CN113988909A

CN113988909A - Integration processing method and device, computer equipment and storage medium

Info

Publication number: CN113988909A
Application number: CN202111205690.0A
Authority: CN
Inventors: 刘恒
Original assignee: Industrial and Commercial Bank of China Ltd ICBC
Current assignee: Industrial and Commercial Bank of China Ltd ICBC
Priority date: 2021-10-15
Filing date: 2021-10-15
Publication date: 2022-01-28

Abstract

The application relates to an integral processing method, an integral processing device, computer equipment and a storage medium. The method comprises the following steps: the method comprises the steps of obtaining integral use data of a user in a preset time period, obtaining values of integral indexes according to the integral use data, and determining integral use activity of the user according to the values of the integral indexes and coefficients of the integral indexes. By adopting the method, the integral use condition can be comprehensively and reasonably measured.

Description

Integration processing method and device, computer equipment and storage medium

Technical Field

The present application relates to the field of big data processing technologies, and in particular, to an integration processing method and apparatus, a computer device, and a storage medium.

Background

With the improvement of the living standard of people, the consumption modes are more and more diversified, and merchants can feed back users in a mode of issuing points. The merchant can guide the user to consume for a long time by researching the credit use condition of the user, and the stickiness of the user is enhanced.

In the prior art, the point use condition of a user is mainly reflected by researching a certain index of the point of the user. Therefore, the method in the prior art cannot comprehensively and reasonably measure the use condition of the integral.

Disclosure of Invention

In view of the above, it is necessary to provide an integration processing method, an apparatus, a computer device and a storage medium for comprehensively and reasonably measuring the use condition of the integration.

A method of integration processing, the method comprising:

acquiring integral use data of a user in a preset time period;

acquiring a value of an integral index according to the integral use data; the integral index comprises at least two of integral using quantity, integral using frequency and integral using duration, and the integral using duration is the duration from the moment when a user uses the integral for the last time to the current moment;

and determining the integral use activity of the user according to the value of each integral index and the coefficient of each integral index.

In one embodiment, the method further comprises the following steps:

acquiring an initial coefficient of each integral index;

and acquiring the coefficient of each integral index according to the initial coefficient of each integral index.

In one embodiment, the obtaining the coefficient of each integral indicator according to the initial coefficient of each integral indicator includes:

acquiring a first weight matrix of the initial coefficient of each integral index according to the initial coefficient of each integral index by adopting an analytic hierarchy process;

and calculating the coefficient of each integral index according to the first weight matrix.

In one embodiment, the calculating the coefficient of each integral indicator according to the first weight matrix includes:

calculating the proportion of each element in the first weight matrix in the column to obtain a second weight matrix;

and summing each row of the second weight matrix, and normalizing the summation result of each row to obtain the coefficient of each integral index.

In one embodiment, the method further comprises the following steps:

calculating a second maximum characteristic root according to the first maximum characteristic root of the first weight matrix, the matrix order of the first weight matrix and the coefficient of each integral index;

and performing consistency verification on the coefficient of each integral index according to the second maximum characteristic root.

In one embodiment, the performing consistency verification on the coefficients of the integral indexes according to the second maximum feature root includes:

determining a value of a random consistency index according to the second maximum characteristic root and the matrix order of the first weight matrix;

determining a random consistency ratio according to the value of the random consistency index and a reference value of the random consistency index;

and if the random consistency ratio is smaller than a preset threshold value, determining that the consistency verification of the coefficient of each integral index passes.

In one embodiment, the obtaining the value of the integral indicator according to the integral usage data includes:

acquiring the integral use duration, the integral use quantity of at least two different time periods and the integral use frequency of the at least two different time periods according to the integral use data;

and acquiring the value of each integral index according to the integral use duration, the integral use number of the at least two different time periods, the integral use frequency of the at least two different time periods and the weight of each time period.

In one embodiment, the obtaining the value of each integral indicator according to the integral usage duration, the integral usage number of the at least two different time periods, the integral usage frequency of the at least two different time periods, and the weight of each time period includes:

respectively performing first preprocessing on the integral use duration, the integral use number of at least two different time periods and the integral use frequency of at least two different time periods to obtain the processed integral use duration, the processed integral use number of each time period and the processed integral use frequency; the first pre-processing includes at least one of normalization processing and normalization processing;

weighting and summing the processed integral use number of each time period and the weight of each time period to obtain a value of the integral use number;

weighting and summing the processed integral use frequency of each time segment and the weight of each time segment to obtain a value of the integral use frequency;

and taking the processed integral use time length as the value of the integral use time length.

In one embodiment, the obtaining of the point usage data of the user within the preset time period includes:

acquiring original point use data of a user in a preset time period from a database;

performing second preprocessing on the original integral using data to obtain the integral using data; the second preprocessing includes padding missing values to zero and/or clearing singular values.

An integration processing apparatus, the apparatus comprising:

the system comprises a first acquisition module, a second acquisition module and a control module, wherein the first acquisition module is used for acquiring integral use data of a user within a preset time period;

the second acquisition module is used for acquiring the value of the integral index according to the integral use data; the integral index comprises at least two of integral using quantity, integral using frequency and integral using duration, and the integral using duration is the duration from the moment when a user uses the integral for the last time to the current moment;

and the determining module is used for determining the integral use activity of the user according to the value of each integral index and the coefficient of each integral index.

A computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:

acquiring integral use data of a user in a preset time period;

A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:

acquiring integral use data of a user in a preset time period;

According to the integral processing method, the integral processing device, the computer equipment and the storage medium, the integral using data of the user in the preset time period is obtained, the value of the integral index is obtained according to the integral using data, and the integral using activity of the user is determined according to the value of each integral index and the coefficient of each integral index. In the embodiment, the integral use activity of the user can be comprehensively and reasonably evaluated by acquiring the values of the at least two integral indexes and the coefficients of the at least two integral indexes.

Drawings

FIG. 1 is a diagram of an embodiment of an application environment of a method for integral processing;

FIG. 2 is a flow diagram illustrating a method of integration processing according to one embodiment;

FIG. 3 is a schematic flow chart showing integration processing steps in another embodiment;

FIG. 4 is a schematic flow chart diagram illustrating an integration processing method according to another embodiment;

FIG. 5 is a flow chart illustrating an integration processing method according to another embodiment;

FIG. 6 is a flow chart illustrating an integration processing method according to another embodiment;

FIG. 7 is a flow chart illustrating an integration processing method according to still another embodiment;

FIG. 8 is a flow chart illustrating an integration processing method according to still another embodiment;

FIG. 9 is a schematic flow chart diagram illustrating an integration processing method according to yet another embodiment;

FIG. 10 is a flowchart illustrating an integration processing method according to another embodiment;

FIG. 11 is a flowchart illustrating an integration processing method according to another embodiment;

FIG. 12 is a flowchart illustrating an integration processing method according to still another embodiment;

FIG. 13 is a block diagram showing the structure of an integration processing device according to an embodiment;

fig. 14 is a block diagram showing the construction of an integration processing device in another embodiment;

FIG. 15 is a block diagram showing the construction of an integration processing device in still another embodiment;

FIG. 16 is a block diagram showing the construction of an integration processing device in still another embodiment;

FIG. 17 is a block diagram showing the construction of an integration processing device in still another embodiment;

FIG. 18 is a diagram illustrating an internal structure of a computer device according to an embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

The integration processing method provided by the application can be applied to the application environment shown in fig. 1. The server includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the server is configured to provide computing and control capabilities. The memory of the server comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing information data related to the stationary user. The network interface of the server is used for communicating with an external terminal through network connection. The server firstly obtains the integral use data of the user in a preset time period, then obtains the value of the integral index according to the integral use data, and finally determines the integral use activity of the user according to the value of each integral index and the coefficient of each integral index. The server may be implemented by an independent server or a server cluster composed of a plurality of servers.

In one embodiment, as shown in fig. 2, there is provided an integration processing method, which is described by taking the application of the method to the server in fig. 1 as an example, and includes the following steps:

s201, obtaining integral use data of a user in a preset time period.

The point usage data may be a detailed description of the point history data of the client, for example, the point usage data may include the number of points used by the user in a certain history time period, the frequency of point usage, the time length of point usage, the specific time of point usage, and the like.

In this embodiment, the server may obtain the point use data of each user within the preset time by extracting the keyword, or may directly obtain the point use data of each user within the preset time by using a network. The preset time period is set manually according to historical experience, for example, the preset time may be set to last one month, three months, or half a year. The corresponding preset time period is one month, three months and half a year. For example, the server obtains point history data details of N clients from the database as point usage data, which may be expressed as X ═ X₁,X₂,…,X_iAnd f, wherein i is greater than or equal to 1 and less than or equal to N, each integral historical data detail comprises an integral use number and an integral use time, the integral use numbers are summarized day by day, integral use frequency is calculated based on time, and the like, which are not limited in the embodiment of the application.

S202, acquiring a value of an integral index according to integral use data; the integral index comprises at least two of integral use number, integral use frequency and integral use duration, and the integral use duration is the duration from the moment when the user uses the integral for the last time to the current moment.

In this embodiment, after obtaining the integral use data in step S201, the integral use data may be classified by a naive bayes classification algorithm according to the integral use data to obtain data of an integral index corresponding to each integral index, and a value of the integral index is determined according to the data of each integral index. Optionally, data values in multiple time periods may be obtained for each integral indicator, and then the data values in multiple time periods are averaged to obtain the value of each integral indicator. Optionally, for each integral index, the data values of multiple time periods may be weighted and averaged to obtain the value of each integral index. For example, the integral use number and the integral use frequency may be selected as the integral indexes, the integral use frequency and the integral use duration may be selected as the integral indexes, the integral use number and the integral use duration may be selected as the integral indexes, and the integral use number, the integral use frequency and the integral use duration may also be selected as the integral indexes. For example, when the number of integration uses, the frequency of integration uses, and the time period of integration uses are selected as the integration indices, the number of integration uses is represented by Q, the frequency of integration uses is represented by F, the time period of integration uses is represented by R, and each integration index P is represented by P ═ Q, F, R }, the data vector of each integration index can be represented as P ═ Q, F, R }.

And S203, determining the integral use activity of the user according to the value of each integral index and the coefficient of each integral index.

The coefficient of each integral index indicates the degree of importance of each integral index.

In the present embodiment, the expert may determine the coefficient of each integral index according to the importance degree of each integral index. Alternatively, the value of each integral index may be obtained in step S202, and the coefficient of each integral index may be calculated, or alternatively, the coefficient of each integral index may be determined by the value of the historical integral index. Optionally, the expert may score each integral index, and obtain the coefficient of each integral index through a corresponding algorithm according to the scoring result. In this embodiment, the method for obtaining the coefficient of each integral index is not limited, and only the coefficient of each integral index needs to be obtained. Each integral index corresponds to the value of each integral index and the coefficient of each integral index, the value of each integral index is multiplied by the coefficient of each integral index to obtain a new value of each integral index, and the new values of each integral index are added to determine the integral use activity of the user. For example, if the coefficient of the number of integration uses is 0.5, the coefficient of the frequency of integration uses is 0.3, and the coefficient of the time period of integration uses is 0.2, the degree of activity of integration uses is 0.5R +0.3F + 0.2Q.

In the integral processing method provided in this embodiment, the integral usage data of the user in the preset time period is obtained, the value of each integral index is obtained according to the integral usage data, and the integral usage activity of the user is determined according to the value of each integral index and the coefficient of each integral index. In the embodiment, the value of at least two integral indexes is obtained, and the integral use activity of the user can be comprehensively, reasonably and accurately evaluated according to the value of at least two integral indexes and the coefficients of at least two integral indexes.

On the basis of the embodiment shown in fig. 2, as shown in fig. 3, in another embodiment, the integration processing method further includes:

s301, acquiring initial coefficients of the integral indexes.

In this embodiment, it is necessary to determine a corresponding initial coefficient for each integral index, and optionally, since the coefficient of each historical integral index is verified to have a certain rationality, the coefficient of each historical integral index may be used as the initial coefficient. Optionally, experts in the field may grade each integral index according to the importance degree of each integral index, and determine the initial coefficient of each integral index according to the grading result. For example, 1 indicates the same importance, 9 indicates the same importance, the integration use frequency is slightly less important than the integration use time, and is represented as 1/2, the integration use number is higher in importance than the integration use time, and is represented as 5, the integration use number is 9, the integration use time is the same as the integration use time, the integration use frequency is the same as the integration use frequency, the integration use number is the same as the integration use number, and is represented as 1.

And S302, acquiring the coefficient of each integral index according to the initial coefficient of each integral index.

In this embodiment, the initial coefficients of the integral indexes are obtained in step S301, and the coefficients of the integral indexes are determined according to the initial coefficients of the integral indexes, optionally, a corresponding neural network model may be constructed, and the initial coefficients of the integral indexes are used as input information, and the coefficients of the integral indexes are output through training of the neural network model. Optionally, the initial coefficient of each integral index may be calculated through a corresponding algorithm, so as to obtain the coefficient of each integral index.

Based on the embodiment shown in fig. 3, as shown in fig. 4, in another embodiment, a detailed implementation process of obtaining the coefficient of each integral indicator according to the initial coefficient of each integral indicator includes:

s401, acquiring a first weight matrix of the initial coefficient of each integral index according to the initial coefficient of each integral index by adopting an analytic hierarchy process.

In the present embodiment, the initial coefficients of the integral indexes are compared with each other by the initial coefficients of the integral indexes obtained in step S301, and the comparison result is used as a first weight matrix of the initial coefficients of the integral indexes. The initial coefficient ratio of the integration use frequency to the integration use time period is 1/2, the initial coefficient ratio of the integration use time period to the integration use frequency is 2/1, the ratio of the integration use number to the integration use time period is 5/1, the ratio of the integration use time period to the integration use number is 1/5, the ratio of the integration use number to the integration use frequency is 9/1, the ratio of the integration use frequency to the integration use number is 1/9, the ratio of the integration use number to the integration use number is 1/1, the ratio of the integration use frequency to the integration use frequency is 1/1, and the ratio of the integration use time period to the integration use time period is 1/1. The above-described proportional results are used as a first weight matrix of the initial coefficients of the respective integral indexes, as shown in table 1.

TABLE 1 first weighting matrix of initial coefficients of each integral index

Each integral index	Integration duration of use (R)	Integral use frequency (F)	Number of integral used (Q)
				Integration duration of use (R)	1	1/2	5
Integral use frequency (F)	2	1	9
				Number of integral used (Q)	1/5	1/9	1

S402, calculating coefficients of the integral indexes according to the first weight matrix.

In this embodiment, a first weight matrix is obtained in step S401, where the first weight matrix represents the importance degree between the integral indexes, and the importance degree between the integral indexes can be calculated through a corresponding algorithm to obtain a matrix corresponding to the coefficient of each integral index, and the coefficient of each integral index is obtained according to the matrix corresponding to the coefficient of each integral index.

In the integration processing method provided in this embodiment, an analytic hierarchy process is used to obtain a first weight matrix of the initial coefficient of each integral index according to the initial coefficient of each integral index, and calculate the coefficient of each integral index according to the first weight matrix. In the embodiment, the first weight matrix is obtained by an analytic hierarchy process, so that the influence of subjective factors in the expert scoring process is weakened, and the coefficient calculation process of each integral index is more accurate.

Based on the embodiment shown in fig. 4, as shown in fig. 5, in another embodiment, a detailed implementation process of calculating the coefficient of each integral indicator according to the first weight matrix includes:

s501, calculating the proportion of each element in the first weight matrix in the column to obtain a second weight matrix.

In this embodiment, a first weight matrix of the initial coefficients of the integral indexes is obtained in step S401, each column in the first weight matrix is summed, and then the ratio of each element in the first weight matrix in the column is calculated, so as to obtain a second weight matrix. For example, the summation of the first column in the first weight matrix is 1+2+1/5 equal to 3.2, the ratio of the first element in the first column is 1/3.2 equal to 0.3125, and according to the rounding, three bits after the decimal point are retained, the ratio of the first element in the first column is 0.313, and 0.313 is taken as the value of the first element in the first column of the second weight matrix. The ratio of the second element in the first column is 2/3.2 to 0.625, and 0.625 is used as the value of the second element in the first column of the second weight matrix. The third element in the first column has a ratio of 1/5/3.2 to 0.0625 in the first column, and three decimal places are retained by rounding, the ratio of the third element in the first column is 0.063, and 0.063 is defined as the numerical value of the third element in the first column of the second weight matrix. The numerical value of the second weight matrix is obtained according to the above method, the fourth row of the second weight matrix is taken as the sum of the ratios of each column, and each value of the fourth row of the second weight matrix is 1. The second weight matrix is shown in table 2.

TABLE 2 second weight matrix

Second weight matrix	Integration duration of use (R)	Integral use frequency (F)	Number of integral used (Q)
				Integration duration of use (R)	0.313	0.310	0.333
Integral use frequency (F)	0.625	0.621	0.600
				Number of integral used (Q)	0.063	0.069	0.067
Summing	1	1	1

And S502, summing each row of the second weight matrix, and normalizing the summation result of each row to obtain the coefficient of each integral index.

The normalization processing is provided for convenience of data processing, data are mapped into a range of 0-1 for processing, and the processing process is simplified. The expression of the normalization process is as follows:

wherein Z' represents data after normalization processing, Z represents data before normalization processing, max (Z) represents the maximum value in the data before normalization processing, and min (Z) represents the minimum value in the data before normalization processing.

In this embodiment, according to the second weight matrix obtained in step S501, each row of the second weight matrix is summed, and the result of the summation is normalized, so that the coefficient of each integral index can be obtained. The first row of the second weight matrix is summed, that is, 0.313+0.310+0.333 is 0.956, the normalized value is 0.319, and the coefficient of the integral usage time length is 0.319; summing the second row of the second weight matrix, namely 0.625+0.621+0.600 equals to 1.846, wherein the normalized value is 0.615, and the coefficient of the integral use frequency is 0.615; the third row of the second weight matrix is summed, i.e. 0.063+0.069+0.067 to 0.066, the normalized value is 0.066, and the coefficient of the integral number used is 0.066. The fourth row of the second weight matrix is the sum of the columns, matrix B includes a matrix of coefficients of the integral indices, and matrix B is shown in table 3, where W% represents the coefficient of each integral index, for example, the coefficient of the integral use duration (R) is 0.319, the coefficient of the integral use frequency (F) is 0.615, and the coefficient of the integral use number (Q) is 0.066.

TABLE 3 matrix B

In the integration processing method provided in this embodiment, the ratio of each element in the first weight matrix in the column is calculated to obtain a second weight matrix, each row of the second weight matrix is summed, and the summed result of each row is normalized to obtain the coefficient of each integral index. In the embodiment, the summation calculation is performed on each row of the second weight matrix, and the result is normalized, so that the whole calculation process is more convenient, and the coefficient of each integral index can be obtained more quickly.

The integration processing method provided in this embodiment obtains the initial coefficient of each integration index, and obtains the coefficient of each integration index according to the initial coefficient of each integration index. The initial coefficients of the integral indexes in the embodiment are obtained through the importance degree scores of the experts, and the influence of subjective factors is weakened through further processing the importance degree scores of the experts, so that the determination process of the coefficients of the integral indexes is more reasonable.

On the basis of the embodiment shown in fig. 3, as shown in fig. 6, the coefficients of the respective integral indexes are verified to ensure the accuracy of the coefficients of the respective integral indexes, and in another embodiment, the integration processing method further includes:

s601, calculating a second maximum characteristic root according to the first maximum characteristic root of the first weight matrix, the matrix order of the first weight matrix and the coefficient of each integral index.

The characteristic polynomial of the matrix can calculate all characteristic values of the matrix, and the maximum characteristic root refers to the maximum characteristic value in the matrix. The solving formula of the maximum characteristic root is as follows:

in the formula, λ_maxDenotes a maximum feature root, A denotes a first maximum feature root of a first weight matrix, W_iAnd n represents the order of the matrix.

In this embodiment, according to the first weight matrix obtained in step S402, the maximum feature root of the first weight matrix is calculated, the coefficient of each integral index is multiplied by the first maximum feature root to obtain a column vector, and each element in the column vector is located in the matrix order and the corresponding coefficient of each integral index to obtain the second maximum feature root. For example, the first maximum eigenroot of the first weight matrix obtained by the maximum eigenroot solving method is 3, the coefficients of the integral indices are multiplied by the first maximum eigenroot to obtain column vectors of 0.957, 1.847, and 0.198, and the second maximum eigenroot of 3.0012 is finally obtained by using the matrix order of each element in the column vector and the corresponding coefficient of each integral index. The matrix C includes the second largest feature root therein, and is shown in table 4.

TABLE 4 matrix C

And S602, performing consistency verification on the coefficient of each integral index according to the second maximum characteristic root.

In this embodiment, the second maximum feature root obtained in step S601 is used to verify the coefficient of each integral index through the maximum feature root, optionally, it may be verified whether the second maximum feature root is smaller than a preset threshold, and if the second maximum feature root is smaller than the preset threshold, it indicates that the coefficient of each integral index passes the consistency verification; if the integral index is larger than the preset threshold value, the consistency verification of the coefficient of each integral index is not passed. Optionally, the random consistency ratio may be calculated through the second largest feature root, and it is determined whether the obtained random consistency ratio is within a preset range, if so, the consistency verification of the coefficient of each integral index is passed, and if not, the consistency verification of the coefficient of each integral index is not passed. The present embodiment is not limited to this, and only needs to determine whether the consistency verification of the coefficients of the integral indexes passes.

In the integration processing method provided in this embodiment, a second maximum feature root is calculated according to a first maximum feature root of a first weight matrix, a matrix order of the first weight matrix, and a coefficient of each integral index, and consistency verification is performed on the coefficient of each integral index according to the second maximum feature root. In this embodiment, the second maximum feature root is obtained by calculating the first maximum feature root of the first weight matrix, and the coefficient of each integral index is verified according to the second maximum feature root, so that the accuracy and the rationality of the coefficient of each integral index can be improved.

On the basis of the embodiment shown in fig. 6, as shown in fig. 7, in another embodiment, a detailed description is given of a specific implementation process of performing consistency verification on the coefficients of the respective integral indexes according to the second maximum feature root, including:

s701, determining the value of the random consistency index according to the second maximum characteristic root and the matrix order of the first weight matrix.

The random consistency index refers to the consistency degree of the matrix which needs to be checked when the maximum characteristic value of the matrix is calculated. The expression for the random consistency index is as follows:

wherein C.I represents a random consistency index, λ_maxRepresenting the maximum eigenroot and n representing the matrix order.

In this embodiment, the second maximum feature root is obtained according to step S601, and the value of the random consistency index can be determined according to the matrix order of the first weight matrix and the second maximum feature root through the expression of the random consistency index. E.g. λ_maxIs 3.0012, n is 3, and lambda can be obtained by the formula calculation_maxThe value of (A) is 0.00062.

S702, determining the random consistency ratio according to the value of the random consistency index and the reference value of the random consistency index.

The random consistency ratio is a ratio of a value of the random consistency index to a reference value of the random consistency index, and can be expressed as:

in the formula, c.r represents a random consistency ratio, and r.i represents a reference value of a random consistency index, which is fixed and can be obtained by looking up a table through a matrix order.

In this embodiment, the value of the random consistency index is obtained in step S701, and the random consistency ratio is obtained by the above expression of the random consistency ratio according to the value of the random consistency index and the reference value of the random consistency index. Wherein, the reference value of the random consistency index obtained by the matrix order table lookup is 0.52, and the value of the random consistency ratio is 0.001186.

And S703, if the random consistency ratio is smaller than a preset threshold value, determining that the consistency verification of the coefficients of the integral indexes is passed.

In this embodiment, the random consistency ratio obtained in step S702 is 0.001186, and the preset threshold value obtained from historical experience is 0.01, that is, the random consistency ratio is smaller than the preset threshold value, which indicates that the coefficient of each integral index has no logic error, that is, the consistency verification of the coefficient of each integral index passes.

In the integration processing method provided in this embodiment, a value of a random consistency index is determined according to a second maximum feature root and a matrix order of a first weight matrix, a random consistency ratio is determined according to the value of the random consistency index and a reference value of the random consistency index, the random consistency ratio is compared with a preset threshold, and if the random consistency ratio is smaller than the preset threshold, it is determined that consistency verification of coefficients of each integration index passes. In the embodiment, the random consistency ratio is obtained through the second maximum feature root, and whether the coefficient of each integral index is reasonable or not is judged through the value of the consistency ratio, so that the accuracy of the coefficient verification process of each integral index is improved.

Based on the embodiment shown in fig. 2, as shown in fig. 8, in another embodiment, a specific implementation process of obtaining a value of an integral indicator according to integral usage data, which includes the following steps:

s801, acquiring the integral use time length, the integral use number of at least two different time periods and the integral use frequency of at least two different time periods according to the integral use data.

In this embodiment, data annotation is performed on the integral usage duration, the integral usage number in different time periods, and the integral usage frequency data, and then the integral usage duration with data annotation, the integral usage number in at least two different time periods, and the integral usage frequency in at least two different time periods are obtained. The at least two different time periods are used for avoiding the problem that the data are single and cannot truly reflect the use condition of the integration. For example, the point use data is data in the last half year, and the last half year refers to data of the last 1 month to 6 months, that is, the point use number and the point use frequency of 2 months and 4 months may be selected, and the point use number and the point use frequency of 1 month, 3 months, and 6 months may be selected.

S802, obtaining the value of each integral index according to the integral use duration, the integral use number of at least two different time periods, the integral use frequency of at least two different time periods and the weight of each time period.

In this embodiment, according to the integral usage duration, the integral usage number of at least two different time periods, and the integral usage frequency of at least two different time periods obtained in step S801, the weight of each time period is determined by the weighting score of the expert, the integral usage number of each time period, the integral usage frequency, and the weight of each time period are multiplied, and then weighted average is performed, so as to obtain the value of each integral index. For example, when the number of integrated uses and the frequency of integrated uses are 1 month, 3 months, and 6 months, the number of integrated uses after the process of 1 month is Q₁ ⁱAnd the number of credits used after 3 months of treatment is Q₂ ⁱAnd the number of credits used after 6 months of treatment is Q₃ ⁱAnd the integral use frequency after 1 month of treatment is F₁ ⁱAnd the integral use frequency after 3 months of treatment is F₂ ⁱAnd the integrated use frequency after 6 months of treatment is F₃ ⁱThe integral usage time is expressed as RⁱThen the data vector of each integral index can be represented as P_i＝{Q₁ ⁱ、Q₂ ⁱ、Q₃ ⁱ、F₁ ⁱ、F₂ ⁱ、F₃ ⁱ、Rⁱ}. The total weight score was 1, the weight score for 1 month was 0.4, the weight score for 3 months was 0.3, and the weight score for 6 months was 0.3. In this case, the value Q ' of the number of integrations to be used may be represented by Q ' 0.4Q '₁+0.3Q′₂+0.3Q′₃The value F 'of the integral use frequency may be expressed as F' 0.4F₁′+0.3F′₂+0.3F′₃。

In the integration processing method provided in this embodiment, the integration use duration, the integration use number in at least two different time periods, and the integration use frequency in at least two different time periods are obtained according to the integration use data, and the value of each integration index is obtained according to the integration use duration, the integration use number in at least two different time periods, the integration use frequency in at least two different time periods, and the weight of each time period. In the embodiment, the integral use number and the integral use frequency of at least two different time periods are obtained, so that the data of the integral use number and the integral use frequency are not too comprehensive, different weights are taken for different time periods, and the determined values of all integral indexes are more reasonable.

On the basis of the embodiment shown in fig. 8, as shown in fig. 9, in another embodiment, a detailed description is given of a specific implementation process for acquiring the value of each integral index according to the integral usage duration, the integral usage number of at least two different time periods, the integral usage frequency of at least two different time periods, and the weight of each time period, including:

s901, respectively carrying out first preprocessing on the integral use duration, the integral use number of at least two different time periods and the integral use frequency of at least two different time periods to obtain the processed integral use duration, the processed integral use number of each time period and the processed integral use frequency; the first preprocessing includes at least one of a normalization processing and a normalization processing.

After the data is normalized, the average value of each feature becomes 0 and the standard deviation becomes 1, and the expression of the normalization process can be represented as:

wherein Z represents normalized data, and X represents_iDenotes the raw data, μ denotes the standard deviation of the raw data, and σ denotes the mean of the raw data.

In this embodiment, after obtaining data of the integral usage duration, the integral usage number of at least two different time periods, and the integral usage frequency of at least two different time periods, normalizing and/or normalizing the data, and mapping the data to a range of 0 to 1 for processing, the processed integral usage duration, the processed integral usage number of each time period, and the processed integral usage frequency are obtained.

And S902, carrying out weighted summation on the processed integral use number of each time period and the weight of each time period to obtain a value of the integral use number.

In this embodiment, according to the integral usage number after each time period processing obtained in step S901, the weight of each time period is obtained in step S802, the integral usage number after each time period processing is multiplied by the weight of each time period, and the multiplication results are subjected to weighted summation, so as to obtain the value of the integral usage number finally. For example, if the weights of 1, 3, and 6 months are 0.4, 0.3, and 0.3, respectively, the integrated usage amount after 1 month of processing is Q₁', the number of integration uses after 3 months of processing is Q₂', the number of integration uses after 6 months of processing is Q₃', the final value of the integral usage quantity, Q, can be expressed as:

Q′＝0.4Q′₁+0.3Q′₂+0.3Q′₃

and S903, carrying out weighted summation on the processed integral use frequency of each time segment and the weight of each time segment to obtain a value of the integral use frequency.

In this embodiment, according to the integral use frequency obtained in step S901 after processing in each time segment, the weight of each time segment is obtained in step S802, the integral use frequency obtained after processing in each time segment is multiplied by the weight of each time segment, and the multiplication results are subjected to weighted summation, so as to obtain the value of the integral use frequency. For example, if the weights of 1 month, 3 months and 6 months are 0.4, 0.3 and 0.3, respectively, the integrated usage frequency after 1 month treatment is F₁', the integral usage frequency after the treatment of 3 months is F₂', the integral usage frequency after the treatment of month 6 is F₃', the value of the final integrated usage frequency, F, can be expressed as:

F′＝0.4F₁′+0.3F′₂+0.3F′₃

and S904, taking the processed integral use time length as the value of the integral use time length.

In this embodiment, each integral usage duration is a fixed value, so the integral usage duration does not relate to a certain time period, and there is no weight corresponding to the time period, and therefore the integral usage duration after normalization and/or normalization processing can be used as the value of the integral usage duration.

The integration processing method provided in this embodiment performs first preprocessing on the integration use duration, the integration use number in at least two different time periods, and the integration use frequency in at least two different time periods, to obtain the processed integration use duration, the processed integration use number in each time period, and the processed integration use frequency, performs weighted summation on the processed integration use number in each time period and the weight in each time period, to obtain a value of the integration use number, performs weighted summation on the processed integration use frequency in each time period and the weight in each time period, to obtain a value of the integration use frequency, and takes the processed integration use duration as the value of the integration use duration. In the embodiment, different weights are set for different time periods, so that the integral use number and the integral use frequency which are closer to the actual situation can be obtained, and the integral use situation can be reflected more reasonably.

On the basis of the embodiment shown in fig. 2, as shown in fig. 10, in another embodiment, a detailed implementation process for acquiring the point usage data of the user within the preset time period is described, which includes:

and S1001, acquiring original point use data of a user in a preset time period from a database.

In this embodiment, data identification is performed according to the user raw point usage data of different time periods, and the user raw point usage data of a preset time period can be acquired from the database through the data identification. Optionally, the raw point usage data of the user may be obtained by means of a keyword. Alternatively, raw point usage data for the user may be obtained over a network. This embodiment is not limited to this.

S1002, carrying out second preprocessing on the original integral using data to obtain integral using data; the second preprocessing includes padding missing values to zero and/or clearing singular values.

The singular value is obtained by the singular value decomposition theorem, and diagonal elements of the matrix are called as singular values of the matrix.

In this embodiment, the original integral usage data obtained in step S1001 needs to be subjected to data cleaning and a second preprocessing, because a part of values in the original data are missing, a missing value needs to be filled with 0, and meanwhile, if a singular value exists in the original data, the singular value needs to be removed, so as to obtain the integral usage data after the second preprocessing.

In the integration processing method provided by this embodiment, the original integration use data of the user in the preset time period is acquired from the database, the missing value of the original integration use data is filled with zero, and the singular value in the original integration use data is removed to obtain the integration use data. According to the embodiment, the quality of the original data can be improved through the preprocessing process of the original data, so that the subsequent calculation process is more accurate.

In one embodiment, as shown in fig. 11, to facilitate understanding of those skilled in the art, the following detailed description of the integration processing method may include:

s1101, acquiring original point use data of a user in a preset time period from a database;

s1102, filling missing values in the original integral using data to be zero, and removing singular values to obtain integral using data;

s1103, acquiring integral use duration, integral use number of at least two different time periods and integral use frequency of at least two different time periods according to the integral use data;

s1104, respectively performing first preprocessing on the integral use duration, the integral use number of at least two different time periods and the integral use frequency of at least two different time periods to obtain the processed integral use duration, the processed integral use number of each time period and the processed integral use frequency;

s1105, weighting and summing the processed integral use number of each time period and the weight of each time period to obtain the value of the integral use number;

s1106, carrying out weighted summation on the processed integral use frequency of each time period and the weight of each time period to obtain a value of the integral use frequency;

s1107, taking the processed integral usage time length as a value of the integral usage time length;

s1108, acquiring initial coefficients of all integral indexes;

s1109, acquiring a first weight matrix of the initial coefficient of each integral index according to the initial coefficient of each integral index by adopting an analytic hierarchy process;

s1110, calculating the proportion of each element in the first weight matrix in the column to obtain a second weight matrix;

s1111, summing each row of the second weight matrix, and normalizing the summation result of each row to obtain the coefficient of each integral index;

s1112, calculating a second maximum characteristic root according to the first maximum characteristic root of the first weight matrix, the matrix order of the first weight matrix and the coefficient of each integral index;

s1113, determining the value of the random consistency index according to the second maximum characteristic root and the matrix order of the first weight matrix;

s1114, determining a random consistency ratio according to the value of the random consistency index and the reference value of the random consistency index;

s1115, judging whether the random consistency ratio is smaller than a preset threshold value, if so, determining that the consistency verification of the coefficients of the integral indexes is passed, and executing the step S1116; if not, executing step S1117;

and S1116, determining the integral use activity of the user according to the value of each integral index and the coefficient of each integral index.

S1117, the initial coefficients of the integral indexes are newly acquired, and step S1109 is executed.

Fig. 12 shows a flow diagram of the point processing procedure, first obtaining the past half year customer point usage data from the database, cleaning and preprocessing the data, extracting data of the number of the use of the credits and the frequency of the use of the credits in 1, 3 and 6 months in the past half year, normalizing the obtained data, respectively giving different weights to 1 month, 3 months and 6 months, calculating the weighted average value of the integral use quantity and the integral use frequency, then, based on the analytic hierarchy process, calculating the coefficients of the integral use quantity, the integral use frequency and the integral use time length, and carrying out consistency check on the obtained coefficients, and if the check is passed, calculating the point use activity of each client according to the coefficients of the number of the used points, the use frequency of the points, the use time length of the points, the use number of the points, the use frequency of the points and the use time length of the points.

The integral processing method provided in this embodiment obtains original integral usage data of a user in a preset time period from a database, performs second preprocessing on the original integral usage data to obtain integral usage data, obtains integral usage time, integral usage number in at least two different time periods, and integral usage frequency in at least two different time periods according to the integral usage data, respectively performs first preprocessing on the integral usage time, the integral usage number in at least two different time periods, and the integral usage frequency in at least two different time periods to obtain processed integral usage time, the processed integral usage number in each time period, and the processed integral usage frequency, performs weighted summation on the processed integral usage number in each time period and the weight in each time period to obtain a value of the integral usage number, and performs weighted summation on the processed integral usage frequency in each time period and the weight in each time period, obtaining the value of integral use frequency, taking the processed integral use time as the value of the integral use time, obtaining the initial coefficient of each integral index, obtaining a first weight matrix of the initial coefficient of each integral index according to the initial coefficient of each integral index by adopting an analytic hierarchy process, calculating the proportion of each element in the first weight matrix in the column to obtain a second weight matrix, summing each row of the second weight matrix, carrying out normalization processing on the summation result of each row to obtain the coefficient of each integral index, calculating a second maximum characteristic root according to the first maximum characteristic root of the first weight matrix, the matrix order of the first weight matrix and the coefficient of each integral index, determining the value of a random consistency index according to the value of the random consistency index and the reference value of the random consistency index, and determining a random consistency ratio, if the random consistency ratio is smaller than a preset threshold value, determining that the consistency of the coefficient of each integral index passes verification, and determining the integral use activity of the user according to the value of each integral index and the coefficient of each integral index. The embodiment determines the corresponding coefficients through different integral use indexes, simultaneously performs consistency check on the determined coefficients, ensures the rationality of the integral use index coefficients, and comprehensively and reasonably evaluates the integral use activity through different integral use indexes and the coefficients corresponding to the integral use indexes.

It should be understood that although the various steps in the flow charts of fig. 2-12 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2-12 may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed in turn or alternately with other steps or at least some of the other steps.

In one embodiment, as shown in fig. 13, there is provided an integration processing device including: a first obtaining module 11, a second obtaining module 12 and a determining module 13, wherein:

the first acquisition module 11 is configured to acquire point usage data of a user within a preset time period;

a second obtaining module 12, configured to obtain a value of an integral index according to the integral usage data; the integral index comprises at least two of integral using quantity, integral using frequency and integral using duration, and the integral using duration is the duration from the moment when the user uses the integral for the last time to the current moment;

and the determining module 13 is configured to determine the integral usage activity of the user according to the value of each integral index and the coefficient of each integral index.

In another embodiment, as shown in fig. 14, the integration processing device further includes:

a third obtaining module 14, configured to obtain an initial coefficient of each integral indicator;

and a fourth obtaining module 15, configured to obtain a coefficient of each integral indicator according to the initial coefficient of each integral indicator.

In one embodiment, as shown in fig. 15, the fourth obtaining module 15 includes: a first acquisition unit 151 and a first calculation unit 152. Wherein:

a first obtaining unit 151, configured to obtain, according to the initial coefficient of each integral index, a first weight matrix of the initial coefficient of each integral index by using an analytic hierarchy process;

the first calculating unit 152 is configured to calculate a coefficient of each integral indicator according to the first weight matrix.

In one embodiment, as shown in fig. 16, the fourth obtaining module 15 further includes: a second calculation unit 153, a verification unit 154. Wherein:

a second calculating unit 153, configured to calculate a second maximum feature root according to the first maximum feature root of the first weight matrix, the matrix order of the first weight matrix, and the coefficient of each integral index;

a verifying unit 154, configured to perform consistency verification on the coefficients of the integration indexes according to the second maximum feature root.

In one embodiment, as shown in fig. 17, the first obtaining module 11 further includes: a second acquisition unit 111, a third acquisition unit 112. Wherein:

a second obtaining unit 111, configured to obtain, from a database, raw point usage data of a user within a preset time period;

a third obtaining unit 112, configured to perform second preprocessing on the original integral usage data to obtain integral usage data; the second preprocessing includes padding missing values to zero and/or clearing singular values.

For the specific limitation of the integration processing device, reference may be made to the above limitation of the integration processing method, which is not described herein again. The respective modules in the integration processing device can be wholly or partially implemented by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 18. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, an operator network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement an integration processing method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.

Those skilled in the art will appreciate that the architecture shown in fig. 18 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:

acquiring integral use data of a user in a preset time period;

acquiring a value of an integral index according to the integral use data; the integral index comprises at least two of integral using quantity, integral using frequency and integral using duration, and the integral using duration is the duration from the moment when the user uses the integral for the last time to the current moment;

In one embodiment, the processor, when executing the computer program, further performs the steps of:

acquiring an initial coefficient of each integral index;

In one embodiment, the processor, when executing the computer program, further performs the steps of: obtaining the coefficient of each integral index according to the initial coefficient of each integral index, comprising:

In one embodiment, the processor, when executing the computer program, further performs the steps of: calculating coefficients of the integral indexes according to the first weight matrix, wherein the calculating comprises the following steps:

In one embodiment, the processor, when executing the computer program, further performs the steps of: calculating a second maximum characteristic root according to the first maximum characteristic root of the first weight matrix, the matrix order of the first weight matrix and the coefficient of each integral index;

In one embodiment, the processor, when executing the computer program, further performs the steps of: and performing consistency verification on the coefficient of each integral index according to the second maximum characteristic root, wherein the consistency verification comprises the following steps:

determining the value of the random consistency index according to the second maximum characteristic root and the matrix order of the first weight matrix;

determining a random consistency ratio according to the value of the random consistency index and the reference value of the random consistency index;

In one embodiment, the processor, when executing the computer program, further performs the steps of: the integral index comprises integral use number, integral use frequency and integral use duration, and the value of the integral index is obtained according to the integral use data, and the integral index comprises the following steps:

acquiring integral use duration, integral use quantity of at least two different time periods and integral use frequency of at least two different time periods according to the integral use data;

and acquiring the value of each integral index according to the integral use duration, the integral use number of at least two different time periods, the integral use frequency of at least two different time periods and the weight of each time period.

In one embodiment, the processor, when executing the computer program, further performs the steps of: obtaining the value of each integral index according to the integral use duration, the integral use number of at least two different time periods, the integral use frequency of at least two different time periods and the weight of each time period, wherein the method comprises the following steps:

respectively performing first pretreatment on the integral use duration, the integral use number of at least two different time periods and the integral use frequency of at least two different time periods to obtain the treated integral use duration, the treated integral use number of each time period and the treated integral use frequency; the first preprocessing includes at least one of a normalization processing and a normalization processing;

weighting and summing the processed integral use frequency of each time period and the weight of each time period to obtain a value of the integral use frequency;

In one embodiment, the processor, when executing the computer program, further performs the steps of: acquiring point use data of a user in a preset time period, wherein the point use data comprises the following steps:

carrying out second preprocessing on the original integral using data to obtain integral using data; the second preprocessing includes padding missing values to zero and/or clearing singular values.

In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:

acquiring integral use data of a user in a preset time period;

In one embodiment, the computer program when executed by the processor further performs the steps of:

acquiring an initial coefficient of each integral index;

In one embodiment, the computer program when executed by the processor further performs the steps of: obtaining the coefficient of each integral index according to the initial coefficient of each integral index, comprising:

In one embodiment, the computer program when executed by the processor further performs the steps of: calculating coefficients of the integral indexes according to the first weight matrix, wherein the calculating comprises the following steps:

In one embodiment, the computer program when executed by the processor further performs the steps of: and performing consistency verification on the coefficient of each integral index according to the second maximum characteristic root, wherein the consistency verification comprises the following steps:

In one embodiment, the computer program when executed by the processor further performs the steps of: the integral index comprises integral use number, integral use frequency and integral use duration, and the value of the integral index is obtained according to the integral use data, and the integral index comprises the following steps:

In one embodiment, the computer program when executed by the processor further performs the steps of: obtaining the value of each integral index according to the integral use duration, the integral use number of at least two different time periods, the integral use frequency of at least two different time periods and the weight of each time period, wherein the method comprises the following steps:

In one embodiment, the computer program when executed by the processor further performs the steps of: acquiring point use data of a user in a preset time period, wherein the point use data comprises the following steps:

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware related to instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above examples only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. An integration processing method, characterized in that the method comprises:

acquiring integral use data of a user in a preset time period;

2. The method of claim 1, further comprising:

acquiring an initial coefficient of each integral index;

3. The method of claim 2, wherein obtaining the coefficient of each integral indicator according to the initial coefficient of each integral indicator comprises:

4. The method of claim 3, wherein said calculating coefficients for each of said integral indicators based on said first weight matrix comprises:

5. The method according to claim 3 or 4, characterized in that the method further comprises:

6. The method of claim 5, wherein the performing consistency verification on the coefficients of the integral indicators according to the second largest root of features comprises:

7. The method according to any one of claims 1-4, wherein the integral indicator comprises an integral usage number, an integral usage frequency and an integral usage duration, and the obtaining a value of the integral indicator from the integral usage data comprises:

8. The method of claim 7, wherein obtaining the value of each of the integral indicators according to the integral usage duration, the integral usage amount of the at least two different time periods, the integral usage frequency of the at least two different time periods, and the weight of each of the time periods comprises:

9. The method according to any one of claims 1-4, wherein the obtaining of the point usage data of the user within the preset time period comprises:

10. An integration processing apparatus, characterized in that the apparatus comprises:

11. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor realizes the steps of the method of any one of claims 1 to 9 when executing the computer program.

12. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 9.