WO2017201905A1 - Data distribution method and device, and storage medium - Google Patents

Abstract

Disclosed are a data distribution method and device and a storage medium. The method comprises: obtaining data feature information of one or more pieces of data to be distributed; obtaining recipient feature information of one or more data recipients; matching the data feature information with the recipient feature information; and distributing the data to be distributed to a corresponding data recipient according to a matching degree. Embodiments of the present invention can achieve targeted data distribution by matching data feature information of data to be distributed with recipient feature information of a data recipient and distributing the data to be distributed to the corresponding data recipient according to a matching degree.

Description

Data distribution method, device and storage medium

This patent application claims priority from the Chinese patent application filed on May 25, 2016, the application number is 201610354138.0, the applicant is Baidu Online Network Technology (Beijing) Co., Ltd., and the invention name is "data distribution method and device". The full text of the application is hereby incorporated by reference.

Technical field

The embodiments of the present invention relate to the field of information processing technologies, and in particular, to a data distribution method, apparatus, and storage medium.

Background technique

In the Customer Relationship Management (CRM) system, customer data is the core data and assets, which are characterized by a large number, varying degrees of quality, timeliness and multidimensional features. In the sales CRM system, the salesperson is the core user, and the salesperson itself has the difference in sales ability, industry accumulation and geographical preference. It is a critical issue for sales personnel to efficiently and accurately find customer profiles that match their business capabilities.

Taking Baidu's sales CRM system as an example, the salesperson will generally check the customer information he wants to follow according to the conditions, and then go to the phone to follow up the specific customer information and reach an order. The way in which such information is obtained requires the salesperson to obtain it manually, and the obtained customer data depends on the subjective understanding of the customer's subjective knowledge of the customer, and the efficiency and matching degree of obtaining the customer's data are relatively low.

Or, the customer information is input into the CRM system by a dedicated staff member, and then distributed to the salesperson by the system randomly or according to the principle of even distribution. This distribution method is less targeted, and the salesperson can only passively accept the customer data, ignoring The difference in the sales staff themselves.

Summary of the invention

The invention provides a data distribution method and device for distributing information in a targeted manner.

In a first aspect, an embodiment of the present invention provides a data distribution method, including:

Obtaining data feature information of one or more materials to be distributed;

Obtaining personnel characteristic information of one or more data receiving personnel;

Matching the material feature information with the person feature information;

The to-be-distributed materials are distributed to corresponding data recipients according to the degree of matching.

In a second aspect, an embodiment of the present invention further provides a data distribution apparatus, including:

a data feature acquiring module, configured to acquire data feature information of one or more materials to be distributed;

a personnel feature obtaining module, configured to acquire personnel characteristic information of one or more data receiving personnel;

a feature matching module, configured to match the data feature information with the personnel feature information;

And a data distribution module, configured to distribute the to-be-distributed data to a corresponding data receiving person according to a matching degree.

In a third aspect, an embodiment of the present invention further provides a storage medium including computer executable instructions for executing a data distribution method when executed by a computer processor, the method comprising:

Obtaining data feature information of one or more materials to be distributed;

Obtaining personnel characteristic information of one or more data receiving personnel;

Matching the material feature information with the person feature information;

The to-be-distributed materials are distributed to corresponding data recipients according to the degree of matching.

In a fourth aspect, the embodiment of the present invention further provides a data distribution device, including:

One or more processors;

a memory that stores one or more programs;

When the one or more programs are executed by the one or more processors, causing the one or more processors to perform the following operations:

Obtaining data feature information of one or more materials to be distributed;

Obtaining personnel characteristic information of one or more data receiving personnel;

Matching the material feature information with the person feature information;

The to-be-distributed materials are distributed to corresponding data recipients according to the degree of matching.

In the embodiment of the present invention, by matching the data feature information of the data to be distributed with the personnel feature information of the data receiving personnel, the data to be distributed is distributed to the corresponding data receiving personnel according to the matching degree, and the data can be distributed in a targeted manner.

DRAWINGS

1 is a flowchart of a data distribution method in Embodiment 1 of the present invention;

2A is a flowchart of a data distribution method in Embodiment 2 of the present invention;

2B is a schematic diagram showing a result of analyzing customer data in a data distribution method according to Embodiment 2 of the present invention;

2C is a schematic diagram showing the distribution result of using the existing data distribution method in Embodiment 2 of the present invention;

2D is a schematic diagram of distribution results provided in a data distribution method according to Embodiment 2 of the present invention;

2E is a schematic diagram of a data distribution display interface in a data distribution method according to Embodiment 2 of the present invention;

3 is a structural diagram of a data distribution apparatus in Embodiment 3 of the present invention;

4 is a structural diagram of a data distribution device in Embodiment 3 of the present invention.

detailed description

The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. It should also be noted that, for ease of description, only some, but not all, of the structures related to the present invention are shown in the drawings.

Embodiment 1

FIG. 1 is a flowchart of a data distribution method according to a first embodiment of the present invention. The present embodiment is applicable to a data distribution apparatus provided by an embodiment of the present invention. Executing, the device may be implemented by using a software or a hardware, and the device may be integrated into a mobile terminal, a tablet computer, a fixed terminal, or a server, as shown in FIG.

S101. Acquire data feature information of one or more materials to be distributed.

The information to be distributed may be one piece of information or multiple pieces of information. The type of the to-be-distributed material may specifically be, but is not limited to, any one of a file, a document, and a customer profile.

The material feature information is composed of at least one keyword, and the keyword represents the content or attribute included in the to-be-distributed material, including but not limited to industry information to which the data belongs, enterprise business information and data of the enterprise to which the data belongs. At least one of a geographical location of the affiliated company, a telephone communication record of the material, and a visit record of the information. The telephone communication record is the number or time of historical telephone communication performed by the relevant person (such as a customer or a salesperson) for the data. The visit record is the number, location or time of historical visits made by the relevant person (eg, customer or salesperson) for the material.

Specifically, the content to be distributed may be obtained by parsing the to-be-distributed data, and the keyword is extracted from the included content as the feature information, or is summarized according to the included content. A keyword capable of characterizing the meaning or attribute of the material to be distributed, and the summarized keyword is used as the material feature information.

S102. Acquire person characteristic information of one or more data receiving personnel.

The data receiving personnel is a receiving object of the data to be distributed. The personnel characteristic information represents related information of the data receiving personnel, including but not limited to at least one of industry information engaged by the data receiving personnel, business operation information of the enterprise, geographical location, telephone communication record, and visit record. The telephone communication record is the historical telephone communication time or time for a data receiving person (such as a customer or a sales person) for a certain data. A visit record is the number, location, or time of historical visits to a profile by a person (such as a customer or salesperson) who receives the data.

S103. Match the data feature information with the personnel feature information.

Specifically, taking the above S102 and S103 as an example, if the data feature information includes industry information to which the data belongs, enterprise business information of the enterprise to which the data belongs, geographic location of the enterprise to which the data belongs, telephone communication record of the data, and visit of the data Recording, the personnel characteristic information includes industry information engaged by the data receiving personnel, business operation information of the enterprise, geographic location, telephone communication record, and visit record. When matching is performed, each feature item included in the material feature information and each feature item included in the person feature information are matched one by one.

S104. Distribute the to-be-distributed data to a corresponding data receiving personnel according to the degree of matching.

Specifically, if the matching degree of a piece of information to be distributed to a certain data receiving person exceeds a preset threshold, or the matching degree of the piece of data with the data receiving person is the maximum value of all matching degrees, the piece of data is Distribute to the recipient of the data. For example, the industry information to which the data belongs is matched with the industry information of the data receiving personnel, and the information is preferentially distributed to the personnel of the same industry; the business operation information of the enterprise to which the data belongs and the business operation information of the enterprise where the data receiving personnel are located are performed. Match, Priority is given to distributing the information to persons with similar business conditions; matching the geographical location of the data to the geographic location of the recipient of the data, prioritizing the distribution of the information to the same geographical location, and so on.

In this embodiment, the data feature information of the data to be distributed is matched with the personnel feature information of the data receiving personnel, and the data to be distributed is distributed to the corresponding data receiving personnel according to the matching degree, so that the data can be distributed in a targeted manner.

Exemplarily, on the basis of the foregoing embodiment, in order to facilitate the calculation of the matching degree between the data feature information and the personnel feature information, the embodiment of the present invention, the data feature information and the personnel feature information are matrixized, and specifically includes:

The data feature information and the person feature information are respectively a data feature matrix and a personnel feature matrix, wherein each of the to-be-distributed materials and each of the data receiving personnel have the same feature dimension.

Correspondingly, matching the data feature information with the personnel feature information includes:

A similarity matrix of the data feature matrix and the person feature matrix is calculated, and the similarity matrix is used as a matching degree.

The material feature matrix is obtained by performing feature modeling on the data to be distributed. The data feature matrix includes n feature values, and the features represented by the feature values include, but are not limited to, at least one of industry information, business operation information, geographic location, telephone communication record, and visit record. The similarity may specifically be, but not limited to, any one of a cosine value, an Euclidean distance, a Pearson correlation coefficient, a Spearman rank correlation coefficient, a Tanimoto coefficient, a log likelihood similarity, and a Manhattan distance.

When assigning the data feature matrix, the following methods can be used:

Parsing the data to be distributed, and if the corresponding feature in the data feature matrix is included, the capital is The corresponding feature value in the material feature matrix is assigned to 1, otherwise the value is assigned to 0.

If the industry information contained in the data feature matrix is IT, chemical, and legal, if the industry is determined to be chemical by analyzing the data to be distributed, the eigenvalue corresponding to the chemical in the data feature matrix is assigned a value of 1, and the remaining features are The eigenvalues corresponding to IT and law are assigned a value of zero. Similarly, for other features such as business operations information, geographic location, telephone communication records, and visit records, similar methods are used for assignment.

When assigning the personnel feature matrix, the following method can be used:

According to the data receiving personnel, the preferences of the features in the personnel feature matrix are respectively assigned.

In the embodiment of the present invention, each feature item in the person feature matrix is the same as each feature item in the data feature matrix. According to the degree of interest of the person to the feature, each feature is assigned, and the assignment range is 0~1. For example, if the person has a degree of interest in Beijing of 1, the feature value corresponding to the feature of Beijing is assigned a value of 1. If the degree of interest of the person to IT is 0.5, the feature value corresponding to the feature of IT is assigned a value of 0.5, etc. .

After the data assignment is performed, the distributing the data to be distributed to the corresponding data receiver according to the degree of matching further includes:

When the plurality of materials to be distributed are multiple, based on the calculated similarity matrix, the plurality of to-be-distributed materials are separately distributed to the corresponding data receiving personnel by using an optimization algorithm, so that one piece of information is distributed to one person and distributed. The sum of the similarities is the largest.

The optimization algorithm is preferably but not limited to a weighted bipartite graph optimal matching algorithm.

Specifically, based on the calculated similarity matrix, using an optimization algorithm to separately distribute the plurality of data to be distributed to the corresponding data receiving personnel mainly includes the following two distribution methods:

The first distribution method: based on the calculated similarity matrix, an optimization algorithm is used to distribute a plurality of data to be distributed to a corresponding data receiving person in a fixed distribution form, that is, when each data is distributed, The same number of items to be distributed are fixedly allocated by personnel. Specifically, in the case of distributing m pieces of data to be distributed to t data receivers, each data receiver is fixedly allocated k pieces of data to be distributed, where m and t are positive integers, and m is greater than t; The similarity matrix C _mt is transformed into a similarity matrix C _m(t×k) of t×k data receivers in the column direction; the m data to be distributed is distributed to t×k virtual data receiving by an optimization algorithm. personnel.

The second distribution method is: based on the calculated similarity matrix, the plurality of to-be-distributed materials are respectively distributed to the corresponding data receiving personnel in a variable distribution amount by using an optimization algorithm, that is, when distributing the data, according to each The data receiver's personal ability to assign a different number of data to be distributed to each data receiver. Specifically, for the case where the to-be-distributed data to be distributed is distributed to t data receiving personnel, a set number of to-be-distributed materials are separately allocated according to the personal capabilities of each data receiving person, wherein the i-th data receiving personnel The number of allocated data to be distributed is denoted as k _i , where m and t are positive integers, and m is greater than t; the similarity matrix C _mt is transformed in the column direction to

Similarity matrix of virtual data receivers

Distribution of m data to the optimization algorithm

A virtual data receiver.

In the above embodiments, the data of the data to be distributed and the personnel characteristic information of the data receiver are matrixed, and the similarity of the two matrices is calculated, and the data to be distributed is distributed to the corresponding data based on the principle of maximum similarity. Receiving personnel, able to distribute information in a targeted manner.

Embodiment 2

2A is a flowchart of a data distribution method according to Embodiment 2 of the present invention. In this embodiment, the data to be distributed is used as customer data, and the data receiving personnel considers the sales as a scenario, and the present invention is described in detail, such as As shown in FIG. 2A, the specific includes:

S201: Analyze customer data, and extract data feature information that is used to represent the customer data, where the data feature information is a data feature matrix that includes n feature values.

By parsing the customer data, the customer data is classified and described in a tree manner as shown in FIG. 2B, and all the leaf nodes in the tree are all feature values. If the customer data can be abstracted into n feature values, Can be expressed as an n × 1 dimensional matrix. Customer profile information can describe a row matrix, ie data profile matrix A{security security, office culture, medical health, ..., Beijing, Shanghai, Guangzhou, ...} to indicate that its value will be similar {0,1,0,...,0,1,0,...} (1 indicates that the feature is met, 0 means no), the matrix indicates that it is an office culture and education industry, customer information in the Shanghai area. Assuming that there are m customer data in the library, all the feature values of the customer data can be represented as an m×n matrix A _mn .

S202. Obtain a person characteristic information of a salesperson, where the person feature information is a data feature matrix including n feature values.

Specifically, based on the previous salesperson's single record and follow-up records, the statistical characteristics of the customer data that the sales follow-up can be analyzed. This is called the salesperson's preference. This preference can also be expressed as a matrix or sales. Personnel corresponding data feature matrix B{security security, office culture, medical health, ..., Beijing, Shanghai, Guangzhou, ...}, its value is similar to {0.3,0.5,0.1...,0} The matrix indicates that the salesperson's interest in the safety and security industry is 0.3, the degree of interest in office culture and education is 0.5, and the degree of interest in the medical and health industry is 0.1. The greater the value, the more interested. If the total number of salespersons is t, then all salespersons' preference values are represented as an _nxt matrix _Bnt .

S203. Calculate a similarity matrix of the data feature matrix and the person feature matrix.

Based on the previously obtained customer data feature matrix A _mn and the salesperson's preference matrix B _nt , the similarity matrix of the two is calculated. Due to the sparsity of the two matrices, the cosine similarity matching algorithm is used to represent the customer data characteristics and sales. The degree of matching of people's interests.

If A{A ₁ , A ₂ ,..., A _n } represents the data feature matrix of a customer profile, B{B ₁ , B ₂ ,..., B _n } represents the personnel identity matrix of a salesperson. The cosine similarity between the two is calculated as follows:

Wherein, the value of i is 1 to n, and the value of the cosine similarity is larger, indicating that the characteristics of the customer data and the preference of the salesperson match.

The two characteristic matrices A _mn and B _{nt are} used to calculate the cosine similarity of the matrix, and finally a similarity matrix C _{mt is obtained} , where C _{ij(i<=m, j<=t)} represents the similarity between the customer data i and the salesperson j. coefficient.

Based on this similarity matrix, two pieces of information can be obtained: (1) the top k salesperson list that is most interested in the customer profile in the customer data dimension; (2) the salesperson dimension is most interested in the salesperson dimension A list of the top k customer profiles.

S204. The plurality of to-be-distributed materials are separately distributed to the corresponding data receiving personnel by using an optimization algorithm based on the calculated similarity matrix.

In the distribution, the following two conditions must be met: (1) exclusive demand: one customer data can only be distributed to one salesperson; (2) optimized demand: the sum of similarity of the collected customer data is the largest .

Suppose there are t sales personnel, and the average number of customer data recommended by each salesperson is k. For the convenience of calculation, t×k virtual sales personnel can be abstracted (k times for each sales), and the similarity matrix C can be used. _Mt is transformed into a matrix of t × k dimensions C _{m(t × k)} in the direction of the column. , indicating m customer data, t × k virtual sales staff similarity matrix.

At this point, the optimal distribution of m customer data to t×k virtual sales personnel is converted into a weighted bipartite graph (t×k is generally less than m), and the set of sales data and the set of virtual sales personnel are composed of two points. The two sets of vertices of the graph, c _ij (i<=m, j<=t×k), represent the weight coefficients between the customer data i and the virtual salesperson j. Based on the Kuhn-Munkres algorithm, a customer data matched by t×k virtual sales personnel can be obtained, so that the overall similarity is maximized, thereby achieving the best matching effect.

As shown in FIG. 2C, after the similar relationship between the initial salesperson and the customer profile is distributed by using the solution of the embodiment, a final distribution relationship as shown in FIG. 2D is obtained.

As shown in FIG. 2E, a schematic diagram of a distribution interface is provided in this embodiment. On the interface, the user can select the customer data to be distributed, and the salesperson who receives the data, and then click the distribution button to start the embodiment of the present invention. The data distribution process and the distribution results as shown below in Figure 2E.

In addition, after the solution of the embodiment of the present invention is used in the sales CRM system of Baidu, the indicators issued by the customer data are greatly improved, as shown in the following table:

Evaluation index	Manual distribution	Intelligent acquisition method
Distribution speed (10,000 customer data)	2 hours	10 minutes
Assign customer data accuracy	*	90%

Sales staff into a single amount (months)

4.5

5.5

It can be seen that the data distribution method proposed by the embodiment of the present invention is adopted, which saves a large amount of distribution manpower, and the data distribution is more targeted. The most important thing is that the salesperson's single-sheet quantity is obviously improved, and a better economy is obtained. benefit.

Embodiment 3

FIG. 3 is a schematic structural diagram of a data distribution apparatus according to Embodiment 3 of the present invention. The apparatus may be implemented by using software or hardware, and the apparatus may be integrated into a mobile terminal, a tablet computer, a fixed terminal, or a server, such as As shown in FIG. 3, the specific structure of the device is as follows: a material feature acquisition module 31, a personnel feature acquisition module 32, a feature matching module 33, and a data distribution module 34.

The data feature obtaining module 31 is configured to acquire data feature information of one or more materials to be distributed;

The person feature obtaining module 32 is configured to acquire personnel feature information of one or more data receiving personnel;

The feature matching module 33 is configured to match the material feature information with the personnel feature information;

The data distribution module 34 is configured to distribute the to-be-distributed materials to corresponding data receiving personnel according to the degree of matching.

The data distribution device in the embodiment is used to perform the data distribution method described in the foregoing embodiments, and the technical principle and the generated technical effects are similar, and details are not described herein again.

Based on the foregoing embodiment, the data feature information and the personnel feature information are respectively a data feature matrix and a personnel feature matrix, wherein each of the to-be-distributed materials and each data receiving person has a special feature The same dimension;

Correspondingly, the feature matching module 33 is specifically configured to calculate a similarity matrix of the data feature matrix and the person feature matrix, and use the similarity matrix as a matching degree.

Based on the foregoing embodiment, the data feature acquiring module is configured to obtain a data feature matrix by performing feature modeling on the data to be distributed.

On the basis of the foregoing embodiment, the data feature obtaining module 31 is specifically configured to parse the data to be distributed; if the corresponding feature in the data feature matrix is included, assign the corresponding feature value in the data feature matrix to 1 Otherwise, the value is 0.

Based on the foregoing embodiment, the personnel feature obtaining module 32 is specifically configured to respectively assign a preference to each feature in the personnel feature matrix according to the data receiving personnel.

On the basis of the foregoing embodiment, the data distribution module 34 is specifically configured to: when the plurality of materials to be distributed are multiple, use an optimization algorithm to distribute multiple to-be-distributed data to each other based on the calculated similarity matrix. Corresponding data receivers, so that a piece of information is distributed to a person and the sum of similarities is maximized after distribution.

Based on the foregoing embodiment, the data distribution module 34 is specifically configured to allocate, to each data receiving person, k pieces of to-be-distributed data for the case where the m pieces of data to be distributed are distributed to t data receiving personnel, where m , t is a positive integer, m is greater than t;

The similarity matrix C _mt is transformed in the column direction into a similarity matrix C _m(t×k) of t×k data receivers;

The m pieces of data to be distributed are distributed to t×k virtual data receivers by using an optimization algorithm.

On the basis of the foregoing embodiment, the data distribution module 34 is specifically configured to allocate, according to the personal capabilities of each data receiving personnel, the distribution of the m to-be-distributed materials to be distributed to the t-receiving personnel. The number of data to be distributed, wherein the number of data to be distributed allocated by the i-th data receiver is recorded as k _i , where m and t are positive integers, and m is greater than t;

Transform the similarity matrix C _mt in the column direction to

Similarity matrix of virtual data receivers

Distribution of m data to the optimization algorithm

A virtual data receiver.

Based on the foregoing embodiment, the optimization algorithm is a weighted bipartite graph optimal matching algorithm.

Based on the above embodiment, the similarity is any one of a cosine value, an Euclidean distance, a Pearson correlation coefficient, a Spearman rank correlation coefficient, a Tanimoto coefficient, a log likelihood similarity, and a Manhattan distance.

The data distribution device described in the above embodiments is used to perform the data distribution method described in the above embodiments, and the technical principle and the generated technical effects are similar, and are not described herein again.

Embodiment 4

A fourth embodiment of the present invention provides a data distribution device, including the data distribution device provided by any embodiment of the present invention, which can be integrated into a mobile terminal, a tablet computer, a fixed terminal, or a server.

Specifically, as shown in FIG. 4, an embodiment of the present invention provides a data distribution device, which includes a processor 40, a memory 41, an input device 42, and an output device 43. The number of processors 40 in the data distribution device may be One or more, one processor 40 is taken as an example in FIG. 4; the processor 40, the memory 41, the input device 42, and the output device 43 in the data distribution device may be connected by a bus or other means, in FIG. Connection is an example.

The memory 41 is used as a computer readable storage medium, and can be used for storing software programs, computer executable programs, and modules, such as program instructions/modules corresponding to the data distribution method in the embodiment of the present invention (for example, data characteristics in the data distribution device) The acquisition module 31, the personnel feature acquisition module 32, the feature matching module 33, and the material distribution module 34). The processor 30 executes various functional applications and data processing of the material distribution device by executing software programs, instructions, and modules stored in the memory 31, that is, implementing the above-described data distribution method.

The memory 41 may mainly include a storage program area and an storage data area, wherein the storage program area may store an operating system, an application required for at least one function; the storage data area may store data created according to usage of the material distribution device, and the like. Further, the memory 41 may include a high speed random access memory, and may also include a nonvolatile memory such as at least one magnetic disk storage device, flash memory device, or other nonvolatile solid state storage device. In some examples, memory 41 may further include memory remotely located relative to processor 40, which may be connected to the data distribution device over a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

Input device 42 can be used to receive input numeric or character information, as well as to generate key signal inputs related to user settings and function control of the data distribution device. The output device 43 may include a display device such as a display screen.

Embodiments of the present invention also provide a storage medium including computer executable instructions for executing a data distribution method when executed by a computer processor, the method comprising:

Obtaining data feature information of one or more materials to be distributed;

Obtaining personnel characteristic information of one or more data receiving personnel;

Matching the material feature information with the person feature information;

The to-be-distributed materials are distributed to corresponding data recipients according to the degree of matching.

Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by software and necessary general hardware, and can also be implemented by hardware, but in many cases, the former is a better implementation. . Based on such understanding, the technical solution of the present invention, which is essential or contributes to the prior art, may be embodied in the form of a software product, which may be stored in a computer readable storage medium, such as a floppy disk of a computer. , Read-Only Memory (ROM), Random Access Memory (RAM), Flash (FLASH), hard disk or optical disk, etc., including a number of instructions to make a computer device (can be a personal computer) The server, or network device, etc.) performs the methods described in various embodiments of the present invention.

It should be noted that, in the embodiment of the foregoing data distribution device, each unit and module included in the data distribution device is divided according to functional logic, but is not limited to the above division, as long as the corresponding function can be implemented; The specific names of the functional units are also for convenience of distinguishing from each other and are not intended to limit the scope of the present invention.

The above is only a specific embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily think of changes or substitutions within the technical scope of the present disclosure. All should be covered by the scope of the present invention. Therefore, the scope of the invention should be determined by the scope of the appended claims.

Claims (22)

1. A data distribution method, comprising:

   Obtaining data feature information of one or more materials to be distributed;

   Obtaining personnel characteristic information of one or more data receiving personnel;

   Matching the material feature information with the person feature information;

   The to-be-distributed materials are distributed to corresponding data recipients according to the degree of matching.
2. The method of claim 1 wherein:

   The data feature information and the person feature information are respectively a data feature matrix and a personnel feature matrix, wherein each of the to-be-distributed materials and each data receiving person has the same feature dimension;

   Correspondingly, matching the data feature information with the personnel feature information includes:

   A similarity matrix of the data feature matrix and the person feature matrix is calculated, and the similarity matrix is used as a matching degree.
3. The method of claim 2 wherein:

   The data feature matrix is obtained by performing feature modeling on the data to be distributed.
4. The method according to claim 3, wherein the obtaining the data feature information of the data to be distributed comprises:

   Parsing the information to be distributed;

   If the corresponding feature in the data feature matrix is included, the corresponding feature value in the data feature matrix is assigned to 1, otherwise the value is 0.
5. The method according to claim 2, wherein the obtaining the personnel characteristic information of the data receiving person comprises:

   According to the data receiving personnel, the preferences of the features in the personnel feature matrix are respectively assigned.
6. The method according to claim 2, wherein said to be distributed is matched according to the degree of matching The information is distributed to the appropriate recipients of the data, including:

   When the plurality of materials to be distributed are multiple, based on the calculated similarity matrix, the plurality of to-be-distributed materials are separately distributed to the corresponding data receiving personnel by using an optimization algorithm, so that one piece of information is distributed to one person and distributed. The sum of the similarities is the largest.
7. The method according to claim 6, wherein, based on the calculated similarity matrix, the plurality of to-be-distributed materials are separately distributed to the corresponding data receiving personnel by using an optimization algorithm, including:

   For the case of distributing m pieces of data to be distributed to t data receivers, each data receiver is fixedly allocated k pieces of data to be distributed, where m and t are positive integers, and m is greater than t;

   The similarity matrix C _mt is transformed in the column direction into a similarity matrix C _m(t×k) of t×k data receivers;

   The m pieces of data to be distributed are distributed to t×k virtual data receivers by using an optimization algorithm.
8. The method according to claim 6, wherein, based on the calculated similarity matrix, the plurality of to-be-distributed materials are separately distributed to the corresponding data receiving personnel by using an optimization algorithm, including:

   For the case where the to-be-distributed materials to be distributed are distributed to t data receiving personnel, a set number of to-be-distributed materials are respectively allocated according to the individual capabilities of each data receiving person, wherein the _i- th data receiving personnel are allocated The number of distribution materials is recorded as k _i , where m and t are positive integers, and m is greater than t;

   Converting the similarity matrix C _mt in the column direction to

   

   Similarity matrix of virtual data receivers

   

   Distribution of m data to the optimization algorithm

   

   A virtual data receiver.
9. The method according to claim 7 or 8, wherein the optimization algorithm is a weighted bipartite graph optimal matching algorithm.
10. The method according to any one of claims 2 to 8, wherein the similarity is a cosine value, an Euclidean distance, a Pearson correlation coefficient, a Spearman rank correlation coefficient, a Tanimoto coefficient, a log likelihood similarity, and Manhattan Any of the distances.
11. A data distribution device, comprising:

    a data feature acquiring module, configured to acquire data feature information of one or more materials to be distributed;

    a personnel feature obtaining module, configured to acquire personnel characteristic information of one or more data receiving personnel;

    a feature matching module, configured to match the data feature information with the personnel feature information;

    And a data distribution module, configured to distribute the to-be-distributed data to a corresponding data receiving person according to a matching degree.
12. The device of claim 11 wherein:

    The data feature information and the person feature information are respectively a data feature matrix and a personnel feature matrix, wherein each of the to-be-distributed materials and each data receiving person has the same feature dimension;

    Correspondingly, the feature matching module is specifically configured to:

    A similarity matrix of the data feature matrix and the person feature matrix is calculated, and the similarity matrix is used as a matching degree.
13. The device of claim 12 wherein:

    The data feature acquisition module is configured to obtain a data feature matrix by performing feature modeling on the data to be distributed.
14. The device according to claim 12, wherein the data feature acquisition module is specifically configured to:

    Parsing the information to be distributed;

    If the corresponding feature in the data feature matrix is included, the data feature matrix is correspondingly The eigenvalue is assigned to 1, otherwise it is assigned a value of 0.
15. The device according to claim 12, wherein the personnel feature acquisition module is specifically configured to:

    According to the data receiving personnel, the preferences of the features in the personnel feature matrix are respectively assigned.
16. The device according to claim 12, wherein the data distribution module is specifically configured to:

    When the plurality of materials to be distributed are multiple, based on the calculated similarity matrix, the plurality of to-be-distributed materials are separately distributed to the corresponding data receiving personnel by using an optimization algorithm, so that one piece of information is distributed to one person and distributed. The sum of the similarities is the largest.
17. The device according to claim 16, wherein the data distribution module is specifically configured to:

    For the case of distributing m pieces of data to be distributed to t data receivers, each data receiver is fixedly allocated k pieces of data to be distributed, where m and t are positive integers, and m is greater than t;

    The similarity matrix C _mt is transformed in the column direction into a similarity matrix C _m(t×k) of t×k data receivers;

    The m pieces of data to be distributed are distributed to t×k virtual data receivers by using an optimization algorithm.
18. The device according to claim 16, wherein the data distribution module is specifically configured to:

    For the case where the to-be-distributed materials to be distributed are distributed to t data receiving personnel, a set number of to-be-distributed materials are respectively allocated according to the individual capabilities of each data receiving person, wherein the i-th data receiving personnel are allocated The number of distribution materials is recorded as k _i , where m and t are positive integers, and m is greater than t;

    Transform the similarity matrix C _mt in the column direction to

    

    Similarity matrix of virtual data receivers

    

    Distribution of m data to the optimization algorithm

    

    A virtual data receiver.
19. The apparatus according to claim 17 or 18, wherein the optimization algorithm is a weighted bipartite graph optimal matching algorithm.
20. The apparatus according to any one of claims 12 to 18, wherein the similarity is a cosine value, an Euclidean distance, a Pearson correlation coefficient, a Spearman rank correlation coefficient, a Tanimoto coefficient, a log likelihood similarity, and Manhattan Any of the distances.
21. A storage medium comprising computer executable instructions for performing a data distribution method when executed by a computer processor, the method comprising:

    Obtaining data feature information of one or more materials to be distributed;

    Obtaining personnel characteristic information of one or more data receiving personnel;

    Matching the material feature information with the person feature information;

    The to-be-distributed materials are distributed to corresponding data recipients according to the degree of matching.
22. A data distribution device, comprising:

    One or more processors;

    a memory that stores one or more programs;

    When the one or more programs are executed by the one or more processors, causing the one or more processors to perform the following operations:

    Obtaining data feature information of one or more materials to be distributed;

    Obtaining personnel characteristic information of one or more data receiving personnel;

    Matching the material feature information with the person feature information;

    The to-be-distributed materials are distributed to corresponding data recipients according to the degree of matching.

Images