CN112214668A - Big data-based personalized financial service recommendation device and method - Google Patents
Big data-based personalized financial service recommendation device and method Download PDFInfo
- Publication number
- CN112214668A CN112214668A CN202011040637.5A CN202011040637A CN112214668A CN 112214668 A CN112214668 A CN 112214668A CN 202011040637 A CN202011040637 A CN 202011040637A CN 112214668 A CN112214668 A CN 112214668A
- Authority
- CN
- China
- Prior art keywords
- service
- user
- matrix
- data
- hidden feature
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 239000011159 matrix material Substances 0.000 claims abstract description 152
- 230000001133 acceleration Effects 0.000 claims abstract description 38
- 239000013598 vector Substances 0.000 claims description 56
- 238000004458 analytical method Methods 0.000 claims description 14
- 238000013500 data storage Methods 0.000 claims description 11
- 230000000694 effects Effects 0.000 claims description 6
- 238000013179 statistical model Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 2
- 230000006399 behavior Effects 0.000 description 89
- 230000006870 function Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 3
- 238000007405 data analysis Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000000670 limiting effect Effects 0.000 description 2
- 238000011084 recovery Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- 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
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/958—Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/06—Asset management; Financial planning or analysis
-
- 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
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Probability & Statistics with Applications (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Accounting & Taxation (AREA)
- Computational Linguistics (AREA)
- Operations Research (AREA)
- Mathematical Physics (AREA)
- Entrepreneurship & Innovation (AREA)
- Fuzzy Systems (AREA)
- Game Theory and Decision Science (AREA)
- Human Resources & Organizations (AREA)
- Software Systems (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a device and a method for recommending personalized financial services based on big data, comprising S1, acquiring user-service behavior statistical data and establishing a user-service behavior statistical matrix T; s2: receiving an instruction and setting initialization parameters; s3: constructing a uniform loss function phi, and carrying out non-negative constraint on the uniform loss function; s4: constructing a Nesterov acceleration user-service behavior recommendation model by adopting a Nesterov acceleration gradient method and training; s5: and outputting the user hidden feature matrix and the service hidden feature matrix, and restoring unknown behavior data in the user-service behavior statistical matrix. By adopting a Nesterov acceleration gradient method, the internal statistical rule of the statistical data of the known user-service behavior is analyzed with smaller computational complexity, so that the unknown behavior data of the personalized financial service based on the big data is accurately restored, and the personalized and accurate financial service is provided for the user.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a personalized financial service recommendation device and method based on big data.
Background
With the continuous development of technologies such as big data, artificial intelligence and the like, the traditional financial services with industrial era structure gradually move to the personalized financial services with scene, individuation and intelligence, such as a great amount of financial services like 'treasure on live "," profit treasure "," balance financing' and the like, and the consumption experience of users is comprehensively met. However, with the continuous expansion of financial services, it is difficult for users to select the services required by themselves from a large number of services, so the financial service industry wants to actively capture the consumption requirements of users through big data technology and machine learning, and provide more accurate services for users.
According to the historical record of the financial service system, the user-service behavior statistical matrix is a common data description structure, wherein each row pair uses one user, each column corresponds to one service, and one element of each matrix corresponds to historical behavior data of one user for one service. In financial services, users face massive services and cannot operate all services, and similarly, a service cannot be operated by all users, so that a user-service behavior statistical matrix is often extremely sparse, and unknown behaviors of users to services are far more than known behaviors.
In a financial service system, behavior data of user services often has a positive natural law, so the financial service system recommends services to users by adopting the existing non-negative matrix hidden feature analysis method, but the convergence rate of the non-negative matrix hidden feature analysis method is low, and the accuracy of data reduction is low, so that the required services cannot be recommended to the users quickly and effectively.
Disclosure of Invention
Aiming at the problem of low matching degree of users and services in the prior art, the invention provides the personalized financial service recommendation device and method based on big data, and non-negative limitation and acceleration are performed on the user-service behavior statistical matrix, so that the recommendation precision and speed are improved, and the matching degree of the users and the services is improved.
In order to achieve the purpose, the invention provides the following technical scheme:
a personalized financial service recommendation device based on big data comprises a data receiving module, a data storage module, an initialization module, a Nesterov training module and a service recommendation module; wherein,
the data receiving module is used for acquiring user-service behavior statistical data from the financial service system, establishing a user-service behavior statistical matrix and storing the user-service behavior statistical matrix in the data storage module;
the initialization module is used for preprocessing the user-service behavior statistical matrix and initializing parameters;
the Nesterov training module is used for constructing a Nesterov acceleration gradient user-service behavior statistical model according to the user-service behavior statistical data and the initialized parameters so as to accelerate and recommend the user-service behavior statistical data;
the service recommendation module is used for restoring unknown user-service behavior data through the constructed Nesterov acceleration gradient user-service behavior statistical model and recommending corresponding services to the user according to the unknown behavior data.
Preferably, the user-service behavior statistical matrix is a matrix of | U | rows and | S | columns, U represents a user set, and S represents a service set.
Preferably, the Nesterov training module comprises a Nesterov updating unit and a training unit; wherein,
a Nesterov updating unit which adopts a Nesterov acceleration gradient method to update parameters of the user hidden feature matrix X, the service hidden feature matrix Y, the user linear deviation vector B and the service linear deviation vector C;
and the training unit is used for training a user hidden feature matrix X, a service hidden feature matrix Y, a user linear deviation vector B and a service linear deviation vector C by adopting non-negative matrix hidden feature analysis.
Preferably, the service recommendation module comprises an output implicit characteristic unit and a recommendation unknown interestingness unit; wherein,
the output hidden feature unit outputs a user hidden feature matrix X and a service hidden feature matrix Y in the user-service behavior statistical data;
and the unknown interestingness recommending unit restores the unknown user-service behavior data according to the predicted user-service behavior statistical matrix T obtained by outputting the user implicit characteristic matrix X and the service implicit characteristic matrix Y in the user-service behavior statistical data.
The invention also provides a personalized financial service recommendation method based on big data, which specifically comprises the following steps:
s1, acquiring user-service behavior statistical data and establishing a user-service behavior statistical matrix T;
s2: receiving an instruction and setting initialization parameters;
s3: constructing a uniform loss function, and carrying out non-negative constraint on the uniform loss function;
s4: constructing a Nesterov acceleration user-service behavior recommendation model by adopting a Nesterov acceleration gradient method and training;
s5: and outputting the user hidden feature matrix and the service hidden feature matrix, performing predictive scoring on unknown behavior data in the user-service behavior statistical matrix, and recommending the service corresponding to the unknown behavior data of the top k to the user.
Preferably, the S1 includes:
the user set is U, the service set is S, a matrix of | U | rows and | S | columns is established as a user-service behavior statistical matrix T, the T is decomposed by using reduced matrix factorization to respectively obtain a user hidden feature matrix X and a service hidden feature matrix Y, wherein X is a matrix of | U | rows and f columns, and each row vector in X corresponds to one user and is a hidden feature vector of the user; y is a matrix of | S | rows and f columns, each row vector in Y corresponds to a service and is a hidden feature vector of the service; f is the dimension of the user implicit characteristic space and the service implicit characteristic space.
Preferably, the initialization parameters include hidden feature matrices X and Y; a hidden feature space dimension f; nesterov momentum control parameter mu and regularization factor lambda; the maximum iteration round number R; controlling a variable r by iteration rounds in the training process; the linear deviations hide the feature vectors B and C.
Preferably, in S3, the unified loss function Φ (X, Y, B, C) is expressed as:
in the formula (1), tu,sRepresenting the entity relationship between the user u and the service s, namely historical behavior data of the user u to the service s; lambda represents a known behavior data set of the user to the financial service in the user-service behavior statistical matrix T;
representing a restoration value of an unknown behavior in the user-service behavior statistical matrix; buRepresenting the characteristic value of the U line of the linear deviation of the user set U; c. CsThe characteristic value of the S line of the linear deviation representing the service set S; x is the number ofu,kRepresenting the characteristic value of the kth row and the kth column of the user hidden characteristic matrix X; y iss,kRepresenting the characteristic value of the kth row and the kth column of the service hidden characteristic matrix Y; λ represents the canonical control parameter and f represents the dimension of the implicit feature space.
Preferably, the S4 includes the following steps:
s4-1: during the first iteration, a non-negative matrix hidden feature analysis method is adopted to train a user hidden feature matrix X, a service hidden feature matrix Y, a user linear deviation vector B and a service linear deviation vector C, and the training formula is as follows:
in formula (2), | Λ (u) | represents a service set in the user-service known behavior data related to the user u, and | Λ(s) | represents a user set in the user-service known behavior data related to the service s;
s4-2: updating a user hidden feature matrix X, a service hidden feature matrix Y, a user linear deviation vector B and a service linear deviation vector C by adopting a Nesterov acceleration gradient method, wherein the updating formula is as follows:
in the formula (3), the first and second groups,
denotes buIn the state at the time t, the state,
indicates updated at time b of t +1u (t+1)The intermediate state value mu is a Nesterov control parameter and controls the acceleration effect, the Nesterov acceleration gradient method is only added for the second time in the training process, namely t is more than or equal to 2, and the iteration times are represented;
s4-3: after updating according to non-negative matrix hidden feature analysis method
Training is carried out, and the training formula is as follows:
trained parameters
Indicating the use of intermediate states
The state value at the moment t +1 after updating;
s4-4: and judging whether the training reaches an iteration termination condition, if so, finishing the iteration training, and if not, continuing the training and updating until the iteration termination condition is reached.
Preferably, the S5 includes the following steps:
s5-1: acquiring hidden features corresponding to a user u, a service s, a user linear deviation b and a service linear deviation c from an output hidden feature unit;
s5-2: the inner product of the vector of the user u to the user hidden feature matrix X and the vector of the service hidden feature matrix Y corresponding to the service s, the user linear deviation b and the service linear deviation c is used as the unknown behavior data value of the user u to the service s
Namely, it is
In summary, due to the adoption of the technical scheme, compared with the prior art, the invention at least has the following beneficial effects:
according to the invention, a Nesterov acceleration gradient method is adopted, and the internal statistical rule of the statistical data of the known user-service behavior is analyzed with smaller computational complexity, so that the unknown behavior data of the personalized financial service based on big data is accurately restored, and the corresponding personalized and accurate financial service is provided for the user; the invention also realizes that: (1) the user-service behavior statistical matrix is subjected to non-negative limitation by a non-negative matrix hidden feature analysis method, so that the recommendation accuracy is improved, and the user features are better ensured; (2) the recommendation precision is improved by a Nesterov acceleration gradient method; (3) through the Nesterov acceleration gradient method, the recommendation speed is accelerated, the time cost is saved, and the matching degree of the user and the service is improved.
Description of the drawings:
fig. 1 is a schematic diagram of a personalized financial service recommendation device based on big data according to an exemplary embodiment of the invention.
Fig. 2 is a schematic diagram of a method for recommending personalized financial services based on big data according to an exemplary embodiment of the present invention.
Fig. 3 is a user-service behavior statistics analysis time versus diagram illustration according to an exemplary embodiment of the present invention.
FIG. 4 is a schematic comparison of RMSE according to an exemplary embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to examples and embodiments. It should be understood that the scope of the above-described subject matter is not limited to the following examples, and any techniques implemented based on the disclosure of the present invention are within the scope of the present invention.
In the description of the present invention, it is to be understood that the terms "longitudinal", "lateral", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like, indicate orientations or positional relationships based on those shown in the drawings, and are used merely for convenience of description and for simplicity of description, and do not indicate or imply that the referenced devices or elements must have a particular orientation, be constructed in a particular orientation, and be operated, and thus, are not to be construed as limiting the present invention.
As shown in fig. 1, the present invention provides a personalized financial service recommendation device based on big data, which includes a data receiving module 10, a data storage module 20, an initialization module 30, a nersterov training module 40, and a service recommendation module 50. The output end of the data receiving module 10 is connected with the input end of the data storage module 20, the output end of the data storage module 20 is connected with the input end of the initialization module 30, the output end of the initialization module 30 is connected with the input end of the Nesterov training module 40, and the output end of the Nesterov training module 40 is connected with the input end of the service recommendation module 50. Wherein,
and the data receiving module 10 is used for acquiring the user-service behavior statistical data from the financial service system and establishing a user-service behavior statistical matrix, and storing the user-service behavior statistical matrix into the data storage module 20.
In the obtained user-service behavior statistical data, the user set is recorded as U, the service set is recorded as S, a matrix of | U | rows and | S | columns is established as a user-service behavior statistical matrix T, and the user-service behavior statistical matrix T is stored in the data storage module 20. Decomposing T by using a reduced matrix factorization to respectively obtain a user hidden feature matrix X and a service hidden feature matrix Y, wherein X is a matrix of | U | rows and f columns, and each row vector in X corresponds to a user and is a hidden feature vector of the user; y is a matrix of | S | rows and f columns, each row vector in Y corresponds to a service and is a hidden feature vector of the service; f is the dimension of the user implicit characteristic space and the service implicit characteristic space; meanwhile, the user hidden feature matrix X and the service hidden feature matrix Y are stored in the data storage module 20.
And the initialization module 30 is used for initializing parameters in the user-service behavior characteristic recommendation process.
Initialized parameters comprise hidden feature matrixes X and Y of users and services; a hidden feature space dimension f; nesterov momentum control parameter mu and regularization factor lambda; the maximum iteration round number R; controlling a variable r by iteration rounds in the training process; the linear deviations hide the feature vectors B and C. Wherein f determines the characteristic space dimension of each hidden characteristic matrix and initializes the hidden characteristic matrixes to positive integers; in latent feature matrices X and Y: x is a hidden feature matrix of | U | row f column, Y is a hidden feature matrix of | S | row f column, and random smaller positive numbers are respectively used for initialization; linear deviation hidden feature vectors B and C: b is a vector of | U | rows, C is a vector of | S | rows, and random smaller positive numbers are respectively used for initialization; the maximum training iteration round number R is a variable for controlling the upper limit of the training times and is initialized to be a larger positive integer; initializing an iteration round number control variable r to be 0; the regularization factor λ is a measure L2The restriction effect of the regularization item on the recommendation model is initialized to a smaller positive number; and the Nesterov momentum control parameter mu is initialized to be a positive number close to 1 in order to control the acceleration effect of the Nesterov momentum on the model and ensure that the data recovery accuracy of the recommended model on the unknown behavior data is highest.
And the Nesterov training module 40 is used for constructing a Nesterov acceleration gradient user-service behavior statistical model according to the user-service behavior statistical data and the initialized parameters so as to accelerate and recommend the user-service behavior statistical data.
In this embodiment, the Nesterov training module 40 includes a Nesterov updating unit and a training unit.
And the Nesterov updating unit is used for updating parameters of the user hidden feature matrix X, the service hidden feature matrix Y, the user linear deviation vector B and the service linear deviation vector C by adopting a Nesterov acceleration gradient method and updating corresponding hidden features in a single element mode.
And the training unit is used for training a user hidden feature matrix X, a service hidden feature matrix Y, a user linear deviation vector B and a service linear deviation vector C by adopting non-negative matrix hidden feature analysis, and updating corresponding hidden features in a unit element mode.
And the service recommendation module 50 is used for restoring the unknown user-service behavior data through the constructed Nesterov acceleration gradient user-service behavior statistical model and outputting the restored user-service unknown behavior data and the hidden feature matrix of the user and the service.
In this embodiment, the service recommendation module 50 includes an output implicit feature unit and a recommendation unknown interestingness unit.
And the output hidden feature unit is used for outputting a user hidden feature matrix X and a service hidden feature matrix Y in the user-service behavior statistical data.
And the unknown interestingness recommending unit restores the unknown user-service behavior data according to the predicted user-service behavior statistical matrix T ^ obtained by outputting the user implicit characteristic matrix X and the service implicit characteristic matrix Y in the user-service behavior statistical data.
Based on the above device, as shown in fig. 2, the invention further provides a personalized financial service recommendation method based on big data, which specifically includes the following steps:
s1, acquiring user-service behavior statistical data and establishing a user-service behavior statistical matrix T, wherein for each element in the matrix, a row represents a corresponding user and a column represents a corresponding service;
in the obtained user-service behavior statistical data, the user set is recorded as U, the service set is recorded as S, a matrix of | U | rows and | S | columns is established as a user-service behavior statistical matrix T, and the user-service behavior statistical matrix T is stored in the data storage module 20. Decomposing T by using a reduced matrix factorization to respectively obtain a user hidden feature matrix X and a service hidden feature matrix Y, wherein X is a matrix of | U | rows and f columns, and each row vector in X corresponds to a user and is a hidden feature vector of the user; y is a matrix of | S | rows and f columns, each row vector in Y corresponds to a service and is a hidden feature vector of the service; f is the dimension of the user implicit characteristic space and the service implicit characteristic space.
S2: receiving an instruction and setting initialization parameters;
in this embodiment, the initialized parameters include:
user-service known behavior data tu,sThe entity relationship between the user u and the service s is represented as historical behavior data of the user u to the service s;
a linear deviation hidden feature vector B which represents a linear deviation feature vector of the user set U;
a linear deviation hidden feature vector C representing a linear deviation feature vector of the service set S;
a hidden feature matrix X which represents a user hidden feature matrix after factorization of a user-service behavior statistical data matrix by adopting a reduction matrix;
the hidden feature matrix Y represents a service hidden feature matrix after a user-service behavior statistical data matrix is factorized by adopting a reduction matrix;
the regular control parameter lambda is used for measuring the limiting effect of the regular term on the model and enhancing the robustness of the model;
and the Nesterov momentum control parameter mu is used for controlling the acceleration effect of the Nesterov momentum on the model, so that the data recovery accuracy of the recommended model on the unknown behavior data is highest.
S3: constructing a uniform loss function phi (X, Y, B and C), and carrying out non-negative constraint on the uniform loss function to ensure the non-negativity of the user and service hidden feature matrix in the training process;
in the formula (1), tu,sRepresenting the entity relationship between the user u and the service s, namely historical behavior data of the user u to the service s; lambda represents a known behavior data set of the user to the financial service in the user-service behavior statistical matrix T;
representing a restoration value of an unknown behavior in the user-service behavior statistical matrix; buRepresenting the characteristic value of the U line of the linear deviation of the user set U; c. CsThe characteristic value of the S line of the linear deviation representing the service set S; x is the number ofu,kRepresenting the characteristic value of the kth row and the kth column of the user hidden characteristic matrix X; y iss,kRepresenting the characteristic value of the kth row and the kth column of the service hidden characteristic matrix Y; λ represents the canonical control parameter and f represents the dimension of the implicit feature space.
S4: and constructing a Nesterov acceleration user-service behavior recommendation model by adopting a Nesterov acceleration gradient method and training.
In this embodiment, the construction expression of the Nesterov accelerated user-service behavior recommendation model is expressed by a uniform loss function Φ (X, Y, B, C).
S4-1: during the first iteration, namely t is 1, a non-negative matrix hidden feature analysis method is adopted to train a user hidden feature matrix X, a service hidden feature matrix Y, a user linear deviation vector B and a service linear deviation vector C, and the training formula is as follows:
in equation (2), | Λ (u) | represents a set of services in the user-service known behavior data related to user u, and | Λ(s) | represents a set of users in the user-service known behavior data related to service s.
S4-2: a Nesterov acceleration gradient method is added to accelerate a Nesterov acceleration user-service behavior recommendation model, firstly, a Nesterov acceleration gradient method is adopted to update a user hidden feature matrix X, a service hidden feature matrix Y, a user linear deviation vector B and a service linear deviation vector C, and the updating formula is as follows:
in the formula (3), the first and second groups,
represents buIn the state at the time t, the state,
indicates that the data is updated at time b of t +1u (t+1)Mu is a Nesterov control parameter and controls the acceleration effect. The Nesterov acceleration gradient method is only added for the second time in the training process, namely t is more than or equal to 2, and the iteration times are represented.
S4-3: after updating according to non-negative matrix hidden feature analysis method
Training is carried out, and the training formula is as follows:
trained parameters
Indicating the use of intermediate states
The state value at time t +1 after the update.
S4-4: and judging whether the training reaches an iteration termination condition, if so, finishing the iteration training, and if not, continuing the training and updating until the iteration termination condition is reached.
And (3) iteration termination conditions: if the iteration round number control variable R exceeds the training iteration round number upper limit R, the iteration training is finished; if the iteration round number control variable R is smaller than the upper limit R of the training iteration round number (R can take a value of 1000), the training and the updating are continued until an iteration termination condition is reached.
S5: outputting a user hidden feature matrix and a service hidden feature matrix, predicting unknown behavior data in the user-service behavior statistical matrix, and predicting the unknown behavior data according to the predicted unknown behavior data
Recommending services, i.e. unknown behavioural data
A higher value of (c) indicates a higher probability that the user will use the service, thereby servicing the service (depending on unknown behavior data)
Value) to the interested user.
S5-1: and acquiring the hidden features corresponding to the user u, the service s, the user linear deviation b and the service linear deviation c from the output hidden feature unit.
S5-2: the inner product of the vector of the user u to the user hidden feature matrix X and the vector of the service hidden feature matrix Y corresponding to the service s, the user linear deviation b and the service linear deviation c is used as the unknown behavior data value of the user u to the service s
Namely, it is
buRepresenting the characteristic value of the U line of the linear deviation of the user set U; c. CsThe characteristic value of the S line of the linear deviation representing the service set S; x is the number ofu,kRepresenting the characteristic value of the kth row and the kth column of the user hidden characteristic matrix X; y iss,kAnd the characteristic value of the kth row and the kth column of the service hidden characteristic matrix Y is represented.
In this embodiment, fig. 3 is a comparison graph of data analysis time of a recommendation model before and after applying the recommendation apparatus and method of the present invention. As can be seen from fig. 3, after the recommendation apparatus and method of the present invention are applied, the execution time for the statistical data of user-service behavior is much less than the case of not applying the recommendation apparatus and method of the present invention. Specifically, as can be seen from fig. 3, the model without using the recommendation apparatus and method of the present invention has a performance time about 7 times that of the recommendation apparatus and method of the present invention. Namely, after the recommendation device and the recommendation method are used, the execution time of the model is improved by 7 times compared with the original execution time, and the operation efficiency of the model is greatly improved.
Fig. 4 is a comparison graph of RMSE during data analysis before and after applying the recommendation apparatus and method of the present invention, where RMSE (root mean square error) is a measure of recommendation error, and the smaller RMSE, the higher the accuracy. As can be seen from FIG. 4, the accuracy of the user personalized service selection in the financial service is greatly improved by applying the recommendation device and method of the present invention. In practical application, the service can better improve personalized requirements for users.
According to the device and the method for recommending the personalized financial service based on the big data, a Nesterov acceleration gradient method is adopted, and the internal statistical rule of the statistical data of the known user-service behavior is analyzed with low computational complexity, so that the unknown behavior data of the personalized financial service based on the big data is accurately restored, and the personalized and accurate financial service is provided for the user; the invention also realizes that: (1) the user-service behavior statistical matrix is subjected to non-negative limitation by a non-negative matrix hidden feature analysis method, so that the recommendation accuracy is ensured, and the user features are better ensured; (2) the recommendation precision of the recommendation model is improved by a Nesterov acceleration gradient method; (3) the calculation speed of the recommendation model is accelerated by a Nesterov acceleration gradient method.
It will be understood by those of ordinary skill in the art that the foregoing embodiments are specific examples for carrying out the invention, and that various changes in form and details may be made therein without departing from the spirit and scope of the invention in practice.
Claims (10)
1. A personalized financial service recommendation device based on big data is characterized by comprising a data receiving module, a data storage module, an initialization module, a Nesterov training module and a service recommendation module; wherein,
the data receiving module is used for acquiring user-service behavior statistical data from the financial service system, establishing a user-service behavior statistical matrix and storing the user-service behavior statistical matrix in the data storage module;
the initialization module is used for preprocessing the user-service behavior statistical matrix and initializing parameters;
the Nesterov training module is used for constructing a Nesterov acceleration gradient user-service behavior statistical model according to the user-service behavior statistical data and the initialized parameters so as to accelerate and recommend the user-service behavior statistical data;
the service recommendation module is used for restoring unknown user-service behavior data through the constructed Nesterov acceleration gradient user-service behavior statistical model and recommending corresponding services to the user according to the unknown behavior data.
2. The apparatus as claimed in claim 1, wherein the statistical matrix of user-service behavior is a matrix of | U | rows and | S | columns, where U represents a user set and S represents a service set.
3. The big-data-based personalized financial service recommendation device of claim 1, wherein the Nesterov training module comprises a Nesterov update unit and a training unit; wherein,
a Nesterov updating unit which adopts a Nesterov acceleration gradient method to update parameters of the user hidden feature matrix X, the service hidden feature matrix Y, the user linear deviation vector B and the service linear deviation vector C;
and the training unit is used for training a user hidden feature matrix X, a service hidden feature matrix Y, a user linear deviation vector B and a service linear deviation vector C by adopting non-negative matrix hidden feature analysis.
4. The big-data-based personalized financial service recommendation device according to claim 1, wherein the service recommendation module comprises an output implicit feature unit and a recommendation unknown interestingness unit; wherein,
the output hidden feature unit outputs a user hidden feature matrix X and a service hidden feature matrix Y in the user-service behavior statistical data;
and the unknown interestingness recommending unit restores the unknown user-service behavior data according to the predicted user-service behavior statistical matrix T obtained by outputting the user implicit characteristic matrix X and the service implicit characteristic matrix Y in the user-service behavior statistical data.
5. A personalized financial service recommendation method based on big data is characterized by comprising the following steps:
s1, acquiring user-service behavior statistical data and establishing a user-service behavior statistical matrix T;
s2: receiving an instruction and setting initialization parameters;
s3: constructing a uniform loss function, and carrying out non-negative constraint on the uniform loss function;
s4: constructing a Nesterov acceleration user-service behavior recommendation model by adopting a Nesterov acceleration gradient method and training;
s5: and outputting the user hidden feature matrix and the service hidden feature matrix, performing predictive scoring on unknown behavior data in the user-service behavior statistical matrix, and recommending the service corresponding to the unknown behavior data of the top k to the user.
6. The method for recommending personalized financial services based on big data according to claim 5, wherein said S1 comprises:
the user set is U, the service set is S, a matrix of | U | rows and | S | columns is established as a user-service behavior statistical matrix T, the T is decomposed by using reduced matrix factorization to respectively obtain a user hidden feature matrix X and a service hidden feature matrix Y, wherein X is a matrix of | U | rows and f columns, and each row vector in X corresponds to one user and is a hidden feature vector of the user; y is a matrix of | S | rows and f columns, each row vector in Y corresponds to a service and is a hidden feature vector of the service; f is the dimension of the user implicit characteristic space and the service implicit characteristic space.
7. The method of claim 5, wherein the initialization parameters comprise hidden feature matrices X and Y; a hidden feature space dimension f; nesterov momentum control parameter mu and regularization factor lambda; the maximum iteration round number R; controlling a variable r by iteration rounds in the training process; the linear deviations hide the feature vectors B and C.
8. The method as claimed in claim 5, wherein in the S3, the uniform loss function Φ (X, Y, B, C) is expressed as:
in the formula (1), tu,sRepresenting the entity relationship between the user u and the service s, namely historical behavior data of the user u to the service s; lambda represents a known behavior data set of the user to the financial service in the user-service behavior statistical matrix T;
representing a restoration value of an unknown behavior in the user-service behavior statistical matrix; buRepresenting the characteristic value of the U line of the linear deviation of the user set U; c. CsThe characteristic value of the S line of the linear deviation representing the service set S; x is the number ofu,kRepresenting the characteristic value of the kth row and the kth column of the user hidden characteristic matrix X; y iss,kRepresenting the characteristic value of the kth row and the kth column of the service hidden characteristic matrix Y; λ represents the canonical control parameter and f represents the dimension of the implicit feature space.
9. The method for recommending personalized financial services based on big data according to claim 5, wherein said S4 comprises the steps of:
s4-1: during the first iteration, a non-negative matrix hidden feature analysis method is adopted to train a user hidden feature matrix X, a service hidden feature matrix Y, a user linear deviation vector B and a service linear deviation vector C, and the training formula is as follows:
in formula (2), | Λ (u) | represents a service set in the user-service known behavior data related to the user u, and | Λ(s) | represents a user set in the user-service known behavior data related to the service s;
s4-2: updating a user hidden feature matrix X, a service hidden feature matrix Y, a user linear deviation vector B and a service linear deviation vector C by adopting a Nesterov acceleration gradient method, wherein the updating formula is as follows:
in the formula (3), the first and second groups,
denotes buIn the state at the time t, the state,
indicates updated at time b of t +1u (t+1)The intermediate state value mu is a Nesterov control parameter and controls the acceleration effect, the Nesterov acceleration gradient method is only added for the second time in the training process, namely t is more than or equal to 2, and the iteration times are represented;
s4-3: after updating according to non-negative matrix hidden feature analysis method
Training is carried out, and the training formula is as follows:
trained parameters
Indicating the use of intermediate states
The state value at the moment t +1 after updating;
s4-4: and judging whether the training reaches an iteration termination condition, if so, finishing the iteration training, and if not, continuing the training and updating until the iteration termination condition is reached.
10. The method for recommending personalized financial services based on big data according to claim 5, wherein said S5 comprises the steps of:
s5-1: acquiring hidden features corresponding to a user u, a service s, a user linear deviation b and a service linear deviation c from an output hidden feature unit;
s5-2: the inner product of the vector of the user u to the user hidden feature matrix X and the vector of the service hidden feature matrix Y corresponding to the service s, the user linear deviation b and the service linear deviation c is used as the unknown behavior data value of the user u to the service s
Namely, it is
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011040637.5A CN112214668B (en) | 2020-09-28 | 2020-09-28 | Personalized financial service recommendation device and method based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011040637.5A CN112214668B (en) | 2020-09-28 | 2020-09-28 | Personalized financial service recommendation device and method based on big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112214668A true CN112214668A (en) | 2021-01-12 |
| CN112214668B CN112214668B (en) | 2023-06-02 |
Family
ID=74051477
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011040637.5A Active CN112214668B (en) | 2020-09-28 | 2020-09-28 | Personalized financial service recommendation device and method based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112214668B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115599598A (en) * | 2022-10-08 | 2023-01-13 | 国网江苏省电力有限公司南通供电分公司（Cn） | Power load sensing data recovery method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170228433A1 (en) * | 2016-02-04 | 2017-08-10 | Microsoft Technology Licensing, Llc | Method and system for diverse set recommendations |
| CN110008404A (en) * | 2019-03-22 | 2019-07-12 | 成都理工大学 | Enigmatic language justice model optimization method based on the optimization of NAG momentum |
| CN111563203A (en) * | 2020-05-08 | 2020-08-21 | 深圳市万佳安人工智能数据技术有限公司 | Intelligent household user-service interest degree personalized prediction device and method based on rapid non-negative implicit characteristic analysis |
-
2020
- 2020-09-28 CN CN202011040637.5A patent/CN112214668B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170228433A1 (en) * | 2016-02-04 | 2017-08-10 | Microsoft Technology Licensing, Llc | Method and system for diverse set recommendations |
| CN110008404A (en) * | 2019-03-22 | 2019-07-12 | 成都理工大学 | Enigmatic language justice model optimization method based on the optimization of NAG momentum |
| CN111563203A (en) * | 2020-05-08 | 2020-08-21 | 深圳市万佳安人工智能数据技术有限公司 | Intelligent household user-service interest degree personalized prediction device and method based on rapid non-negative implicit characteristic analysis |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115599598A (en) * | 2022-10-08 | 2023-01-13 | 国网江苏省电力有限公司南通供电分公司（Cn） | Power load sensing data recovery method and device |
| CN115599598B (en) * | 2022-10-08 | 2023-08-15 | 国网江苏省电力有限公司南通供电分公司 | Power load sensing data recovery method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112214668B (en) | 2023-06-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kumar et al. | Pruning filters with L1-norm and capped L1-norm for CNN compression | |
| US10304008B2 (en) | Fast distributed nonnegative matrix factorization and completion for big data analytics | |
| Alain et al. | Variance reduction in sgd by distributed importance sampling | |
| CN110929164A (en) | Interest point recommendation method based on user dynamic preference and attention mechanism | |
| US20210256375A1 (en) | Reduced computation real time recurrent learning | |
| CN112052948B (en) | Network model compression method and device, storage medium and electronic equipment | |
| CN109543112A (en) | A kind of sequence of recommendation method and device based on cyclic convolution neural network | |
| CN111178986B (en) | User-commodity preference prediction method and system | |
| Dai et al. | Hybrid deep model for human behavior understanding on industrial internet of video things | |
| CN111311324B (en) | User-commodity preference prediction system and method based on stable neural collaborative filtering | |
| CN111563203A (en) | Intelligent household user-service interest degree personalized prediction device and method based on rapid non-negative implicit characteristic analysis | |
| CN113641907B (en) | Super-parameter self-adaptive depth recommendation method and device based on evolutionary algorithm | |
| CN112214668A (en) | Big data-based personalized financial service recommendation device and method | |
| CN114693993A (en) | Image processing and image classification method, device, equipment and storage medium | |
| CN114861671A (en) | Model training method and device, computer equipment and storage medium | |
| Liu et al. | High-performance tensor learning primitives using GPU tensor cores | |
| CN110633417A (en) | Web service recommendation method and system based on service quality | |
| US20230342626A1 (en) | Model processing method and related apparatus | |
| Yang et al. | Learning to generalize provably in learning to optimize | |
| CN113485107B (en) | Reinforced learning robot control method and system based on consistency constraint modeling | |
| CN113762737B (en) | Method and system for predicting network service quality | |
| CN113570037A (en) | Neural network compression method and device | |
| Du et al. | Evolutionary NAS in light of model stability for accurate continual learning | |
| CN112037850B (en) | Momentum acceleration-based device and method for predicting interaction between missing proteins | |
| CN112037849B (en) | Device and method for predicting protein-protein interaction based on alternative direction multiplier method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |