CN112035541A - Client image drawing method and device, computer readable storage medium and terminal equipment - Google Patents
Client image drawing method and device, computer readable storage medium and terminal equipment Download PDFInfo
- Publication number
- CN112035541A CN112035541A CN202010888273.XA CN202010888273A CN112035541A CN 112035541 A CN112035541 A CN 112035541A CN 202010888273 A CN202010888273 A CN 202010888273A CN 112035541 A CN112035541 A CN 112035541A
- Authority
- CN
- China
- Prior art keywords
- client
- evaluation
- dimension
- value
- preset
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/64—Protecting data integrity, e.g. using checksums, certificates or signatures
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Software Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Databases & Information Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Bioethics (AREA)
- General Health & Medical Sciences (AREA)
- Computer Hardware Design (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the technical field of artificial intelligence, and particularly relates to a client image drawing method and device, a computer readable storage medium and terminal equipment. The method comprises the steps of receiving a client portrait instruction, and extracting a client identifier of a client to be pictured from the client portrait instruction; acquiring information of the client to be imaged on each preset evaluation dimension from a preset database according to the client identification, wherein each evaluation dimension comprises a plurality of evaluation factors; on each evaluation dimension, weighting and summing the normalization information on each evaluation factor according to preset factor weight to obtain an evaluation value on each evaluation dimension; carrying out weighted summation on the evaluation values in all the evaluation dimensions according to preset dimension weights to obtain the image value of the client to be imaged; and selecting a client label corresponding to the image value of the client to be imaged, and determining the selected client label as the image result of the client to be imaged, so that the accuracy of the result is greatly improved.
Description
Technical Field
The invention belongs to the technical field of artificial intelligence, and particularly relates to a client image drawing method and device, a computer readable storage medium and terminal equipment.
Background
In the prior art, when a client needs to be portrait, several common information such as sex, age, income and the like are generally used as the basis, although the method is widely applied, the characteristics of the client can be reflected to a certain extent, but the method is too monotonous in the current big data environment, the utilization of the client information is not sufficient, and the obtained portrait result of the client is not accurate enough.
Disclosure of Invention
In view of this, embodiments of the present invention provide a client image method, a client image device, a computer-readable storage medium, and a terminal device, so as to solve the problem that the existing client image method is not sufficient in utilizing client information and the obtained client image result is often not accurate enough.
A first aspect of an embodiment of the present invention provides a client representation method, which may include:
receiving a client portrait instruction, and extracting a client identifier of a client to be pictured from the client portrait instruction;
acquiring information of the client to be imaged on each preset evaluation dimension from a preset database according to the client identification, wherein each evaluation dimension comprises a plurality of evaluation factors;
normalizing the information of the client to be portrait on each evaluation dimension to obtain normalized information;
on each evaluation dimension, weighting and summing the normalization information on each evaluation factor according to preset factor weight to obtain an evaluation value on each evaluation dimension;
carrying out weighted summation on the evaluation values in all the evaluation dimensions according to preset dimension weights to obtain the image value of the client to be imaged;
and selecting a client label corresponding to the image value of the client to be imaged, and determining the selected client label as the image result of the client to be imaged.
Further, the selecting a customer label corresponding to the image value of the customer to be imaged includes:
respectively acquiring reference image value sets corresponding to preset client labels from a preset label database, wherein each reference image value set comprises a plurality of reference image value subsets, and each reference image value subset comprises reference image values of a plurality of historical clients in a preset time interval;
for each client label, respectively calculating a first distance between the portrait value of the client to be portrait and each reference portrait value subset, and calculating a second distance between the portrait value of the client to be portrait and the reference portrait value set according to the first distance of each reference portrait value subset and a preset time interval coefficient; the time interval coefficient of each reference image value subset is inversely related to the time length of the time interval from the current system time;
and selecting the customer label with the shortest second distance as the customer label corresponding to the image value of the customer to be imaged.
Further, the setting process of the factor weight comprises:
determining an optimization target set, wherein the optimization target set comprises at least one optimization target;
acquiring a training sample set from a preset database, wherein each training sample in the training sample set comprises information of a historical client on each evaluation dimension and a label value on each optimization target;
performing iterative training on factor weights on target dimensions according to the training sample set, and respectively calculating a first output value of each training sample on each optimization target on the target dimensions in each iterative training process, wherein the target dimensions are any one evaluation dimension;
calculating a first global error of the training sample set according to a first output value and a label value of each training sample on each optimization target;
and determining the factor weight corresponding to the minimum value of the first global error as the factor weight trained on the target dimension.
Further, the setting process of the dimension weight comprises:
performing iterative training on the dimension weights according to the training sample set, and respectively calculating a second output value of each training sample on each optimization target in each iterative training process;
calculating a second global error of the training sample set according to a second output value and a label value of each training sample on each optimization target;
and determining the dimension weight corresponding to the minimum value of the second global error as the trained dimension weight.
Further, the calculating the first output value of each training sample on each optimization target in the target dimension respectively includes:
calculating a first output value of any one training sample on the k-th optimization target according to the following formula:
k is more than or equal to 1 and less than or equal to K, K is the total number of optimization targets in the optimization target set, and MkA set of sequence numbers, w, for each evaluation factor in the target dimensioniFactor weight for the ith evaluation factor of the target dimension, i ∈ Mk，x′iFor the normalized information of the training sample on the i-th evaluation factor of the target dimension, bkAnd is the bias term of the k-th optimization target, sigma is a preset activation function,
a first output value on the kth optimization objective for the training sample.
Further, the calculating a first global error of the training sample set according to the first output value and the label value of each training sample on the respective optimization target includes:
calculating a first sample error for any one of the training samples according to:
wherein, ykFor the label value of the sample on the kth optimization target, θkThe relative weight of a preset kth optimization target is beta, a preset regularization weight is beta, and L is a first sample error of the sample;
and summing the first sample errors of the training samples to obtain a first global error of the training sample set.
Further, after selecting the customer label corresponding to the image value of the customer to be imaged, the method further comprises the following steps:
and opening a system authority corresponding to the selected client label for the client to be imaged in a preset client management system. A second aspect of an embodiment of the present invention provides a client representation apparatus, which may include:
the client identification extraction module is used for receiving a client portrait instruction and extracting a client identification of a client to be pictured from the client portrait instruction;
the client information acquisition module is used for acquiring information of the client to be imaged on each preset evaluation dimension from a preset database according to the client identification, wherein each evaluation dimension comprises a plurality of evaluation factors;
the normalization processing module is used for performing normalization processing on the information of the client to be portrait on each evaluation dimension to obtain normalized information;
the first processing module is used for weighting and summing the normalization information on each evaluation factor on each evaluation dimension according to preset factor weight to obtain an evaluation value on each evaluation dimension;
the second processing module is used for carrying out weighted summation on the evaluation values in all the evaluation dimensions according to preset dimension weights to obtain the image value of the client to be imaged;
and the client label selecting module is used for selecting a client label corresponding to the image value of the client to be imaged and determining the selected client label as the image result of the client to be imaged.
Further, the customer tag selecting module may include:
the system comprises a reference image value set acquisition unit, a reference image value set acquisition unit and a display unit, wherein the reference image value set acquisition unit is used for respectively acquiring reference image value sets corresponding to preset client labels from a preset label database, each reference image value set comprises a plurality of reference image value subsets, and each reference image value subset comprises reference image values of a plurality of historical clients in a preset time interval;
the distance calculation unit is used for calculating a first distance between the image value of the client to be imaged and each reference image value subset for each client label, and calculating a second distance between the image value of the client to be imaged and the reference image value set according to the first distance of each reference image value subset and a preset time interval coefficient; the time interval coefficient of each reference image value subset is inversely related to the time length of the time interval from the current system time;
and the label selecting unit is used for selecting the customer label with the shortest second distance as the customer label corresponding to the image value of the customer to be imaged. Further, the client representation apparatus may further comprise:
the optimization target determination module is used for determining an optimization target set, wherein the optimization target set comprises at least one optimization target;
the training sample acquisition module is used for acquiring a training sample set from a preset database, wherein each training sample in the training sample set comprises information of a historical client on each evaluation dimension and a label value on each optimization target;
the first output value calculation module is used for carrying out iterative training on the factor weight on a target dimension according to the training sample set, and in each iterative training process, calculating a first output value of each training sample on each optimization target on the target dimension respectively, wherein the target dimension is any one evaluation dimension;
the first global error calculation module is used for calculating a first global error of the training sample set according to a first output value and a label value of each training sample on each optimization target;
and the factor weight determining module is used for determining the factor weight corresponding to the minimum value of the first global error as the factor weight trained on the target dimension.
Further, the client representation apparatus may further comprise:
the second output value calculation module is used for carrying out iterative training on the dimension weight according to the training sample set, and calculating a second output value of each training sample on each optimization target in each iterative training process;
the second global error calculation module is used for calculating a second global error of the training sample set according to a second output value and a label value of each training sample on each optimization target;
and the dimension weight determining module is used for determining the dimension weight corresponding to the minimum value of the second global error as the trained dimension weight.
Further, the first output value calculating module is specifically configured to calculate a first output value of any one training sample on the kth optimization target according to the following formula:
k is more than or equal to 1 and less than or equal to K, K is the total number of optimization targets in the optimization target set, and MkA set of sequence numbers, w, for each evaluation factor in the target dimensioniFactor weight for the ith evaluation factor of the target dimension, i ∈ Mk，x′iFor the normalized information of the training sample on the i-th evaluation factor of the target dimension, bkAnd is the bias term of the k-th optimization target, sigma is a preset activation function,
a first output value on the kth optimization objective for the training sample.
Further, the first global error calculation module may include:
a first sample error calculation unit for calculating a first sample error of any one of the training samples according to:
wherein, ykFor the label value of the sample on the kth optimization target, θkThe relative weight of a preset kth optimization target is beta, a preset regularization weight is beta, and L is a first sample error of the sample;
and the first global error calculation unit is used for summing the first sample errors of the training samples to obtain a first global error of the training sample set.
Further, the client representation apparatus may further comprise:
and the system authority control module is used for opening the system authority corresponding to the selected client label for the client to be imaged in a preset client management system.
A third aspect of embodiments of the present invention provides a computer readable storage medium having computer readable instructions stored thereon which, when executed by a processor, implement the steps of any of the above-described client representation methods.
A fourth aspect of an embodiment of the present invention provides a terminal device, including a memory, a processor, and computer readable instructions stored in the memory and executable on the processor, where the processor implements any one of the steps of the client representation method when executing the computer readable instructions.
Compared with the prior art, the embodiment of the invention has the following beneficial effects: the embodiment of the invention receives a client portrait command and extracts a client identifier of a client to be pictured from the client portrait command; acquiring information of the client to be imaged on each preset evaluation dimension from a preset database according to the client identification, wherein each evaluation dimension comprises a plurality of evaluation factors; normalizing the information of the client to be portrait on each evaluation dimension to obtain normalized information; on each evaluation dimension, weighting and summing the normalization information on each evaluation factor according to preset factor weight to obtain an evaluation value on each evaluation dimension; carrying out weighted summation on the evaluation values in all the evaluation dimensions according to preset dimension weights to obtain the image value of the client to be imaged; and selecting a client label corresponding to the image value of the client to be imaged, and determining the selected client label as the image result of the client to be imaged. According to the embodiment of the invention, the sufficient and comprehensive customer information can be obtained through a plurality of evaluation factors on a plurality of evaluation dimensions, the customer information is uniformly measured through normalization processing, and a more accurate customer portrait result can be obtained through processing of the customer portrait model.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise.
FIG. 1 is a flow chart of an embodiment of a method for imaging a customer in an embodiment of the present invention;
FIG. 2 is a schematic diagram of a machine learning model that determines specific values for various weights in a customer representation model;
FIG. 3 is a schematic flow diagram of a process for setting factor weights;
FIG. 4 is a block diagram of one embodiment of a client imaging device in accordance with one embodiment of the present invention;
fig. 5 is a schematic block diagram of a terminal device in an embodiment of the present invention.
Detailed Description
In order to make the objects, features and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the embodiments described below are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, an embodiment of a method for client image rendering according to an embodiment of the present invention may include:
step S101, receiving a client portrait command, and extracting a client identification of a client to be portrait from the client portrait command.
When a relevant worker needs to perform client portrait for a certain client, a client portrait instruction can be issued to a terminal device executing the client portrait, and the client portrait instruction carries a client identifier of the client to be portrait. The customer identification may include, but is not limited to, an identification number, a cell phone number, a social security number, a public deposit number, a policy number, and other identification that may uniquely identify the customer.
After receiving the client portrait command, the terminal equipment can extract the client identification of the client to be portrait from the client portrait command, and perform client portrait according to the subsequent steps.
And S102, acquiring information of the customer to be imaged on each preset evaluation dimension from a preset database according to the customer identification.
Wherein each evaluation dimension comprises a plurality of evaluation factors. Specifically, which evaluation dimensions are selected and which evaluation factors can be included in each evaluation dimension can be set according to actual conditions.
In a specific implementation of the embodiment of the present invention, three dimensions, namely, a time dimension, a breadth dimension, and a frequency dimension, may be selected.
The time dimension mainly refers to the time when a customer purchases a product or registers an enterprise APP for the first time, and the earlier the customer contacts the enterprise, the more familiar the enterprise and the more compact the relationship. In the embodiment of the invention, the time when the customer purchases the product for the last time can be added, and the closer the last purchase is, the more compact the relationship is.
The breadth dimension mainly refers to the communication breadth of the client and the enterprise, and for a diversified group, each branch company and each product must be considered to comprehensively evaluate the intimacy of the client. In the embodiment of the present invention, whether a customer is a customer of each service Unit (BU), whether the customer is an effective customer of each BU, and the number of contracts across the BU may be considered.
The frequency dimension is mainly the number of times of interaction between a client and an enterprise, and is also an important dimension for measuring the intimacy between the client and the enterprise. The data used in this dimension is the most abundant and not just a simple "number of times" concept. In the embodiment of the invention, the contract number of the client, the purchase times and the amount of various products, the number of times of participating in the enterprise activities of the client, the number of times of interaction between the client and the marketer and the like can be comprehensively considered.
In a specific implementation of the embodiment of the present invention, each evaluation dimension includes an evaluation factor specifically:
in the time dimension may include, but is not limited to: evaluating factors such as the current time of APP registration, the current time of a product purchased for the first time, the current time of recent APP activity, the current time of a product purchased for the recent time and the like;
in the breadth dimension may include, but is not limited to: holding the total number of product types, newly increasing the number of group series, reducing the number of group series, whether to debit card customers, whether to debit card valid customers, whether to stock valid customers, whether to trust valid customers and the like;
in the frequency dimension may include, but is not limited to: the number of contracts, the number of times of product logout, total assets, assets newly added in about 6 months, total liabilities, liabilities newly added in about 6 months, the number of short message sending times, the number of telephone communication times, the number of referrals, APP login days, information reading numbers, activity sign-in times, after-sales service times and other evaluation factors.
According to actual conditions, the information of the client on each evaluation dimension can be stored in the same database in advance, or can be distributed and stored in different databases in advance, and when the information needs to be used, the corresponding information is retrieved from the database according to the unique client identifier of the client, and is summarized so as to perform subsequent processing.
Step S103, carrying out normalization processing on the information of the customer to be portrait on each evaluation dimension to obtain normalized information.
In the embodiment of the present invention, the normalization processing manner may be selected according to actual situations, and the calculation result of a single evaluation factor is limited to 0 to 1. For different evaluation factors, the same normalization processing mode may be adopted, and different normalization processing modes may also be adopted, which is not specifically limited in the embodiment of the present invention.
In a specific implementation of the embodiment of the present invention, for a larger evaluation factor, the normalization process may be performed using the following formula:
wherein, x is the evaluation factor before normalization, x' is the evaluation factor after normalization, μ is the center of the distribution of the evaluation factors, which can be set as the mean or median of the evaluation factors, and when the evaluation factor takes the value μ, the processed value is 0.5. d is a scaling coefficient, the change speed of the evaluation factor after transformation can be adjusted, and can be set as the standard deviation of the evaluation factor or (3/4 quantile-1/4 quantile)/2. For example, the total asset is often a large value and the difference between customers is large, where μ can be equal to 5000 as the mean value of the customer's assets and d equal to 2000 as the standard deviation, so that the changed factors will be more uniform and suitable for final scoring.
Further, to handle the relationship between historical accumulated data and recent incremental data, further processing may be performed using the following equation:
xtotalfor normalizing the processed historical accumulated data, xincreFor the recent incremental data after normalization processing, x' is a result of normalizing the two together, and α is a preset coefficient for balancing the relative importance of the two, and the specific value can be set according to actual conditions. Take the total asset as an example, xtotalAs total assets, xincreFor the newly added assets in the last 6 months, the total assets and the newly added assets in the last 6 months can be normalized together by the formula.
And S104, respectively carrying out weighted summation on the normalization information on each evaluation factor according to preset factor weight in each evaluation dimension to obtain an evaluation value on each evaluation dimension.
And S105, carrying out weighted summation on the evaluation values in all the evaluation dimensions according to preset dimension weights to obtain the image value of the customer to be imaged.
Recording the number of the evaluation dimensions as D, recording the serial number of each evaluation dimension as D, wherein D is more than or equal to 1 and less than or equal to D, and recording the dimension weight of the D-th evaluation dimension as wd', the number of evaluation factors of the d-th evaluation dimension is denoted as NdWherein the sequence number of each evaluation dimension is marked as nd，1≤nd≤NdN is to bedThe information normalized by each evaluation factor is recorded as
Will n bedThe factor weight of each evaluation factor is recorded as
The calculation process of step S104 and step S105 can be expressed as a customer image model as shown in the following formulaType (2):
result is the final calculated customer portrait Result.
As can be seen from the above formula, the key to establishing the customer portrait model is to determine the specific values of the weights, and there are two ways to determine the weights: one is to manually determine the weight according to expert experience; the other is to use a machine learning algorithm to automatically learn weights with big data and to automatically iterate updating. The two methods of determining weights may be combined, because the machine-learned weights may be too concentrated, and a few evaluation factors may give too high weight, or some evaluation factors with low data quality may give too high weight. Expert experience can be utilized to set artificial constraints on the weights, and a machine learning method is utilized to determine specific numerical values within the constraints.
In an embodiment of the present invention, the specific values of the weights in the customer representation model are preferably determined by a machine learning model as shown in FIG. 2. Taking the setting process of the factor weight as an example, the process may specifically include the process shown in fig. 3:
and S301, determining an optimization target set.
The set of optimization objectives includes at least one optimization objective. Preferably, the optimization goal set may include more than two optimization goals, which may be set according to specific situations, and may include, but are not limited to, whether to purchase long-term insurance, whether to purchase short-term insurance, whether to purchase financing, whether to purchase trust, and the like, for example, within a specified fixed period of time.
Step S302, a training sample set is obtained from a preset database.
Each training sample in the training sample set comprises information of a historical client on each evaluation dimension and a label value on each optimization target.
For example, if the optimization objective is to purchase long-term insurance, when a certain historical customer does not purchase long-term insurance within a specified fixed time period, the label value on the optimization objective is 0, and when a certain historical customer purchases long-term insurance within a specified fixed time period, the label value on the optimization objective is 1; if the optimization target is the purchase of short-term insurance, when a certain historical client does not purchase the short-term insurance in a specified fixed time period, the label value of the historical client on the optimization target is 0, and when the certain historical client purchases the short-term insurance in the specified fixed time period, the label value of the historical client on the optimization target is 1; if the optimization target is purchasing financing, when a certain historical customer does not purchase financing within a specified fixed time period, the label value of the historical customer on the optimization target is 0, and when the certain historical customer purchases financing within the specified fixed time period, the label value of the historical customer on the optimization target is 1; if the optimization target is a purchase trust, when a certain historical client does not purchase trust in a specified fixed time period, the label value of the historical client on the optimization target is 0, and when the certain historical client purchases trust in the specified fixed time period, the label value of the historical client on the optimization target is 1; other optimization objectives may be analogized.
And step S303, performing iterative training on the factor weight on the target dimension according to the training sample set, and respectively calculating a first output value of each training sample on each optimization target on the target dimension in each iterative training process.
The target dimension is any one evaluation dimension. As shown in fig. 2, the information on each evaluation factor of the target dimension acquired by the input layer is normalized by the normalization layer to obtain normalized information, each evaluation factor is configured with a pending factor weight in the weight layer, and these processes are shared by each optimization target, that is, all the optimization targets use the same input layer, normalization layer and weight layer.
In the subsequent Mask layer and output layer, the optimization objectives are different, and the k-th optimization objective is taken as an example for explanation. On a Mask layer, carrying out weighted summation on the normalization information on each evaluation factor according to the factor weight, and adding a preset bias term; and processing the processing result of the Mask layer by using a preset activation function at an output layer of the device to obtain an output value, wherein the output value is recorded as a first output value. That is, the first output value of any training sample on the kth optimization target can be calculated according to the following formula:
k is more than or equal to 1 and less than or equal to K, K is the total number of optimization targets in the optimization target set, and MkA set of sequence numbers, w, for each evaluation factor in the target dimensioniFactor weight for the ith evaluation factor of the target dimension, i ∈ Mk，x′iFor the normalized information of the training sample on the i-th evaluation factor of the target dimension, bkThe bias term for the kth optimization objective, denoted herein as the first bias term, σ is a preset activation function, which may include, but is not limited to, a sigmoid function,
a first output value on the kth optimization objective for the training sample.
As can be seen from the above process, multiple optimization objectives share the parameters of the weighting layer and normalization layer, but have the freedom to choose the evaluation factors and bias terms.
And step S304, calculating a first global error of the training sample set according to the first output value and the label value of each training sample on each optimization target.
Specifically, the first sample error of any one training sample can be calculated according to a loss function as shown in the following formula:
wherein, ykIs thatThe label value of the sample on the kth optimization target, θkIn the embodiment of the present invention, in order to avoid the problem of over-concentration and over-fitting of the weights, an L2 regularization term is added to the equation, where β is a preset regularization weight, and may be adjusted by using a method such as grid search during training, where L is an error of the sample, and is denoted as a first sample error.
It can be seen that if the importance of each optimization objective is equivalent, the loss function is the equal weight addition of each optimization objective; if business places great importance on one of the optimization objectives, then multitask learning degenerates to single-task learning, with the loss function dominated by this optimization objective.
And summing the first sample errors of the training samples to obtain a global error of the training sample set, wherein the global error is marked as a first global error.
Step S305, determining the factor weight corresponding to the minimum value of the first global error as the factor weight trained on the target dimension.
The larger the first global error is, the larger the difference is, and the training of the model becomes a process of minimizing the first global error. In the embodiment of the invention, the training sample set is used for carrying out iterative training on the model, the value of each factor weight is continuously corrected, and the optimal factor weight is finally determined.
The setting process of the dimension weight is similar to the setting process of the factor weight. Specifically, the dimension weights may be iteratively trained according to the training sample set, and in each iterative training process, an output value of each training sample on each optimization target is calculated, where this output value is denoted as a second output value. Then the second output value of any one training sample on the kth optimization objective can be calculated according to the following formula:
this formula is the same as the formula for calculating the first output value, but the specific symbolic meanings are slightly different, here, MkSet of sequence numbers for each evaluation dimension, wiDimension weight for the ith evaluation dimension, i ∈ Mk，x′iFor the image value of the training sample in the i-th evaluation dimension, bkThe bias term for the kth optimization objective, denoted herein as the second bias term, σ is a preset activation function, which may include, but is not limited to, a sigmoid function,
and a second output value of the training sample on the k-th optimization target.
Then, a global error of the training sample set is calculated according to the second output value and the label value of each training sample on each optimization target, where this global error is denoted as a second global error, and a specific calculation process of the second global error is similar to a calculation process of the first global error, and details thereof can be referred to the foregoing contents, and are not described herein again. And finally, determining the corresponding dimension weight when the second global error obtains the minimum value as the trained dimension weight.
After the specific values of the weights are determined, the client portrait model can be used for analyzing and processing the client information, so that a portrait value is obtained, and the portrait value represents the intimacy between the client and the enterprise.
And S106, selecting a client label corresponding to the image value of the client to be imaged, and determining the selected client label as the image result of the client to be imaged.
In a specific implementation of the embodiment of the present invention, the reference image value sets corresponding to the preset client tags may be respectively obtained from a preset tag database. Each reference image value set comprises a plurality of reference image value subsets, each reference image value subset comprises a plurality of image values of historical clients in a preset time interval, and the image values are marked as reference image values.
For each customer tag, a first distance may be calculated between the representation value of the customer to be represented and each of the subsets of reference representation values, respectively. Taking any one of the reference image value subsets as an example, absolute values of differences between the image value of the client to be imaged and each reference image value in the subset are respectively calculated, and an average value of the absolute values is used as a first distance between the image value of the client to be imaged and the subset.
Then, a second distance between the image value of the client to be rendered and the reference image value set may be calculated according to the first distance of each reference image value subset and a preset time interval coefficient.
The time interval coefficient of each reference image value subset is inversely related to the time length of the time interval from the current system time, namely, the closer the subset is to the current time, the larger the reference value of the data is, the larger the coefficient is correspondingly given, while the farther the subset is from the current time, the smaller the reference value of the data is, the smaller the coefficient is correspondingly given, and the first distance of each reference image value subset is weighted and averaged according to the time interval coefficient, so that the second distance between the image value of the client to be imaged and the reference image value set can be obtained.
And finally, selecting the customer label with the shortest second distance as the customer label corresponding to the image value of the customer to be imaged. The shorter the distance is, the higher the similarity is, and the characteristics of the customer can be reflected more accurately by selecting the customer label with the shortest distance.
In another specific implementation of the embodiment of the present invention, the client tags in three levels of "intimacy", "friendly" and "acquaintance" may be further divided according to the image values from high to low, and each client corresponds to one of the client tags. Two thresholds are preset and respectively recorded as a first threshold and a second threshold, wherein the first threshold is larger than the second threshold, when the portrait value of a certain client is larger than the first threshold, the client label of the client is determined to be the highest-level intimacy, when the portrait value of the certain client is smaller than the first threshold and larger than the second threshold, the client label of the client is determined to be the next-higher-level friendly, and when the portrait value of the certain client is smaller than the second threshold, the client label is determined to be the lowest-level acquaintance. The image value and the client label can be refreshed periodically to reflect the real-time state of the client.
Further, after the client label corresponding to the image value of the client to be imaged is selected, a system authority corresponding to the selected client label can be opened for the client to be imaged in a preset client management system. The system authority corresponding to the client label with higher grade is also larger, and conversely, the system authority corresponding to the client label with lower grade is also smaller.
Further, after the client portrait result of the client to be pictured is determined, the client portrait result can be uploaded to a block chain (Blockchain), so that the security and the fair transparency to the client are guaranteed. The customer may use his terminal device to download the customer portrait results from the blockchain to verify that the customer portrait results have been tampered with. The blockchain referred to in this example is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm, and the like. The blockchain is essentially a decentralized database, which is a string of data blocks associated by using cryptography, each data block contains information of a batch of network transactions, and the information is used for verifying the validity (anti-counterfeiting) of the information and generating the next block. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.
In summary, the embodiment of the present invention receives a client representation instruction, and extracts a client identifier of a client to be represented from the client representation instruction; acquiring information of the client to be imaged on each preset evaluation dimension from a preset database according to the client identification, wherein each evaluation dimension comprises a plurality of evaluation factors; normalizing the information of the client to be portrait on each evaluation dimension to obtain normalized information; on each evaluation dimension, weighting and summing the normalization information on each evaluation factor according to preset factor weight to obtain an evaluation value on each evaluation dimension; carrying out weighted summation on the evaluation values in all the evaluation dimensions according to preset dimension weights to obtain the image value of the client to be imaged; and selecting a client label corresponding to the image value of the client to be imaged, and determining the selected client label as the image result of the client to be imaged. According to the embodiment of the invention, the sufficient and comprehensive customer information can be obtained through a plurality of evaluation factors on a plurality of evaluation dimensions, the customer information is uniformly measured through normalization processing, and a more accurate customer portrait result can be obtained through processing of the customer portrait model.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
FIG. 4 is a block diagram of an embodiment of a client rendering device, in accordance with the present invention, corresponding to the client rendering method described in the previous embodiments.
In this embodiment, a client representation apparatus may include:
a client identifier extracting module 401, configured to receive a client portrait instruction, and extract a client identifier of a client to be pictured from the client portrait instruction;
a client information obtaining module 402, configured to obtain, from a preset database according to the client identifier, information of the client to be imaged in each preset evaluation dimension, where each evaluation dimension includes a plurality of evaluation factors;
a normalization processing module 403, configured to perform normalization processing on information of the to-be-portrait client in each evaluation dimension to obtain normalized information;
a first processing module 404, configured to perform weighted summation on the normalization information of each evaluation factor according to a preset factor weight on each evaluation dimension, respectively, to obtain an evaluation value on each evaluation dimension;
the second processing module 405 is configured to perform weighted summation on the evaluation values in each evaluation dimension according to preset dimension weights, so as to obtain an image value of the customer to be imaged;
and the client tag selecting module 406 is configured to select a client tag corresponding to the image value of the client to be imaged, and determine the selected client tag as the image result of the client to be imaged.
Further, the customer tag selecting module may include:
the system comprises a reference image value set acquisition unit, a reference image value set acquisition unit and a display unit, wherein the reference image value set acquisition unit is used for respectively acquiring reference image value sets corresponding to preset client labels from a preset label database, each reference image value set comprises a plurality of reference image value subsets, and each reference image value subset comprises reference image values of a plurality of historical clients in a preset time interval;
the distance calculation unit is used for calculating a first distance between the image value of the client to be imaged and each reference image value subset for each client label, and calculating a second distance between the image value of the client to be imaged and the reference image value set according to the first distance of each reference image value subset and a preset time interval coefficient; the time interval coefficient of each reference image value subset is inversely related to the time length of the time interval from the current system time;
and the label selecting unit is used for selecting the customer label with the shortest second distance as the customer label corresponding to the image value of the customer to be imaged.
Further, the client representation apparatus may further comprise:
the optimization target determination module is used for determining an optimization target set, wherein the optimization target set comprises at least one optimization target;
the training sample acquisition module is used for acquiring a training sample set from a preset database, wherein each training sample in the training sample set comprises information of a historical client on each evaluation dimension and a label value on each optimization target;
the first output value calculation module is used for carrying out iterative training on the factor weight on a target dimension according to the training sample set, and in each iterative training process, calculating a first output value of each training sample on each optimization target on the target dimension respectively, wherein the target dimension is any one evaluation dimension;
the first global error calculation module is used for calculating a first global error of the training sample set according to a first output value and a label value of each training sample on each optimization target;
and the factor weight determining module is used for determining the factor weight corresponding to the minimum value of the first global error as the factor weight trained on the target dimension.
Further, the client representation apparatus may further comprise:
the second output value calculation module is used for carrying out iterative training on the dimension weight according to the training sample set, and calculating a second output value of each training sample on each optimization target in each iterative training process;
the second global error calculation module is used for calculating a second global error of the training sample set according to a second output value and a label value of each training sample on each optimization target;
and the dimension weight determining module is used for determining the dimension weight corresponding to the minimum value of the second global error as the trained dimension weight.
Further, the first output value calculating module is specifically configured to calculate a first output value of any one training sample on the kth optimization target according to the following formula:
k is more than or equal to 1 and less than or equal to K, K is the total number of optimization targets in the optimization target set, and MkA set of sequence numbers, w, for each evaluation factor in the target dimensioniFactor weight for the ith evaluation factor of the target dimension, i ∈ Mk，x′iNormalizing information of the training sample on the ith evaluation factor of the target dimension，bkAnd is the bias term of the k-th optimization target, sigma is a preset activation function,
a first output value on the kth optimization objective for the training sample.
Further, the first global error calculation module may include:
a first sample error calculation unit for calculating a first sample error of any one of the training samples according to:
wherein, ykFor the label value of the sample on the kth optimization target, θkThe relative weight of a preset kth optimization target is beta, a preset regularization weight is beta, and L is a first sample error of the sample;
and the first global error calculation unit is used for summing the first sample errors of the training samples to obtain a first global error of the training sample set.
Further, the client representation apparatus may further comprise:
and the system authority control module is used for opening the system authority corresponding to the selected client label for the client to be imaged in a preset client management system.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described apparatuses, modules and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and reference may be made to the related descriptions of other embodiments for parts that are not described or illustrated in a certain embodiment.
Fig. 5 shows a schematic block diagram of a terminal device according to an embodiment of the present invention, and for convenience of description, only the relevant parts related to the embodiment of the present invention are shown.
In this embodiment, the terminal device 5 may be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The terminal device 5 may include: a processor 50, a memory 51, and computer readable instructions 52 stored in the memory 51 and executable on the processor 50, such as computer readable instructions to perform the client representation method described above. The processor 50, when executing the computer readable instructions 52, implements the steps in the various client representation method embodiments described above, such as steps S101-S106 shown in FIG. 1. Alternatively, the processor 50, when executing the computer readable instructions 52, implements the functions of the modules/units in the above-mentioned device embodiments, such as the functions of the modules 401 to 406 shown in fig. 4.
Illustratively, the computer readable instructions 52 may be partitioned into one or more modules/units that are stored in the memory 51 and executed by the processor 50 to implement the present invention. The one or more modules/units may be a series of computer-readable instruction segments capable of performing specific functions, which are used for describing the execution process of the computer-readable instructions 52 in the terminal device 5.
The Processor 50 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory 51 may be an internal storage unit of the terminal device 5, such as a hard disk or a memory of the terminal device 5. The memory 51 may also be an external storage device of the terminal device 5, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, which are provided on the terminal device 5. Further, the memory 51 may also include both an internal storage unit and an external storage device of the terminal device 5. The memory 51 is used for storing the computer readable instructions and other instructions and data required by the terminal device 5. The memory 51 may also be used to temporarily store data that has been output or is to be output.
Each functional unit in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes a plurality of computer readable instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and the like, which can store computer readable instructions.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A method for customer imaging, comprising:
receiving a client portrait instruction, and extracting a client identifier of a client to be pictured from the client portrait instruction;
acquiring information of the client to be imaged on each preset evaluation dimension from a preset database according to the client identification, wherein each evaluation dimension comprises a plurality of evaluation factors;
normalizing the information of the client to be portrait on each evaluation dimension to obtain normalized information;
on each evaluation dimension, weighting and summing the normalization information on each evaluation factor according to preset factor weight to obtain an evaluation value on each evaluation dimension;
carrying out weighted summation on the evaluation values in all the evaluation dimensions according to preset dimension weights to obtain the image value of the client to be imaged;
and selecting a client label corresponding to the image value of the client to be imaged, and determining the selected client label as the image result of the client to be imaged.
2. The customer imaging method according to claim 1, wherein the selecting a customer label corresponding to the image value of the customer to be imaged comprises:
respectively acquiring reference image value sets corresponding to preset client labels from a preset label database, wherein each reference image value set comprises a plurality of reference image value subsets, and each reference image value subset comprises reference image values of a plurality of historical clients in a preset time interval;
for each client label, respectively calculating a first distance between the portrait value of the client to be portrait and each reference portrait value subset, and calculating a second distance between the portrait value of the client to be portrait and the reference portrait value set according to the first distance of each reference portrait value subset and a preset time interval coefficient; the time interval coefficient of each reference image value subset is inversely related to the time length of the time interval from the current system time;
and selecting the customer label with the shortest second distance as the customer label corresponding to the image value of the customer to be imaged.
3. The customer portrayal method of claim 1, wherein the factor weight setting process comprises:
determining an optimization target set, wherein the optimization target set comprises at least one optimization target;
acquiring a training sample set from a preset database, wherein each training sample in the training sample set comprises information of a historical client on each evaluation dimension and a label value on each optimization target;
performing iterative training on factor weights on target dimensions according to the training sample set, and respectively calculating a first output value of each training sample on each optimization target on the target dimensions in each iterative training process, wherein the target dimensions are any one evaluation dimension;
calculating a first global error of the training sample set according to a first output value and a label value of each training sample on each optimization target;
and determining the factor weight corresponding to the minimum value of the first global error as the factor weight trained on the target dimension.
4. A client representation method as claimed in claim 3 wherein said dimension weight setting process comprises:
performing iterative training on the dimension weights according to the training sample set, and respectively calculating a second output value of each training sample on each optimization target in each iterative training process;
calculating a second global error of the training sample set according to a second output value and a label value of each training sample on each optimization target;
and determining the dimension weight corresponding to the minimum value of the second global error as the trained dimension weight.
5. A client representation method as claimed in claim 3 wherein said calculating a first output value for each training sample over a respective optimization objective in said objective dimension comprises:
calculating a first output value of any one training sample on the k-th optimization target according to the following formula:
k is more than or equal to 1 and less than or equal to K, K is the total number of optimization targets in the optimization target set, and MkA set of sequence numbers, w, for each evaluation factor in the target dimensioniFactor weight for the ith evaluation factor of the target dimension, i ∈ Mk，x′iFor the normalized information of the training sample on the i-th evaluation factor of the target dimension, bkAnd is the bias term of the k-th optimization target, sigma is a preset activation function,
a first output value on the kth optimization objective for the training sample.
6. The customer portrayal method of claim 5, wherein the computing of the first global error for the set of training examples based on the first output value and the label value of each training example on the respective optimization objective comprises:
calculating a first sample error for any one of the training samples according to:
wherein, ykFor the label value of the sample on the kth optimization target, θkThe relative weight of a preset kth optimization target is beta, a preset regularization weight is beta, and L is a first sample error of the sample;
and summing the first sample errors of the training samples to obtain a first global error of the training sample set.
7. The customer imaging method according to any one of claims 1 to 6, further comprising, after selecting a customer label corresponding to an imaging value of the customer to be imaged:
and opening a system authority corresponding to the selected client label for the client to be imaged in a preset client management system.
8. A client rendering device, comprising:
the client identification extraction module is used for receiving a client portrait instruction and extracting a client identification of a client to be pictured from the client portrait instruction;
the client information acquisition module is used for acquiring information of the client to be imaged on each preset evaluation dimension from a preset database according to the client identification, wherein each evaluation dimension comprises a plurality of evaluation factors;
the normalization processing module is used for performing normalization processing on the information of the client to be portrait on each evaluation dimension to obtain normalized information;
the first processing module is used for weighting and summing the normalization information on each evaluation factor on each evaluation dimension according to preset factor weight to obtain an evaluation value on each evaluation dimension;
the second processing module is used for carrying out weighted summation on the evaluation values in all the evaluation dimensions according to preset dimension weights to obtain the image value of the client to be imaged;
and the client label selecting module is used for selecting a client label corresponding to the image value of the client to be imaged and determining the selected client label as the image result of the client to be imaged.
9. A computer readable storage medium having computer readable instructions stored thereon which, when executed by a processor, implement the steps of a client representation method as claimed in any one of claims 1 to 7.
10. A terminal device comprising a memory, a processor and computer readable instructions stored in said memory and executable on said processor, wherein said processor when executing said computer readable instructions implements the steps of a client representation method as claimed in any one of claims 1 to 7.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010888273.XA CN112035541A (en) | 2020-08-28 | 2020-08-28 | Client image drawing method and device, computer readable storage medium and terminal equipment |
| PCT/CN2020/135552 WO2021147557A1 (en) | 2020-08-28 | 2020-12-11 | Customer portrait method, apparatus, computer-readable storage medium, and terminal device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010888273.XA CN112035541A (en) | 2020-08-28 | 2020-08-28 | Client image drawing method and device, computer readable storage medium and terminal equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112035541A true CN112035541A (en) | 2020-12-04 |
Family
ID=73587635
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010888273.XA Pending CN112035541A (en) | 2020-08-28 | 2020-08-28 | Client image drawing method and device, computer readable storage medium and terminal equipment |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN112035541A (en) |
| WO (1) | WO2021147557A1 (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112580887A (en) * | 2020-12-25 | 2021-03-30 | 百果园技术(新加坡)有限公司 | Weight determination method, device and equipment for multi-target fusion evaluation and storage medium |
| CN112598090A (en) * | 2021-03-08 | 2021-04-02 | 北京冠新医卫软件科技有限公司 | Method, device, equipment and system for health portrait |
| CN112801287A (en) * | 2021-01-26 | 2021-05-14 | 商汤集团有限公司 | Neural network performance evaluation method and device, electronic equipment and storage medium |
| CN112950225A (en) * | 2021-02-25 | 2021-06-11 | 中国工商银行股份有限公司 | Customer category determination method, device and storage medium |
| CN112991110A (en) * | 2021-04-25 | 2021-06-18 | 湖南知名未来科技有限公司 | Multi-dimensional portrait standard client type identification method and intellectual property monitoring system |
| CN113033444A (en) * | 2021-03-31 | 2021-06-25 | 北京金山云网络技术有限公司 | Age estimation method and device and electronic equipment |
| CN113064927A (en) * | 2021-03-24 | 2021-07-02 | 深圳市道通科技股份有限公司 | Client screening method and device, electronic equipment and computer readable storage medium |
| WO2021147557A1 (en) * | 2020-08-28 | 2021-07-29 | 平安科技（深圳）有限公司 | Customer portrait method, apparatus, computer-readable storage medium, and terminal device |
| CN113486041A (en) * | 2021-08-02 | 2021-10-08 | 南京邮电大学 | Client portrait management method and system based on block chain |
| CN113706019A (en) * | 2021-08-30 | 2021-11-26 | 平安银行股份有限公司 | Service capability analysis method, device, equipment and medium based on multidimensional data |
| CN113707296A (en) * | 2021-08-25 | 2021-11-26 | 平安国际智慧城市科技股份有限公司 | Medical treatment scheme data processing method, device, equipment and storage medium |
| CN113761134A (en) * | 2021-09-16 | 2021-12-07 | 平安国际智慧城市科技股份有限公司 | User portrait construction method and device, computer equipment and storage medium |
| CN116883048A (en) * | 2023-07-12 | 2023-10-13 | 广州朝辉智能科技有限公司 | Customer data processing method and device based on artificial intelligence and computer equipment |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113626499B (en) * | 2021-08-09 | 2024-05-10 | 青软创新科技集团股份有限公司 | Student portrait mining implementation method based on big data number bin technology |
| CN114119058B (en) * | 2021-08-10 | 2023-09-26 | 国家电网有限公司 | User portrait model construction method, device and storage medium |
| CN113780415B (en) * | 2021-09-10 | 2023-08-15 | 平安科技（深圳）有限公司 | User portrait generating method, device, equipment and medium based on applet game |
| CN114417955B (en) * | 2021-12-02 | 2023-10-31 | 国网浙江省电力有限公司 | Method, device and storage medium for generating power index image suitable for village in real time |
| CN114722252A (en) * | 2022-03-18 | 2022-07-08 | 深圳市小满科技有限公司 | Foreign trade user classification method based on user portrait and related equipment |
| CN114826682A (en) * | 2022-03-30 | 2022-07-29 | 京东科技信息技术有限公司 | Information delivery management method, device and system |
| CN115458099A (en) * | 2022-09-14 | 2022-12-09 | 广州商学院 | Person psychological image obtaining method and system based on questionnaire evaluation and electronic equipment |
| CN116089401B (en) * | 2023-02-17 | 2023-09-05 | 国网浙江省电力有限公司营销服务中心 | User data management method and system |
| CN115827934B (en) * | 2023-02-21 | 2023-05-09 | 四川省计算机研究院 | Enterprise portrait intelligent analysis system and method based on unified social credit code |
| CN116108086B (en) * | 2023-02-27 | 2023-09-26 | 广州汇通国信科技有限公司 | Time sequence data evaluation method and device, electronic equipment and storage medium |
| CN116452165B (en) * | 2023-03-22 | 2024-05-24 | 北京游娱网络科技有限公司 | Talent information recommendation method, service system and storage medium |
| CN116705337B (en) * | 2023-08-07 | 2023-10-27 | 山东第一医科大学第一附属医院(山东省千佛山医院) | Health data acquisition and intelligent analysis method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108399575A (en) * | 2018-01-24 | 2018-08-14 | 大连理工大学 | A kind of five-factor model personality prediction technique based on social media text |
| CN110096526A (en) * | 2019-04-30 | 2019-08-06 | 秒针信息技术有限公司 | A kind of prediction technique and prediction meanss of user property label |
| CN110442761A (en) * | 2019-06-21 | 2019-11-12 | 深圳中琛源科技股份有限公司 | A kind of user draws a portrait construction method, electronic equipment and storage medium |
| CN110490729A (en) * | 2019-08-16 | 2019-11-22 | 南京汇银迅信息技术有限公司 | A kind of financial user classification method based on user's portrait model |
| US20200125669A1 (en) * | 2018-10-17 | 2020-04-23 | Clari Inc. | Method for classifying and grouping users based on user activities |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2016362126A1 (en) * | 2015-12-02 | 2018-07-19 | Tata Consultancy Services Limited | Method and system for purchase behavior prediction of customers |
| CN111352962B (en) * | 2018-12-24 | 2024-03-29 | 网智天元科技集团股份有限公司 | Customer portrait construction method and device |
| CN110232150B (en) * | 2019-05-21 | 2023-04-14 | 平安科技（深圳）有限公司 | User data analysis method and device, readable storage medium and terminal equipment |
| CN110348879A (en) * | 2019-06-17 | 2019-10-18 | 阿里巴巴集团控股有限公司 | For determining the method and device of user behavior value |
| CN112035541A (en) * | 2020-08-28 | 2020-12-04 | 平安科技（深圳）有限公司 | Client image drawing method and device, computer readable storage medium and terminal equipment |
-
2020
- 2020-08-28 CN CN202010888273.XA patent/CN112035541A/en active Pending
- 2020-12-11 WO PCT/CN2020/135552 patent/WO2021147557A1/en active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108399575A (en) * | 2018-01-24 | 2018-08-14 | 大连理工大学 | A kind of five-factor model personality prediction technique based on social media text |
| US20200125669A1 (en) * | 2018-10-17 | 2020-04-23 | Clari Inc. | Method for classifying and grouping users based on user activities |
| CN110096526A (en) * | 2019-04-30 | 2019-08-06 | 秒针信息技术有限公司 | A kind of prediction technique and prediction meanss of user property label |
| CN110442761A (en) * | 2019-06-21 | 2019-11-12 | 深圳中琛源科技股份有限公司 | A kind of user draws a portrait construction method, electronic equipment and storage medium |
| CN110490729A (en) * | 2019-08-16 | 2019-11-22 | 南京汇银迅信息技术有限公司 | A kind of financial user classification method based on user's portrait model |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021147557A1 (en) * | 2020-08-28 | 2021-07-29 | 平安科技（深圳）有限公司 | Customer portrait method, apparatus, computer-readable storage medium, and terminal device |
| CN112580887A (en) * | 2020-12-25 | 2021-03-30 | 百果园技术(新加坡)有限公司 | Weight determination method, device and equipment for multi-target fusion evaluation and storage medium |
| CN112580887B (en) * | 2020-12-25 | 2023-12-01 | 百果园技术(新加坡)有限公司 | Weight determination method, device, equipment and storage medium for multi-target fusion evaluation |
| CN112801287A (en) * | 2021-01-26 | 2021-05-14 | 商汤集团有限公司 | Neural network performance evaluation method and device, electronic equipment and storage medium |
| CN112950225A (en) * | 2021-02-25 | 2021-06-11 | 中国工商银行股份有限公司 | Customer category determination method, device and storage medium |
| CN112598090A (en) * | 2021-03-08 | 2021-04-02 | 北京冠新医卫软件科技有限公司 | Method, device, equipment and system for health portrait |
| CN112598090B (en) * | 2021-03-08 | 2021-05-18 | 北京冠新医卫软件科技有限公司 | Method, device, equipment and system for health portrait |
| CN113064927A (en) * | 2021-03-24 | 2021-07-02 | 深圳市道通科技股份有限公司 | Client screening method and device, electronic equipment and computer readable storage medium |
| CN113033444A (en) * | 2021-03-31 | 2021-06-25 | 北京金山云网络技术有限公司 | Age estimation method and device and electronic equipment |
| CN112991110B (en) * | 2021-04-25 | 2024-02-02 | 湖南知名未来科技有限公司 | Customer type identification method of multi-dimensional portrait standard and intellectual property monitoring system |
| CN112991110A (en) * | 2021-04-25 | 2021-06-18 | 湖南知名未来科技有限公司 | Multi-dimensional portrait standard client type identification method and intellectual property monitoring system |
| CN113486041A (en) * | 2021-08-02 | 2021-10-08 | 南京邮电大学 | Client portrait management method and system based on block chain |
| CN113707296A (en) * | 2021-08-25 | 2021-11-26 | 平安国际智慧城市科技股份有限公司 | Medical treatment scheme data processing method, device, equipment and storage medium |
| CN113707296B (en) * | 2021-08-25 | 2024-04-02 | 深圳平安智慧医健科技有限公司 | Medical scheme data processing method, device, equipment and storage medium |
| CN113706019A (en) * | 2021-08-30 | 2021-11-26 | 平安银行股份有限公司 | Service capability analysis method, device, equipment and medium based on multidimensional data |
| CN113706019B (en) * | 2021-08-30 | 2024-06-07 | 平安银行股份有限公司 | Service capability analysis method, device, equipment and medium based on multidimensional data |
| CN113761134A (en) * | 2021-09-16 | 2021-12-07 | 平安国际智慧城市科技股份有限公司 | User portrait construction method and device, computer equipment and storage medium |
| CN116883048A (en) * | 2023-07-12 | 2023-10-13 | 广州朝辉智能科技有限公司 | Customer data processing method and device based on artificial intelligence and computer equipment |
| CN116883048B (en) * | 2023-07-12 | 2024-03-15 | 卓盛科技(广州)有限公司 | Customer data processing method and device based on artificial intelligence and computer equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2021147557A1 (en) | 2021-07-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112035541A (en) | Client image drawing method and device, computer readable storage medium and terminal equipment | |
| TW202022769A (en) | Risk identification model training method and device and server | |
| CN110830448B (en) | Target event flow abnormity detection method and device, electronic equipment and medium | |
| CN113011646B (en) | Data processing method, device and readable storage medium | |
| CN110796539A (en) | Credit investigation evaluation method and device | |
| Snihovyi et al. | Implementation of robo-advisor services for different risk attitude investment decisions using machine learning techniques | |
| CN111369080B (en) | Intelligent customer service solution rate prediction method and system and multi-service prediction model | |
| CN116596659A (en) | Enterprise intelligent credit approval method, system and medium based on big data wind control | |
| CN115619270A (en) | Data asset management method, device, computer equipment and storage medium | |
| CN112950347B (en) | Resource data processing optimization method and device, storage medium and terminal | |
| CN114266664A (en) | Transaction information prediction method and device, computer equipment and storage medium | |
| US20140344020A1 (en) | Competitor pricing strategy determination | |
| CN110264368A (en) | Method and device for generating emotion index of target cryptocurrency | |
| Lin et al. | Empirical of the Taiwan stock index option price forecasting model–applied artificial neural network | |
| Manzo et al. | Deep learning credit risk modeling | |
| CN115358894A (en) | Intellectual property life cycle trusteeship management method, device, equipment and medium | |
| CN116091200A (en) | Scene credit granting system and method based on machine learning, electronic equipment and medium | |
| Alvarez Diaz | Speculative strategies in the foreign exchange market based on genetic programming predictions | |
| CN115600226A (en) | Method for encrypting warehouse pledge data | |
| CN114418776A (en) | Data processing method, device, terminal equipment and medium | |
| Zhang et al. | Modeling customers' loyalty using ten years' automobile repair and maintenance data: Machine learning approaches | |
| CN114358949A (en) | Stock right amortization prediction method, device and equipment | |
| CN115293867A (en) | Financial reimbursement user portrait optimization method, device, equipment and storage medium | |
| CN115018608A (en) | Risk prediction method and device and computer equipment | |
| CN114298825A (en) | Method and device for extremely evaluating repayment volume |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201204 |