CN117670399A - Customer marketing grade evaluation method, device, electronic equipment and storage medium - Google Patents
Customer marketing grade evaluation method, device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN117670399A CN117670399A CN202311666513.1A CN202311666513A CN117670399A CN 117670399 A CN117670399 A CN 117670399A CN 202311666513 A CN202311666513 A CN 202311666513A CN 117670399 A CN117670399 A CN 117670399A
- Authority
- CN
- China
- Prior art keywords
- marketing
- encryption
- characteristic information
- customer
- grade
- 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
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 34
- 238000000034 method Methods 0.000 claims abstract description 55
- 238000005516 engineering process Methods 0.000 claims abstract description 18
- 238000013210 evaluation model Methods 0.000 claims abstract description 13
- 238000004590 computer program Methods 0.000 claims description 16
- 238000012549 training Methods 0.000 claims description 15
- 238000013528 artificial neural network Methods 0.000 claims description 8
- 238000004422 calculation algorithm Methods 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 7
- 238000010801 machine learning Methods 0.000 claims description 7
- 239000013598 vector Substances 0.000 description 25
- 238000012545 processing Methods 0.000 description 16
- 238000010586 diagram Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 11
- 238000004891 communication Methods 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 230000007246 mechanism Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000002457 bidirectional effect Effects 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
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 client marketing grade assessment method, a client marketing grade assessment device, electronic equipment and a storage medium. The method comprises the following steps: acquiring historical transaction data of a customer to be evaluated, and extracting target characteristic information in the historical transaction data; encrypting the target characteristic information based on the homomorphic encryption technology to generate encrypted characteristic information; analyzing the encryption characteristic information through a pre-trained client marketing grade evaluation model, and determining an encryption marketing grade evaluation result corresponding to the encryption characteristic information; and decrypting the encryption marketing grade assessment result to generate the marketing grade of the customer to be assessed. Through this scheme, not only improved the discernment precision of bank to potential customer marketing rating greatly, solved the problem that data can not trust each other between enterprise and the bank through the mode of encryption moreover, both reduced the risk that enterprise data revealed, improved the efficiency that the bank obtained customer marketing grade again.
Description
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to a method and apparatus for evaluating a client marketing level, an electronic device, and a storage medium.
Background
Ranking refers to ordering customers by their value and potential benefit. The high-value customers refer to customers capable of bringing more profits and profits to the bank, and the customers with high potential profits refer to customers who may become good customers of the bank in the future. The bank needs to order the marketing customers in a rank order so that the marketing can be done with priority. The bank may evaluate and rank the customer based on factors such as the customer's size, industry, financial status, historical transaction records, etc., to determine the customer's value and potential revenue. However, the existing customer marketing evaluation method is usually implemented through manual analysis and experience judgment.
Specifically, a bank staff can evaluate the value of the client according to the basic information, the historical transaction record, the credit rating and other factors of the client, and formulate a corresponding marketing strategy based on the evaluation result. However, the traditional method has the problems of strong subjectivity, low analysis efficiency, inaccurate evaluation result and the like; and privacy problems caused by data leakage exist, and banks have barriers in acquiring relevant information of potential clients and cannot adapt to increasingly complex and diversified market environments and client requirements.
Disclosure of Invention
The invention provides a client marketing grade assessment method, a client marketing grade assessment device, electronic equipment and a storage medium, which can accurately determine a bank client marketing grade.
According to an aspect of the present invention, there is provided a customer marketing rating assessment method, comprising:
acquiring historical transaction data of a customer to be evaluated, and extracting target characteristic information in the historical transaction data;
encrypting the target characteristic information based on the homomorphic encryption technology to generate encrypted characteristic information;
analyzing the encryption characteristic information through a pre-trained client marketing grade evaluation model, and determining an encryption marketing grade evaluation result corresponding to the encryption characteristic information;
and decrypting the encryption marketing grade assessment result to generate the marketing grade of the customer to be assessed.
According to another aspect of the present invention, there is provided a customer marketing rating assessment device comprising:
the target feature information extraction module is used for acquiring historical transaction data of the clients to be evaluated and extracting target feature information in the historical transaction data;
the encryption characteristic information generation module is used for encrypting the target characteristic information based on the homomorphic encryption technology to generate encryption characteristic information;
the encryption marketing grade evaluation result determining module is used for analyzing the encryption characteristic information through a pre-trained client marketing grade evaluation model and determining an encryption marketing grade evaluation result corresponding to the encryption characteristic information;
and the marketing grade generating module is used for decrypting the encryption marketing grade evaluation result to generate the marketing grade of the client to be evaluated.
According to another aspect of the present invention, there is provided an electronic apparatus including:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the customer marketing rating assessment method of any of the embodiments of the present invention.
According to another aspect of the present invention, there is provided a computer readable storage medium storing computer instructions for causing a processor to implement the customer marketing rating assessment method of any of the embodiments of the present invention when executed.
The client marketing grade assessment scheme of the embodiment of the invention comprises the following steps: acquiring historical transaction data of a customer to be evaluated, and extracting target characteristic information in the historical transaction data; encrypting the target characteristic information based on the homomorphic encryption technology to generate encrypted characteristic information; analyzing the encryption characteristic information through a pre-trained client marketing grade evaluation model, and determining an encryption marketing grade evaluation result corresponding to the encryption characteristic information; and decrypting the encryption marketing grade assessment result to generate the marketing grade of the customer to be assessed. By the technical scheme provided by the embodiment of the invention, the identification accuracy of the bank to the marketing rating of the potential customer is greatly improved, the problem that data between enterprises and banks cannot be mutually trusted is solved by an encryption mode, the risk of enterprise data leakage is reduced, and the efficiency of the bank for acquiring the marketing rating of the customer is improved.
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the invention or to delineate the scope of the invention. Other features of the present invention will become apparent from the description that follows.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for customer marketing rating assessment provided in accordance with a first embodiment of the present invention;
fig. 2 is a schematic structural diagram of a Tranformer in a Bert model according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a process for obtaining a customer marketing rating according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a process for obtaining customer marketing levels for a plurality of participants provided by an embodiment of the invention;
FIG. 5 is a schematic diagram of a client marketing rating system according to a second embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device implementing a client marketing rating assessment method according to an embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention 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 invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention 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 invention 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, system, 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
Fig. 1 is a flowchart of a method for evaluating a customer marketing grade according to an embodiment of the present invention, which is applicable to evaluating a bank customer marketing grade, the method may be performed by a customer marketing grade evaluation device, the customer marketing grade evaluation device may be implemented in hardware and/or software, and the customer marketing grade evaluation device may be configured in an electronic device. As shown in fig. 1, the method includes:
s110, acquiring historical transaction data of a customer to be evaluated, and extracting target characteristic information in the historical transaction data.
Where the customer under evaluation may be understood as a business (also referred to as a participant). In the embodiment of the invention, the historical transaction data of the clients to be evaluated are obtained, the historical transaction data are analyzed, and the target characteristic information in the historical transaction data is extracted. For example, target feature information in historical transaction data may be extracted based on a pre-trained feature information extraction model.
Optionally, extracting the target feature information in the historical transaction data includes: inputting the historical transaction data into a Bert model, and acquiring output initial characteristic information of the Bert model; and inputting the initial characteristic information into a Bi-GRU network model, and acquiring target characteristic information output by the Bi-GRU network model.
The Bert model is trained by a large-scale unmarked corpus. The Bert model can be directly applied to training of various text classification and matching tasks, and can also be used as a text feature extraction mode. The Bi-GRU (Bidirectional Gated Recurrent Unit) model is a deep learning model for processing sequence data. Compared with the traditional RNN model, the Bi-GRU can simultaneously consider historical and future information when processing the sequence data, so that the characteristics of the sequence data are better extracted.
In the embodiment of the invention, the historical transaction data are input into the Bert model, so that the initial characteristic information in the historical transaction data is extracted through the Bert model, then the initial characteristic information is input into the Bi-GRU network model, and the target characteristic information is obtained through the Bi-GRU network model. It can be understood that in the marketing of bank clients, historical transaction data of the clients can be analyzed and predicted based on the Bert model and the Bi-GRU network model, so that the consumption behaviors and consumption trends of the clients can be better known, and corresponding marketing strategies can be formulated.
It will be appreciated that the Bert model is utilized as a wordAnd the embedding layer acquires rich text semantic vector representations (namely initial characteristic information) in the historical transaction data, and then inputs the initial characteristic information into the Bi-GRU network for secondary embedding representation. Because of the strong feature extraction function of the Bert model, the Bert model is directly used for extracting the text vector of the client information. The input representation function of Bert is also more powerful than the traditional embedded representation method, and not only comprises word vectors of the text, but also segment vectors and position vectors in the text where words are located. E for word vector of text token Representing segment vectors with E seg Representing the position vector with E pos Representing, the Bert model finally sums three vector representations to obtain semantic representation characteristics E= { E of the text token +E seg +E pos }. After the input representation feature is obtained, the Bert model is mainly extracted layer by layer through the stacking of the internal bidirectional transducer encoders to obtain the final vectorization representation of the text, and as shown in formula (1), l represents the corresponding layer of transducer encoder in the Bert model, E l The corresponding feature extraction vector is represented:
E l ＝Transformer l (E l-1 ) (1)
the Tranformer encoder is composed of a plurality of overlapped units, and each unit is composed of a Self-Attention mechanism and a feedforward neural network. When the input vector enters a Tranformer unit, firstly, the vector passes through a Self-Attention layer, then residual error and standardization processing are carried out, then the output of the Self-Attention layer is input into a feedforward neural network layer for processing, and finally the same residual error and standardization processing are carried out. Fig. 2 is a schematic structural diagram of Tranformer in the Bert model according to an embodiment of the present invention. The Self-Attention layer processing mainly comprises the following three steps:
(1) The Query, key, and value vector representations are obtained. The attention mechanism represents the target word in Query and the individual words of the context in Key. The self-attention mechanism takes Query and Key as input, and the target word vector representation Q, the upper and lower word vector representation K and the original vector representation V of the target word and the upper and lower words are obtained through matrix calculation and linear transformation according to a formula (2).
Q,K,V＝liner(Q,K,V) (2)
(2) Calculating the Attention weight, and carrying out normalization processing:
in the formula (4), K i Key value representing the ith word, W i Weight vector representing the i-th word, f (Q, K) i ) Is the similarity.
(3) Weighted summation, resulting in an enhanced semantic representation:
in the formula (5), self-Attention represents the probability distribution of Attention, J represents the dimension, and J represents the dimension upper bound.
The historical transaction data is embedded through the Bert model to obtain feature vectors integrated with the content information of the whole chapter, then secondary embedding is carried out, and the feature vectors (namely initial feature information) output by the Bert model are input into the Bi-GRU network model. For the Bi-GRU network model, its update gate and reset gate are central. Update door U t Controlling the degree to which the information of the previous moment is kept to the current moment, and calculating the degree as formula (6); reset gate F t Indicating the degree to which the previous time information was forgotten, its calculation formula is as (7).
U t ＝σ[W u (h t-1 ,x t )] (6)
F t ＝σ[W f (h t-1 ,x t )] (7)
For equations (8) and (9),indicating candidate activation state at t time, h t Then the candidate state is represented; w represents a weight matrix, x t Representing the network input at time t.
S120, encrypting the target characteristic information based on the homomorphic encryption technology to generate encrypted characteristic information.
Among them, the full homomorphic encryption (Fully Homomorphic Encryption, FHE) technique is an encryption technique that computes and processes data without exposing the data content. And encrypting the target characteristic information based on the full homomorphic encryption technology to generate encrypted characteristic information. It will be appreciated that using FHE technology, banks can perform data analysis and processing with customer data encrypted, thereby protecting the privacy and security of customer data.
S130, analyzing the encryption characteristic information through a pre-trained client marketing grade evaluation model, and determining an encryption marketing grade evaluation result corresponding to the encryption characteristic information.
For example, if the pre-trained client marketing level assessment model is configured locally, the encryption characteristic information may be input into the pre-trained client marketing level assessment model, and the encryption marketing level assessment result output by the client marketing level assessment model may be obtained. Optionally, the encryption feature information is analyzed through a pre-trained client marketing level assessment model, and an encryption marketing level assessment result corresponding to the encryption feature information is determined, which includes: the encryption characteristic information is sent to a server, so that the server inputs the encryption characteristic information into a pre-trained client marketing grade assessment model, and an encryption marketing grade assessment result is generated; and acquiring the encryption marketing grade evaluation result fed back by the server. It can be understood that if the pre-trained client marketing grade assessment model is configured on the server, the encryption characteristic information is sent to the server, the server inputs the received encryption characteristic information into the client marketing grade assessment model, obtains the encryption marketing grade assessment information output by the client marketing grade assessment model, and obtains the encryption marketing grade assessment information fed back by the server.
It will be appreciated that the encrypted encryption characteristic information is input into a customer marketing rating assessment model by which the customer marketing rating is predicted. Assume that the vector after being embedded by the Bert model and the Bi-GRU network model is encrypted to be X= [ X ] 1 ,x 2 ,…x n ]The customer marketing rating thereof can be predicted by a customer marketing rating assessment model, calculated as formula (10),representing the calculated evaluation score corresponding to the consumer marketing rating.
Because the relationship between the text feature vector (and the encryption feature information) and the client marketing grade is complex and is not a simple linear relationship, the embodiment of the invention can select the nonlinear regression model constructed based on the neural network as the client marketing grade evaluation model. Exemplary, fig. 3 is a schematic diagram of a process for obtaining a client marketing grade according to an embodiment of the present invention, where target feature information in historical transaction data is extracted through Bert model embedding and Bi-GRU network model as shown in fig. 3, that is, feature vectors embedded twice are obtained through Bert model embedding and Bi-GRU network model, and the feature vectors are input into a neural network including two hidden layers, and only one neuron is used in an output layer to output a client marketing grade evaluation result, so that a nonlinear relationship between the input feature vectors and the client marketing grade evaluation result is realized.
Wherein, when training the client marketing grade assessment model, the mean square error calculation loss can be adopted:
in the formula (11), y i The true value is represented by a value that is true,representing the predicted value through the nonlinear regression model, m representing the number of samples. When the regression model is trained, the mean square error is selected as a loss function, and the model is optimized by adopting a gradient descent algorithm, so that the loss is reduced.
And S140, decrypting the encryption marketing grade assessment result to generate the marketing grade of the customer to be assessed.
In the embodiment of the invention, the encryption marketing grade assessment result is decrypted to generate the marketing grade of the customer to be assessed. Optionally, encrypting the target feature information based on the homomorphic encryption technology to generate encrypted feature information, including: calculating public and private key pairs of the target characteristic information based on a key generation algorithm; the public and private key pairs comprise a public key and a private key; encrypting the target characteristic information based on the public key in the public-private key pair by adopting an homomorphic encryption technology to generate encrypted characteristic information; and decrypting the encryption marketing grade assessment result to generate the marketing grade of the customer to be assessed, which comprises the following steps: and decrypting the encryption marketing grade assessment result based on the private key in the public-private key pair to generate the marketing grade of the customer to be assessed.
For example, the client basic information of the client to be evaluated can be obtained, the client basic information is processed based on a key generation algorithm, and a public key pair of the target characteristic information is generated. And encrypting the target characteristic information based on the public key in the public-private key pair by adopting an homomorphic encryption technology to generate encrypted characteristic information. And when the encryption marketing grade evaluation result is obtained, decrypting the encryption marketing grade evaluation result based on the private key in the public-private key pair to obtain the marketing grade of the client to be evaluated. The electronic device of the customer to be evaluated may then send its own marketing grade directly to the banking system.
Fig. 4 is a schematic diagram illustrating a process for obtaining a client marketing rating of a plurality of participants according to an embodiment of the present invention. As shown in fig. 4, if there are n participants, each participant terminal device calculates a public-private key pair (pk, sk) using a key generation algorithm Keyenc (·) (i=1, 2. Each participant terminal device embeds corresponding historical transaction data through a Bert model and a Bi-GERU network model for the second time to generate a low-dimension feature vector (namely target feature information) x with key information i (i=1, 2,) n. Each participant terminal device P 1 ,P 2 ,...,P n The identical encryption algorithm FHE is respectively used for the respective target characteristic information x i (i=1, 2,., n) to obtain respective encryption characteristic information h i →FHE(x i ) And the encryption characteristic information is uploaded to the server. The server respectively inputs each piece of encryption characteristic information into a pre-trained client marketing grade evaluation model, acquires an encryption marketing grade evaluation result, and feeds the encryption marketing grade evaluation result back to the corresponding participant terminal equipment. The terminal equipment of the party obtains the corresponding marketing grade Eval (y') - & gt f (h) through decryption of the key sk 1 ′,h 2 ′,...,h n '). Alternatively, the server may also feed the encryption marketing rating assessment results of the respective participants back to the banking system.
It can be understood that through the above process, each participant terminal device can transmit the ciphertext input by each participant terminal device to the server for secure multiparty calculation processing, and a returned calculation result is obtained from the server. In the whole process, the server does not know the original input of the terminal equipment of the participator, and only can process the ciphertext uploaded by the terminal equipment of the participator. In this way, the terminal equipment and the server of other participants are not aware of the original input information, so that the privacy of each participant is protected.
The client marketing grade assessment method provided by the embodiment of the invention comprises the following steps: acquiring historical transaction data of a customer to be evaluated, and extracting target characteristic information in the historical transaction data; encrypting the target characteristic information based on the homomorphic encryption technology to generate encrypted characteristic information; analyzing the encryption characteristic information through a pre-trained client marketing grade evaluation model, and determining an encryption marketing grade evaluation result corresponding to the encryption characteristic information; and decrypting the encryption marketing grade assessment result to generate the marketing grade of the customer to be assessed. By the technical scheme provided by the embodiment of the invention, the identification accuracy of the bank to the marketing rating of the potential customer is greatly improved, the problem that data between enterprises and banks cannot be mutually trusted is solved by an encryption mode, the risk of enterprise data leakage is reduced, and the efficiency of the bank for acquiring the marketing rating of the customer is improved. .
In some embodiments, before analyzing the encrypted feature information by a pre-trained customer marketing rating assessment model to determine an encrypted marketing rating assessment result corresponding to the encrypted feature information, further comprising: respectively acquiring encrypted sample characteristics transmitted by at least two participant terminal devices, and generating an encrypted sample characteristic set; the encrypted sample features are generated by encrypting sample features extracted from sample transaction data of a participant by corresponding participant terminal equipment; marking the corresponding encrypted sample features in the encrypted sample feature set based on the marketing grade of the participant sample transaction data to generate an encrypted training feature set; training a preset machine learning model based on the encryption training feature set to generate a client marketing grade assessment model. Wherein the preset machine learning model comprises a nonlinear regression model based on a neural network.
Optionally, the encrypted sample feature is a feature generated by extracting sample features in corresponding sample transaction data of the participant by the participant terminal device based on a Bert model and a Bi-GRU network model, and encrypting the sample features based on a full homomorphic encryption technology.
Illustratively, each of the at least two participant terminal devices extracts sample features in its own participant transaction data based on the Bert model and the Bi-GRU network model, encrypts the sample features, and generates encrypted sample features. And obtaining the encrypted sample characteristics transmitted by at least two participant terminal devices, generating an encrypted sample characteristic set, and marking the encrypted sample characteristics corresponding to the encrypted sample characteristic set based on the marketing grade of the participant sample transaction data, so as to generate an encrypted training characteristic set. And carrying out iterative training on a pre-constructed nonlinear regression model based on the neural network based on the encryption training feature set to generate a client marketing grade assessment model.
Example two
Fig. 5 is a schematic structural diagram of a customer marketing rating assessment device according to a second embodiment of the present invention. As shown in fig. 5, the apparatus includes:
the target feature information extraction module 510 is configured to obtain historical transaction data of a customer to be evaluated, and extract target feature information in the historical transaction data;
the encryption feature information generation module 520 is configured to encrypt the target feature information based on an homomorphic encryption technique, and generate encryption feature information;
an encryption marketing rating assessment result determining module 530, configured to analyze the encryption feature information through a pre-trained client marketing rating assessment model, and determine an encryption marketing rating assessment result corresponding to the encryption feature information;
and the marketing grade generating module 540 is configured to decrypt the encrypted marketing grade evaluation result and generate a marketing grade of the customer to be evaluated.
Optionally, the target feature information extracting module is configured to:
inputting the historical transaction data into a Bert model, and acquiring output initial characteristic information of the Bert model;
and inputting the initial characteristic information into a Bi-GRU network model, and acquiring target characteristic information output by the Bi-GRU network model.
Optionally, the encryption characteristic information generating module is configured to:
calculating public and private key pairs of the target characteristic information based on a key generation algorithm; the public and private key pairs comprise a public key and a private key;
encrypting the target characteristic information based on the public key in the public-private key pair by adopting an homomorphic encryption technology to generate encrypted characteristic information;
the marketing grade generating module is used for:
and decrypting the encryption marketing grade assessment result based on the private key in the public-private key pair to generate the marketing grade of the customer to be assessed.
Optionally, the encryption marketing rating evaluation result determining module is configured to:
the encryption characteristic information is sent to a server, so that the server inputs the encryption characteristic information into a pre-trained client marketing grade assessment model, and an encryption marketing grade assessment result is generated;
and acquiring the encryption marketing grade evaluation result fed back by the server.
Optionally, the method further comprises:
the encryption sample feature set generation module is used for respectively acquiring encryption sample features transmitted by at least two participant terminal devices before the encryption feature information is analyzed through a pre-trained client marketing level evaluation model and an encryption marketing level evaluation result corresponding to the encryption feature information is determined, so as to generate an encryption sample feature set; the encrypted sample features are generated by encrypting sample features extracted from sample transaction data of a participant by corresponding participant terminal equipment;
the encryption training feature set generation module is used for marking the corresponding encryption sample features in the encryption sample feature set based on the marketing grade of the participant sample transaction data to generate an encryption training feature set;
and the client marketing grade evaluation model generation module is used for training a preset machine learning model based on the encryption training feature set to generate a client marketing grade evaluation model.
Optionally, the preset machine learning model includes a nonlinear regression model based on a neural network.
Optionally, the encrypted sample feature is a feature generated by extracting sample features in corresponding sample transaction data of the participant by the participant terminal device based on a Bert model and a Bi-GRU network model, and encrypting the sample features based on a full homomorphic encryption technology.
The client marketing grade assessment device provided by the embodiment of the invention can execute the client marketing grade assessment method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
Example III
Fig. 6 shows a schematic diagram of the structure of an electronic device 10 that may be used to implement an embodiment of the invention. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. Electronic equipment may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.
As shown in fig. 6, the electronic device 10 includes at least one processor 11, and a memory, such as a Read Only Memory (ROM) 12, a Random Access Memory (RAM) 13, etc., communicatively connected to the at least one processor 11, in which the memory stores a computer program executable by the at least one processor, and the processor 11 may perform various appropriate actions and processes according to the computer program stored in the Read Only Memory (ROM) 12 or the computer program loaded from the storage unit 18 into the Random Access Memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 may also be stored. The processor 11, the ROM 12 and the RAM 13 are connected to each other via a bus 14. An input/output (I/O) interface 15 is also connected to bus 14.
Various components in the electronic device 10 are connected to the I/O interface 15, including: an input unit 16 such as a keyboard, a mouse, etc.; an output unit 17 such as various types of displays, speakers, and the like; a storage unit 18 such as a magnetic disk, an optical disk, or the like; and a communication unit 19 such as a network card, modem, wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.
The processor 11 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, digital Signal Processors (DSPs), and any suitable processor, controller, microcontroller, etc. The processor 11 performs the various methods and processes described above, such as the customer marketing rating method.
In some embodiments, the customer marketing rating method may be implemented as a computer program tangibly embodied on a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 10 via the ROM 12 and/or the communication unit 19. One or more of the steps of the customer marketing rating method described above may be performed when the computer program is loaded into the RAM 13 and executed by the processor 11. Alternatively, in other embodiments, processor 11 may be configured to perform the customer marketing rating method by any other suitable means (e.g., by means of firmware).
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.
A computer program for carrying out methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be implemented. The computer program may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which a user can provide input to the electronic device. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.
The computing system may include clients and servers. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service are overcome.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present invention may be performed in parallel, sequentially, or in a different order, so long as the desired results of the technical solution of the present invention are achieved, and the present invention is not limited herein.
The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.
Claims (10)
1. A method of customer marketing rating assessment comprising:
acquiring historical transaction data of a customer to be evaluated, and extracting target characteristic information in the historical transaction data;
encrypting the target characteristic information based on the homomorphic encryption technology to generate encrypted characteristic information;
analyzing the encryption characteristic information through a pre-trained client marketing grade evaluation model, and determining an encryption marketing grade evaluation result corresponding to the encryption characteristic information;
and decrypting the encryption marketing grade assessment result to generate the marketing grade of the customer to be assessed.
2. The method of claim 1, wherein extracting target feature information in the historical transaction data comprises:
inputting the historical transaction data into a Bert model, and acquiring output initial characteristic information of the Bert model;
and inputting the initial characteristic information into a Bi-GRU network model, and acquiring target characteristic information output by the Bi-GRU network model.
3. The method of claim 1, wherein encrypting the target feature information based on a fully homomorphic encryption technique generates encrypted feature information, comprising:
calculating public and private key pairs of the target characteristic information based on a key generation algorithm; the public and private key pairs comprise a public key and a private key;
encrypting the target characteristic information based on the public key in the public-private key pair by adopting an homomorphic encryption technology to generate encrypted characteristic information;
and decrypting the encryption marketing grade assessment result to generate the marketing grade of the customer to be assessed, which comprises the following steps:
and decrypting the encryption marketing grade assessment result based on the private key in the public-private key pair to generate the marketing grade of the customer to be assessed.
4. The method of claim 1, wherein analyzing the encrypted feature information by a pre-trained customer marketing rating assessment model to determine an encrypted marketing rating assessment result corresponding to the encrypted feature information comprises:
the encryption characteristic information is sent to a server, so that the server inputs the encryption characteristic information into a pre-trained client marketing grade assessment model, and an encryption marketing grade assessment result is generated;
and acquiring the encryption marketing grade evaluation result fed back by the server.
5. The method of claim 1, further comprising, prior to analyzing the encrypted feature information by a pre-trained customer marketing rating assessment model to determine an encrypted marketing rating assessment result corresponding to the encrypted feature information:
respectively acquiring encrypted sample characteristics transmitted by at least two participant terminal devices, and generating an encrypted sample characteristic set; the encrypted sample features are generated by encrypting sample features extracted from sample transaction data of a participant by corresponding participant terminal equipment;
marking the corresponding encrypted sample features in the encrypted sample feature set based on the marketing grade of the participant sample transaction data to generate an encrypted training feature set;
training a preset machine learning model based on the encryption training feature set to generate a client marketing grade assessment model.
6. The method of claim 5, wherein the pre-set machine learning model comprises a neural network-based nonlinear regression model.
7. The method of claim 5, wherein the encrypted sample feature is a feature generated by the participant terminal device extracting sample features in corresponding participant sample transaction data based on a Bert model and a Bi-GRU network model and encrypting the sample features based on a fully homomorphic encryption technique.
8. A consumer marketing rating system, comprising:
the target feature information extraction module is used for acquiring historical transaction data of the clients to be evaluated and extracting target feature information in the historical transaction data;
the encryption characteristic information generation module is used for encrypting the target characteristic information based on the homomorphic encryption technology to generate encryption characteristic information;
the encryption marketing grade evaluation result determining module is used for analyzing the encryption characteristic information through a pre-trained client marketing grade evaluation model and determining an encryption marketing grade evaluation result corresponding to the encryption characteristic information;
and the marketing grade generating module is used for decrypting the encryption marketing grade evaluation result to generate the marketing grade of the client to be evaluated.
9. An electronic device, the electronic device comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the customer marketing rating assessment method of any of claims 1-7.
10. A computer readable storage medium storing computer instructions for causing a processor to perform the customer marketing rating method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311666513.1A CN117670399A (en) | 2023-12-06 | 2023-12-06 | Customer marketing grade evaluation method, device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311666513.1A CN117670399A (en) | 2023-12-06 | 2023-12-06 | Customer marketing grade evaluation method, device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117670399A true CN117670399A (en) | 2024-03-08 |
Family
ID=90076622
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311666513.1A Pending CN117670399A (en) | 2023-12-06 | 2023-12-06 | Customer marketing grade evaluation method, device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117670399A (en) |
-
2023
- 2023-12-06 CN CN202311666513.1A patent/CN117670399A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020253358A1 (en) | Service data risk control analysis processing method, apparatus and computer device | |
| TWI689841B (en) | Data encryption, machine learning model training method, device and electronic equipment | |
| TWI706333B (en) | Fraud transaction identification method, device, server and storage medium | |
| WO2022126970A1 (en) | Method and device for financial fraud risk identification, computer device, and storage medium | |
| CN108876133A (en) | Risk assessment processing method, device, server and medium based on business information | |
| CN112862298B (en) | Credit evaluation method for user portrait | |
| Gupta et al. | HiSA-SMFM: historical and sentiment analysis based stock market forecasting model | |
| US11682018B2 (en) | Machine learning model and narrative generator for prohibited transaction detection and compliance | |
| CN112348660A (en) | Method and device for generating risk warning information and electronic equipment | |
| WO2023000794A1 (en) | Service prediction model training method and apparatus for protecting data privacy | |
| CN113361962A (en) | Method and device for identifying enterprise risk based on block chain network | |
| CN114611008B (en) | User service strategy determination method and device based on federal learning and electronic equipment | |
| WO2022142903A1 (en) | Identity recognition method and apparatus, electronic device, and related product | |
| CN112200382A (en) | Training method and device of risk prediction model | |
| CN114186256A (en) | Neural network model training method, device, equipment and storage medium | |
| CN115545216A (en) | Service index prediction method, device, equipment and storage medium | |
| CN112101609B (en) | Prediction system, method and device for user repayment timeliness and electronic equipment | |
| CN115982654B (en) | Node classification method and device based on self-supervision graph neural network | |
| CN117114901A (en) | Method, device, equipment and medium for processing insurance data based on artificial intelligence | |
| Vidanelage et al. | Study on machine learning techniques with conventional tools for payment fraud detection | |
| WO2023165145A1 (en) | Time sequence traffic prediction method and apparatus, storage medium, and electronic device | |
| CN117670399A (en) | Customer marketing grade evaluation method, device, electronic equipment and storage medium | |
| US20190130254A1 (en) | Population Anomaly Detection Through Deep Gaussianization | |
| CN115018608A (en) | Risk prediction method and device and computer equipment | |
| CN113887214A (en) | Artificial intelligence based wish presumption method and related equipment thereof |
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 |