used for determining a target scheduling container which is currently executed before the initializing module initializes the service processor classes of a plurality of scheduling containers; a third parsing module, configured to parse the container token of the target scheduling container; and the positioning module is used for positioning the calling time sequence of the target scheduling container in a first-in first-out FIFO pipeline by adopting the container token.
According to another aspect of the embodiments of the present application, there is also provided a storage medium including a stored program that performs the above steps when running.
According to another aspect of the embodiment of the present application, there is also provided an electronic device including a processor, a communication interface, a memory, and a communication bus, where the processor, the communication interface, and the memory complete communication with each other through the communication bus; wherein: a memory for storing a computer program; and a processor for executing the steps of the method by running a program stored on the memory.
Embodiments of the present application also provide a computer program product comprising instructions which, when run on a computer, cause the computer to perform the steps of the above method.
According to the application, original sample data is read from a customer relationship management CRM system, data cleaning and feature extraction are carried out on the original sample data, intermediate sample data meeting preset requirements is output, the intermediate sample data is input into a pre-trained label prediction model, target label data is output on the CRM system, the target label data is returned to the service end in response to a label calling request of the service end, and dynamic labels required by the service end are output under the condition that sensitive sample data such as transaction data cannot be out of a domain, so that the technical problem that dynamic labels cannot be generated in real time in the related art is solved, and the real-time requirement of the service end on the label data is met.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:
FIG. 1 is a block diagram of the hardware architecture of a server according to an embodiment of the present application;
FIG. 2 is a flow chart of a method of outputting a dynamic tag according to an embodiment of the present application;
FIG. 3 is a schematic block diagram of an embodiment of the present application;
FIG. 4 is a block diagram of a dynamic tag output apparatus according to an embodiment of the present application;
fig. 5 is a block diagram of an electronic device embodying an embodiment of the present application.
Detailed Description
In order that those skilled in the art will better understand the present application, a technical solution in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present application without making any inventive effort, shall fall within the scope of the present application. It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the application described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
The method according to the first embodiment of the present application may be implemented in a server, a computer, a mobile phone, or a similar computing device. Taking the operation on a server as an example, fig. 1 is a block diagram of a hardware structure of a server according to an embodiment of the present application. As shown in fig. 1, the server may include one or more (only one is shown in fig. 1) processors 102 (the processor 102 may include, but is not limited to, a microprocessor MCU or a processing device such as a programmable logic device FPGA) and a memory 104 for storing data, and optionally, a transmission device 106 for communication functions and an input-output device 108. It will be appreciated by those of ordinary skill in the art that the architecture shown in fig. 1 is merely illustrative and is not intended to limit the architecture of the server described above. For example, the server may also include more or fewer components than shown in FIG. 1, or have a different configuration than shown in FIG. 1.
The memory 104 may be used to store a server program, for example, a software program of application software and a module, such as a server program corresponding to a method for outputting a dynamic tag in an embodiment of the present application, and the processor 102 executes the server program stored in the memory 104, thereby performing various functional applications and data processing, that is, implementing the method described above. Memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory remotely located with respect to the processor 102, which may be connected to a server via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The transmission device 106 is used to receive or transmit data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of a server. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, simply referred to as NIC) that can connect to other network devices through a base station to communicate with the internet. In one example, the transmission device 106 may be a Radio Frequency (RF) module, which is configured to communicate with the internet wirelessly.
In this embodiment, a method for outputting a dynamic tag is provided, and fig. 2 is a flowchart of a method for outputting a dynamic tag according to an embodiment of the present invention, as shown in fig. 2, where the flowchart includes the following steps:
step S202, original sample data is read from a customer relationship management CRM system;
the CRM system in this embodiment is operated on an e-commerce platform (such as the ali, the jingdong, etc.), and the original sample data may be transaction data, sales data, order data, member data, commodity data, etc. of the consumer of the target merchant on the e-commerce platform, and the CRM system may be an independent CRM server or may be integrated on a central server of the e-commerce platform.
Step S204, data cleaning and feature extraction are carried out on the original sample data, and intermediate sample data meeting preset requirements is output;
the intermediate sample data in this embodiment refers to sample data meeting the requirements of the label prediction model, and the preset requirements include format requirements, data volume requirements, data quality requirements, and the like.
Step S206, inputting the intermediate sample data into a pre-trained label prediction model, and outputting target label data on the CRM system;
the tag prediction model of this embodiment is configured on the server where the CRM system is located.
Step S208, the target label data is returned to the service end in response to the label calling request of the service end.
The business end of the embodiment can be a merchant end needing to use the target tag data, and can also be other clients needing to call the tag data, and the target tag data does not relate to the privacy information of the user, so that the target tag data can be quickly and dynamically exported, and the tag data synchronization is realized.
Through the steps, original sample data are read from a customer relationship management CRM system, data cleaning and feature extraction are carried out on the original sample data, middle sample data meeting preset requirements are output, the middle sample data are input into a pre-trained label prediction model, target label data are output on the CRM system, the target label data are returned to the service end in response to a label calling request of the service end, and dynamic labels required by the service end are output under the condition that sensitive sample data such as transaction data cannot be out of a domain, so that the technical problem that dynamic labels cannot be generated in real time in related technologies is solved, and the real-time requirement of the service end on the label data is met.
In one implementation manner of the embodiment, performing data cleaning and feature extraction on the original sample data, and outputting intermediate sample data meeting preset requirements includes: performing deduplication cleaning and format content cleaning on the original sample data to obtain first sample data; selecting and extracting a sample attribution identifier and sample characteristics from the first sample data; and constructing a feature table by taking the sample attribution identifier as a key and the corresponding sample feature as a value, and determining the feature table as intermediate sample data.
Alternatively, the sample home identifier may be a user identifier, a commodity identifier, an item identifier, etc., such as a user_id, a item_id, etc., and in one example, the feature table is in the format { user_id/item_id, feature_1, feature_2, … }, feature_1, feature_2, … is a sequence of sample features.
In one example of this embodiment, inputting intermediate sample data into a pre-trained tag prediction model, outputting target tag data on the CRM system includes: determining the type of the label to be output; if the tag type is a probability tag, selecting a probability prediction model from a model library, taking the intermediate sample data as input data of the probability prediction model, and outputting first target tag data from the probability prediction model locally in the CRM system; and if the label type is a preference label, selecting a commodity recommendation model from a model library, taking the intermediate sample data as input data of the commodity recommendation model, and outputting second target label data from the commodity recommendation model locally by the CRM system, wherein the model library comprises a probability prediction model and a commodity recommendation model.
According to the service requirement, in the data modeling module, a corresponding model is selected according to the required label type for modeling, for example, selecting a 'purchase probability label' can call a purchase probability prediction model, or a 'commodity preference label' can call a commodity recommendation model and the like.
In some examples, after outputting the target tag data on the CRM system, post-processing may also be performed on the target tag data, and discretizing the continuous value predicted by the tag, for example, discretizing the purchase probability (value between 0 and 1) into three purchase probabilities (high, medium and low); and storing the label result in a database in the form of { user_id: label value } for subsequent call of a label output module.
In the implementation scenario of this embodiment, in response to the tag call request of the service end, the returning of the target tag data to the service end may be, but is not limited to,:
mode one: responding to a first label calling request of a service end, performing variable length character coding on target label data to generate json data, and returning the json data to the service end by adopting an Application Programming Interface (API) interface;
mode two: and responding to a second label calling request of the service end, performing variable-length character coding on target label data to generate file data, and returning the file data to the service end by adopting a file calling interface.
The scheme of this embodiment supports two ways of label output: the method (1) is used for consuming a label library of a merchant own system in an API (application program interface) mode, the coding mode adopts utf-8 coding, and the data format is json; the mode (2) is used for a merchant user to apply in a form of downloading a csv file, the coding mode adopts utf-8 coding, and the data format is the file.
In one implementation of the present embodiment, before reading the raw sample data from the customer relationship management CRM system, further comprising:
s11, initializing service processor classes of a plurality of scheduling containers, wherein the scheduling containers comprise a plurality of container spaces, each container space corresponds to one service processor class, each service processor class corresponds to one tag type, and each service processor class comprises a plurality of code data connected by a responsibility chain;
in some examples, initializing the traffic processor class of the scheduling container includes: comparing the scheduling minimized actions, and extracting an abstract template of the service processor according to a comparison result to obtain a virtual class of the service processor; generating service processor classes by adopting service processor virtual classes; the main program of the scheduling container is called to initialize the service processor classes of all the service processors.
Optionally, in this embodiment, FIFO pipes are used to manage the call timing of multiple scheduling containers, so as to ensure ordered call of hardware resources and prevent overflow of hardware overhead. Before initializing the service processor class of the plurality of scheduling containers, further comprising: determining a target scheduling container currently executed; analyzing a container token of the target scheduling container; the container token is used to locate the call timing of the target dispatch container in a first in first out FIFO pipe.
The embodiment, when configuring the execution queue based on the FIFO pipe, includes the following steps:
step 1, configuring FIFO (First Input First Output, first-in first-out) queue enqueuing;
step 2, configuring basic attributes of tokens (including but not limited to storage format, encryption mode and the like), wherein each token is used for identifying each object in the FIFO pipeline, namely a scheduling container;
step 3, configuring a queue execution environment (including but not limited to pipeline files, logs, token paths, timeout thresholds, software and hardware versions, etc.);
step 4, setting queue initialization parameters (including, but not limited to, occupation of hardware resources (cpu, memory), operating system version, number of queue initialization tokens, token encryption mode, etc.);
step 5, setting a queue service command list (text file), and initializing a queue (operating system process);
Step 6, starting a queue (a command line background running mode);
and 7, re-executing the steps 4-6 after modifying the service command list according to the requirement.
S12, after initialization is completed, acquiring service demand information, and analyzing a processor list and code environment parameters of a scheduling container according to the service demand information;
s13, if the processor list and the code environment parameters are successfully analyzed, analyzing class parameters of the processor list;
s14, after the class parameter analysis is successful, starting a scheduling container.
In some scenarios of this embodiment, a scheduling container (scheduling framework) is used to control the output flow of the dynamic labels, where the scheduling container includes multiple service processor classes, each service processor class corresponding to a label type.
In one example, the configuration flow of the scheduling container includes:
step 1, comparing the scheduling minimization actions, and extracting a processor abstract template to form a virtual class (such as a base class in a Python code);
step 2, realizing processor class (abstract template subclass) corresponding to a plurality of services;
step 3, initializing all service processor classes by a main program;
step 4, the main program starts to analyze the command line parameters, and the first stage analyzes the public parameters such as the list of the service processor and the environment required at the time; analyzing service processor class parameters in the second stage; loading into a parameter table;
Step 5, processing the first-stage error, and if the error occurs, not entering the second stage; after entering the second stage, starting to execute the processor core scheduling method (serial execution and concurrent execution of the dependent execution queue);
and 6, executing the steps 1-5 again after adding and deleting the service processor class.
The invention provides a label prediction method and a label prediction system for a consumer sensitive data not outputting an e-commerce platform domain, wherein data modeling is performed on a CRM system of the e-commerce platform, dynamic labels required by digital marketing of merchants are output, and the requirement of business instantaneity is met.
Fig. 3 is a schematic flow chart of a framework of an embodiment of the present invention, which includes a data loading module, a data modeling module, a tag output module, and a master scheduling module, and the functions of the respective modules are described below:
and the data loading module is used for loading the original data required by modeling from a Customer Relationship Management (CRM) system deployed in the platform domain, and performing data cleaning and feature engineering according to the modeling requirement so as to output feature data meeting the model requirement.
And the data modeling module is used for selecting a proper model according to the label requirements of the merchant to perform model training and model prediction, and outputting a label result to the label output module.
And the label output module is used for outputting labels in two modes, wherein the mode (1) is used for consuming a label library of a merchant own system in an API (application program interface) and the mode (2) is used for being applied by a merchant user in a form of downloading a csv file.
And the main scheduling module provides overall flow scheduling and performs flow management in a timing mode and a random mode. The module consists of two parts: (1) the high-performance and scalable scheduling framework based on the responsibility chain mode supports diversity modeling (including feature engineering). (2) A controlled (finite token, timing) multithreaded, highly concurrent execution queue based on FIFOs pipeline communications.
The data loading module comprises the following steps when executing data loading:
step one: the data loading module loads the original data required by modeling from a Customer Relationship Management (CRM) system deployed in a platform domain, wherein the original data mainly comprises order data, member data and commodity data;
step two: the data loading module performs data cleaning on the original data, performs feature engineering according to the requirement of a model algorithm in 3.2.2, and stores the features after feature processing in a database in the form of a feature table for consumption by the model, wherein the format of the feature table is { user_id/item_id, feature_1, feature_2, … }.
The data modeling module, when performing data modeling, includes the steps of:
step one: according to business requirements, in a data modeling module, a merchant user selects a corresponding model to model according to the type of a required label, for example, selecting a purchase probability label can call a purchase probability prediction model, or a commodity preference label can call a commodity recommendation model;
step two: training a model;
step three: predicting labels;
step four: discretizing the continuous value predicted by the label, for example, discretizing the purchase probability (value between 0 and 1) into three purchase probabilities (high, medium and low);
step five: the label result is stored in a database in the form of { user_id: label value } for being called by a 3.2.3 label output module;
the label output module comprises the following steps when outputting labels:
step one: setting label calculation and output task scheduling through a main scheduling module;
step two: tag output is performed in two ways: the method (1) is used for consuming a label library of a merchant own system in an API (application program interface) mode, the coding mode adopts utf-8 coding, and the data format is json; the mode (2) is used for a merchant user to apply in a form of downloading a csv file, the coding mode adopts utf-8 coding, and the data format is the file.
The scheme of the embodiment overcomes the difficulty that the sensitive data needed by modeling can be acquired to the local and then the modeling prediction is carried out after the data domain-out pre-application is required to be provided for the e-commerce platform and a plurality of workday audits are waited for passing; the dynamic label required by the marketing of the merchant can be output under the condition that sensitive data such as transaction data and the like cannot go out of the domain, and the requirement of business instantaneity is met. The dynamic label can be called by an intelligent operation tool and can be independently output to business personnel for use; the output can be calculated at regular time through the main dispatching module, and also can be calculated according to the requirement. The data modeling can be performed under the condition that transaction data does not go out of an electronic commerce domain; the method can meet the real-time requirement of business of merchants.
From the description of the above embodiments, it will be clear to a person skilled in the art that the method according to the above embodiments may be implemented by means of software plus the necessary general hardware platform, but of course also by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method according to the embodiments of the present invention.
Example 2
The embodiment also provides an output device of the dynamic tag, which is used for implementing the above embodiment and the preferred implementation, and is not described in detail. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. While the means described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.
Fig. 4 is a block diagram of a dynamic tag output apparatus according to an embodiment of the present invention, as shown in fig. 4, including: an acquisition module 40, a processing module 42, a prediction module 44, a return module 46, wherein,
an acquisition module 40 for reading raw sample data from the customer relationship management CRM system;
the processing module 42 is configured to perform data cleaning and feature extraction on the raw sample data, and output intermediate sample data that meets a preset requirement;
a prediction module 44 for inputting the intermediate sample data into a pre-trained tag prediction model, outputting target tag data on the CRM system;
and the return module 46 is configured to return the target tag data to the service end in response to a tag call request of the service end.
Optionally, the processing module includes: the processing unit is used for carrying out deduplication cleaning and format content cleaning on the original sample data to obtain first sample data; an extraction unit, configured to select and extract a sample home identifier and a sample feature from the first sample data; and the storage unit is used for constructing a feature table by taking the sample attribution identifier as a key and the corresponding sample feature as a value, and determining the feature table as intermediate sample data.
Optionally, the prediction module includes: a determining unit for determining a type of tag to be output; the prediction unit is used for selecting a probability prediction model from a model library if the label type is a probability label, taking the intermediate sample data as input data of the probability prediction model, and outputting first target label data from the probability prediction model locally in the CRM system; and if the label type is a preference label, selecting a commodity recommendation model from a model library, taking the intermediate sample data as input data of the commodity recommendation model, and outputting second target label data from the commodity recommendation model locally by the CRM system, wherein the model library comprises the probability prediction model and the commodity recommendation model.
Optionally, the return module includes one of: the first return unit is used for responding to a first tag call request of a service end, carrying out variable-length character coding on the target tag data, generating json data, and returning the json data to the service end by adopting an Application Programming Interface (API) interface; and the second return unit is used for responding to a second tag call request of the service end, carrying out variable-length character coding on the target tag data, generating file data, and returning the file data to the service end by adopting a file call interface.
Optionally, the apparatus further includes: the system comprises an acquisition module, an initialization module and a service processing module, wherein the acquisition module is used for acquiring original sample data from a customer relationship management CRM system, and the service processing module is used for initializing service processor classes of a plurality of scheduling containers, wherein the scheduling containers comprise a plurality of container spaces, each container space corresponds to one service processor class, each service processor class corresponds to one label type, and each service processor class comprises a plurality of code data connected by a responsibility chain; the first analysis module is used for acquiring service demand information after initialization is completed and analyzing a processor list and code environment parameters of the scheduling container according to the service demand information; the second analysis module is used for analyzing the class parameters of the processor list if the processor list and the code environment parameters are successfully analyzed; and the starting module is used for starting the scheduling container after the class parameter analysis is successful.
Optionally, the initialization module includes: the extraction unit is used for comparing the scheduling minimization actions and extracting an abstract template of the service processor according to a comparison result to obtain a virtual class of the service processor; the generating unit is used for generating service processor classes by adopting the service processor virtual classes; and the initialization unit is used for calling the main program of the scheduling container to initialize the service processor classes of all the service processors.
Optionally, the apparatus further includes: the determining module is used for determining a target scheduling container which is currently executed before the initializing module initializes the service processor classes of a plurality of scheduling containers; a third parsing module, configured to parse the container token of the target scheduling container; and the positioning module is used for positioning the calling time sequence of the target scheduling container in a first-in first-out FIFO pipeline by adopting the container token.
It should be noted that each of the above modules may be implemented by software or hardware, and for the latter, it may be implemented by, but not limited to: the modules are all located in the same processor; alternatively, the above modules may be located in different processors in any combination.
Example 3
An embodiment of the invention also provides a storage medium having a computer program stored therein, wherein the computer program is arranged to perform the steps of any of the method embodiments described above when run.
Alternatively, in the present embodiment, the above-described storage medium may be configured to store a computer program for performing the steps of:
s1, original sample data is read from a Customer Relationship Management (CRM) system;
s2, carrying out data cleaning and feature extraction on the original sample data, and outputting intermediate sample data meeting preset requirements;
s3, inputting the intermediate sample data into a pre-trained label prediction model, and outputting target label data on the CRM system;
s4, responding to a label calling request of the service end, and returning the target label data to the service end.
Alternatively, in the present embodiment, the storage medium may include, but is not limited to: a usb disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing a computer program.
An embodiment of the invention also provides an electronic device comprising a memory having stored therein a computer program and a processor arranged to run the computer program to perform the steps of any of the method embodiments described above.
Optionally, the electronic device may further include a transmission device and an input/output device, where the transmission device is connected to the processor, and the input/output device is connected to the processor.
Alternatively, in the present embodiment, the above-described processor may be configured to execute the following steps by a computer program:
s1, original sample data is read from a Customer Relationship Management (CRM) system;
s2, carrying out data cleaning and feature extraction on the original sample data, and outputting intermediate sample data meeting preset requirements;
s3, inputting the intermediate sample data into a pre-trained label prediction model, and outputting target label data on the CRM system;
s4, responding to a label calling request of the service end, and returning the target label data to the service end.
Alternatively, specific examples in this embodiment may refer to examples described in the foregoing embodiments and optional implementations, and this embodiment is not described herein.
Fig. 5 is a block diagram of an electronic device according to an embodiment of the present invention, as shown in fig. 5, including a processor 501, a communication interface 502, a memory 503, and a communication bus 504, where the processor 501, the communication interface 502, and the memory 503 perform communication with each other through the communication bus 504, and the memory 503 is used to store a computer program; a processor 501 for executing programs stored on a memory 503.
The foregoing embodiment numbers of the present application are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
In the foregoing embodiments of the present application, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed technology may be implemented in other manners. The above-described embodiments of the apparatus are merely exemplary, and the division of the units, such as the division of the units, is merely a logical function division, and may be implemented in another manner, for example, multiple units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing is merely a preferred embodiment of the present application and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present application, which are intended to be comprehended within the scope of the present application.
Claims (9)
1. A method for outputting a dynamic tag, comprising:
reading raw sample data from a customer relationship management CRM system;
performing data cleaning and feature extraction on the original sample data, and outputting intermediate sample data meeting preset requirements;
inputting the intermediate sample data into a pre-trained tag prediction model, outputting target tag data on the CRM system;
responding to a label calling request of a service end, and returning the target label data to the service end;
wherein prior to reading the raw sample data from the customer relationship management CRM system, the method further comprises: initializing service processor classes of a plurality of scheduling containers, wherein the scheduling containers comprise a plurality of container spaces, each container space corresponds to one service processor class, each service processor class corresponds to one tag type, and each service processor class comprises a plurality of code data connected by a responsibility chain; after initialization is completed, acquiring service demand information, and analyzing a processor list and code environment parameters of the scheduling container according to the service demand information; if the processor list and the code environment parameters are successfully analyzed, analyzing class parameters of the processor list; and after the class parameters are successfully parsed, starting the scheduling container.
2. The method of claim 1, wherein performing data cleaning and feature extraction on the raw sample data, and outputting intermediate sample data meeting preset requirements comprises:
performing deduplication cleaning and format content cleaning on the original sample data to obtain first sample data;
selecting and extracting a sample attribution identifier and a sample characteristic from the first sample data;
and constructing a feature table by taking the sample attribution identifier as a key and the corresponding sample feature as a value, and determining the feature table as intermediate sample data.
3. The method of claim 1, wherein inputting the intermediate sample data into a pre-trained tag prediction model, outputting target tag data on the CRM system comprises:
determining the type of the label to be output;
if the tag type is a probability tag, selecting a probability prediction model from a model library, taking the intermediate sample data as input data of the probability prediction model, and outputting first target tag data from the probability prediction model locally by the CRM system; and if the label type is a preference label, selecting a commodity recommendation model from a model library, taking the intermediate sample data as input data of the commodity recommendation model, and outputting second target label data from the commodity recommendation model locally by the CRM system, wherein the model library comprises the probability prediction model and the commodity recommendation model.
4. The method of claim 1, wherein returning the target tag data to the service end in response to a tag call request from the service end comprises one of:
responding to a first label calling request of a service end, performing variable length character coding on target label data to generate json data, and returning the json data to the service end by adopting an Application Programming Interface (API) interface;
and responding to a second tag call request of the service end, performing variable length character coding on the target tag data to generate file data, and returning the file data to the service end by adopting a file call interface.
5. The method of claim 1, wherein initializing a traffic processor class of a scheduling container comprises:
comparing the scheduling minimized actions, and extracting an abstract template of the service processor according to a comparison result to obtain a virtual class of the service processor;
generating a service processor class by adopting the service processor virtual class;
and calling a main program of the scheduling container to initialize service processor classes of all service processors.
6. The method of claim 1, wherein prior to initializing the traffic processor class for the number of scheduling containers, the method further comprises:
Determining a target scheduling container currently executed;
parsing the container token of the target scheduling container;
and positioning the calling time sequence of the target scheduling container in a first-in first-out FIFO pipeline by adopting the container token.
7. An output device for a dynamic tag, comprising:
the acquisition module is used for reading the original sample data from the Customer Relationship Management (CRM) system;
the processing module is used for carrying out data cleaning and feature extraction on the original sample data and outputting intermediate sample data meeting preset requirements;
the prediction module is used for inputting the intermediate sample data into a pre-trained label prediction model and outputting target label data on the CRM system;
the return module is used for responding to a label calling request of a service end and returning the target label data to the service end;
wherein the apparatus further comprises: the system comprises an acquisition module, an initialization module and a service processing module, wherein the acquisition module is used for acquiring original sample data from a customer relationship management CRM system, and the service processing module is used for initializing service processor classes of a plurality of scheduling containers, wherein the scheduling containers comprise a plurality of container spaces, each container space corresponds to one service processor class, each service processor class corresponds to one label type, and each service processor class comprises a plurality of code data connected by a responsibility chain; the first analysis module is used for acquiring service demand information after initialization is completed and analyzing a processor list and code environment parameters of the scheduling container according to the service demand information; the second analysis module is used for analyzing the class parameters of the processor list if the processor list and the code environment parameters are successfully analyzed; and the starting module is used for starting the scheduling container after the class parameter analysis is successful.
8. A storage medium comprising a stored program, wherein the program when run performs the steps of the method of any of the preceding claims 1 to 6.
9. An electronic device comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory are communicated with each other through the communication bus; wherein:
a memory for storing a computer program;
a processor for executing the steps of the method of any one of claims 1 to 6 by running a program stored on a memory.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211536454.1A CN115934675B (en) | 2022-12-01 | 2022-12-01 | Dynamic tag output method and device, storage medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211536454.1A CN115934675B (en) | 2022-12-01 | 2022-12-01 | Dynamic tag output method and device, storage medium and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115934675A CN115934675A (en) | 2023-04-07 |
| CN115934675B true CN115934675B (en) | 2023-10-03 |
Family
ID=86653768
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211536454.1A Active CN115934675B (en) | 2022-12-01 | 2022-12-01 | Dynamic tag output method and device, storage medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115934675B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108764663A (en) * | 2018-05-15 | 2018-11-06 | 广东电网有限责任公司信息中心 | A kind of power customer portrait generates the method and system of management |
| CN115344757A (en) * | 2022-02-07 | 2022-11-15 | 花瓣云科技有限公司 | Label prediction method, electronic equipment and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112988727B (en) * | 2021-03-25 | 2022-09-16 | 北京百度网讯科技有限公司 | Data annotation method, device, equipment, storage medium and computer program product |
-
2022
- 2022-12-01 CN CN202211536454.1A patent/CN115934675B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108764663A (en) * | 2018-05-15 | 2018-11-06 | 广东电网有限责任公司信息中心 | A kind of power customer portrait generates the method and system of management |
| CN115344757A (en) * | 2022-02-07 | 2022-11-15 | 花瓣云科技有限公司 | Label prediction method, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115934675A (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10142106B2 (en) | System and method for securing sensitive data | |
| CN107436844B (en) | Method and device for generating interface use case aggregate | |
| CN112527528A (en) | Data transmission method, device and storage medium based on message queue | |
| CN110445860B (en) | Message sending method, device, terminal equipment and storage medium | |
| CN109815256A (en) | A kind of data processing method, device, electronic equipment and storage medium | |
| CN107273979B (en) | Method and system for performing machine learning prediction based on service level | |
| CN111651140B (en) | Service method and device based on workflow | |
| CN111367690A (en) | Method and device for supporting combined API service calling | |
| CN113592593B (en) | Training and application method, device, equipment and storage medium of sequence recommendation model | |
| CN112749323A (en) | Method and device for constructing user portrait | |
| CN106033551A (en) | Data parsing method, apparatus and system thereof | |
| CN115934675B (en) | Dynamic tag output method and device, storage medium and electronic equipment | |
| CN115328457A (en) | Method and device for realizing form page based on parameter configuration | |
| CN114363268B (en) | Message delay consumption method and related product | |
| CN115495658A (en) | Data processing method and device | |
| US20220214864A1 (en) | Efficient deployment of machine learning and deep learning model's pipeline for serving service level agreement | |
| CN112799797B (en) | Task management method and device | |
| CN113973139A (en) | Message processing method and device | |
| CN113112326A (en) | User identification method, method for displaying data to user and related device | |
| CN112965767A (en) | Job processing method and apparatus, electronic device, and storage medium | |
| CN111601291A (en) | RFID middleware, publish-subscribe system and data transmission method | |
| CN107368597B (en) | Information output method and device | |
| US10580021B2 (en) | Product offering analytics | |
| CN110750563A (en) | Multi-model data processing method, system, device, electronic equipment and storage medium | |
| CN116501365B (en) | Resource calling method, device and equipment based on algorithm platform |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |