CN115619142A

CN115619142A - Method, device, equipment and computer readable medium for matching data

Info

Publication number: CN115619142A
Application number: CN202211246473.0A
Authority: CN
Inventors: 林建进; 陈兰欢; 丁长林; 刘昊
Original assignee: Beijing Jingdong Zhenshi Information Technology Co Ltd
Current assignee: Beijing Jingdong Zhenshi Information Technology Co Ltd
Priority date: 2022-10-12
Filing date: 2022-10-12
Publication date: 2023-01-17

Abstract

The invention discloses a method, a device, equipment and a computer readable medium for matching data, and relates to the technical field of computers. One embodiment of the method comprises: after the business characteristics of the salespeople are coded by a Transformer model, the business characteristics are input into a deep FM model by combining the clue characteristics of the sales clue so as to obtain clue joint characteristics of the clue characteristics; inputting the business characteristics of the salespeople into a Transformer model to obtain the business combination characteristics of the business characteristics; and extracting a clue business combination characteristic value from the characteristic set constructed by the clue joint characteristic and the business combination characteristic by using a residual error network, and outputting the matching degree of the salesperson and the sales thread based on the clue business combination characteristic value so as to match the salesperson and the sales thread. The method and the system can improve the matching degree of sales leads and sales personnel, and further improve the order number.

Description

Method, device, equipment and computer readable medium for matching data

Technical Field

The present invention relates to the field of computer technologies, and in particular, to a method, an apparatus, a device, and a computer-readable medium for matching data.

Background

In the sales management system, sales leads are the forefront of the opportunities generated by the clients, and generally, sales leads are obtained in various ways such as marketing activities, network information, telephone consultation, consumer interview and the like.

And issuing the sales leads to the salespersons according to the areas. The sales force continues to follow and drive the continued extension of the thread, and the sales thread switches to a sales opportunity after reaching the maturity stage. By formally establishing items in a company, a salesperson can be used as a sales opportunity to conduct funnel type management and promotion, and finally reach an agreement with a client and formally sign a contract order through negotiation, business, product and technical communication in several stages.

In the process of implementing the invention, the inventor finds that at least the following problems exist in the prior art: the sales leads are distributed according to the sales regions, so that the distribution of the sales leads is not matched, and the order quantity cannot be increased.

Disclosure of Invention

In view of this, embodiments of the present invention provide a method, an apparatus, a device, and a computer readable medium for matching data, which can improve the matching degree between a sales lead and a salesperson, and further improve the order quantity.

To achieve the above object, according to an aspect of an embodiment of the present invention, there is provided a method of matching data, including:

after the business characteristics of the salespeople are coded by a Transformer model, the business characteristics are input into a deep FM model by combining the clue characteristics of the sales clue so as to obtain clue joint characteristics of the clue characteristics;

inputting the business characteristics of the salespersons into a Transformer model to obtain the business combination characteristics of the business characteristics;

and extracting a clue business combination characteristic value from the characteristic set constructed by the clue joint characteristic and the business combination characteristic by using a residual error network, and outputting the matching degree of the salesperson and the sales thread based on the clue business combination characteristic value so as to match the salesperson and the sales thread.

The cue features include one or more of the following parameters: company information, lead business scope, type of goods sold, business area and area where the company is located.

The service characteristics include one or more of the following parameters: the area, the processing rate of historical clues of the work information, the conversion rate, the contract signing rate and the abandonment rate.

After the business characteristics of the salespersons are coded by a Transformer model, the business characteristics are input into a deep FM model by combining the thread characteristics of the sales thread so as to obtain the thread joint characteristics of the thread characteristics, and the method comprises the following steps:

identifying the weight of the business characteristics of the salespersons through an index matrix of an encoder in the Transformer model;

and inputting the weight of the business characteristic and the thread characteristic of the sales thread into a deep FM model to obtain the thread joint characteristic of the thread characteristic.

The sales person is the same as a trained sales person,

the step of inputting the service characteristics of the salespersons into a Transformer model to obtain the service combination characteristics of the service characteristics comprises the following steps:

in the process of training the Transformer model, inputting the business characteristics of the training salesman into the Transformer model to acquire and store training business combination characteristics;

and taking the training service combination characteristic as a service combination characteristic of the service characteristic.

Extracting clue service combination characteristic values from the characteristic set constructed by the clue joint characteristics and the service combination characteristics by using a residual error network, wherein the method comprises the following steps:

splicing the clue joint characteristics and the service combination characteristics to establish the characteristic set;

and inputting the feature set into the residual error network, and extracting a clue service combination feature value.

The Transformer model, the deep FM model and the residual error network are obtained by training historical service features and historical cue features.

According to a second aspect of the embodiments of the present invention, there is provided an apparatus for matching data, including:

the clue module is used for coding the business characteristics of the salespersons through a Transformer model and inputting the business characteristics of the sales clues into the deep FM model in combination with the clue characteristics to obtain clue joint characteristics of the clue characteristics;

the business module is used for inputting the business characteristics of the salesman into a Transformer model to obtain the business combination characteristics of the business characteristics;

and the matching module is used for extracting a clue service combination characteristic value in a characteristic set constructed by the clue joint characteristic and the service combination characteristic by using a residual error network, and outputting the matching degree of the salesman and the sales lead based on the clue service combination characteristic value so as to match the salesman and the sales lead.

According to a third aspect of embodiments of the present invention, there is provided an electronic device for matching data, including:

one or more processors;

a storage device to store one or more programs,

when executed by the one or more processors, cause the one or more processors to implement the method as described above.

According to a fourth aspect of embodiments of the present invention, there is provided a computer readable medium, on which a computer program is stored, which program, when executed by a processor, performs the method as described above.

One embodiment of the above invention has the following advantages or benefits: after the business characteristics of the salespeople are coded by a Transformer model, the business characteristics are input into a deep FM model in combination with the thread characteristics of the sales thread so as to obtain the thread joint characteristics of the thread characteristics; inputting the business characteristics of the salespeople into a Transformer model to obtain the business combination characteristics of the business characteristics; and extracting a clue business combination characteristic value from the characteristic set constructed by the clue joint characteristic and the business combination characteristic by using a residual error network, and outputting the matching degree of the salesperson and the sales thread based on the clue business combination characteristic value so as to match the salesperson and the sales thread. The clue service combination characteristic value can represent the correlation between the salesperson and the sales clue, and then the salesperson and the sales clue are matched based on the matching degree, so that the matching degree between the sales clue and the salesperson is improved, and the order number is further improved.

Further effects of the above-mentioned non-conventional alternatives will be described below in connection with the embodiments.

Drawings

The drawings are included to provide a better understanding of the invention and are not to be construed as unduly limiting the invention. Wherein:

FIG. 1 is a schematic main flow diagram of a method of matching data according to an embodiment of the invention;

FIG. 2 is a schematic diagram of a process for obtaining cue joint characteristics of cue features according to an embodiment of the present invention;

FIG. 3 is a schematic flow chart of extracting characteristic values of thread service combinations according to an embodiment of the present invention;

FIG. 4 is a flow chart illustrating extracting service composition characteristics according to an embodiment of the present invention;

FIG. 5 is a schematic diagram of a method of matching data using a model according to an embodiment of the invention;

FIG. 6 is a schematic diagram of the main structure of an apparatus for matching data according to an embodiment of the present invention;

FIG. 7 is an exemplary system architecture diagram in which embodiments of the present invention may be applied;

fig. 8 is a schematic block diagram of a computer system suitable for use in implementing a terminal device or server according to an embodiment of the present invention.

Detailed Description

Exemplary embodiments of the present invention are described below with reference to the accompanying drawings, in which various details of embodiments of the invention are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

The "sales lead assignment" is the assignment of sales leads that need to be transferred to the management of the sales force after the approval. The information related to the sales lead is automatically shared with the relevant sales personnel. Reassignment may be made to the sales force return thread.

Currently, after determining a sales lead, the sales lead is issued to the sales staff according to the historical sales data of the region. The salespersons utilize sales leads to facilitate sign-up for sales.

The sales capability is different according to individual capability, and the processing capability, the conversion capability, the signing capability and the industry which is good at signing are different. How to distribute the lead which is most suitable for sale to the salesperson needs algorithm capability and insight capability, and the purpose of matching the lead to the salesperson is to maximize sales processing capability, contract capability and amount of endorsement, thereby increasing the amount of endorsement after the overall sales lead is issued.

The distribution of sales leads according to the sales regions results in unbalanced distribution of sales leads and failure to increase the number of orders.

In order to solve the technical problem of the distribution imbalance of the sales leads, the following technical solutions in the embodiments of the present invention may be adopted.

Referring to fig. 1, fig. 1 is a schematic diagram of a main flow of a method for matching data according to an embodiment of the present invention, and a deep fm model is used to extract a clue joint feature, a transform model is used to extract a service combination feature, and then a matching degree between a salesperson and a sales clue is determined. As shown in fig. 1, the method specifically comprises the following steps:

s101, after the business characteristics of the salespeople are coded by a Transformer model, the business characteristics are input into a deep FM model in combination with the thread characteristics of the sales thread so as to obtain the thread joint characteristics of the thread characteristics.

In the embodiment of the invention, on one hand, the business characteristics of the salespersons are extracted from the salespersons; and on the other hand, starting from the sales lead, extracting the lead characteristics of the sales lead. And determining the matching degree of the salesperson and the sales lead on the basis of the business characteristics and the lead characteristics, and further realizing the matching of the salesperson and the sales lead according to the matching degree.

In one embodiment of the invention, the service characteristics include one or more of the following parameters: the region, the work information, the historical clue processing rate, the conversion rate, the contract signing rate and the abandonment rate. Wherein the area includes the city where the salesperson is located. The work information is information involved in the process of operating the business by the salesperson. As one example, the operational information includes one or more of the following: age, gender, job level, and working age. It should be noted that the working information is obtained according to relevant laws and regulations. As one example, a salesperson actively uploads work information. As an example, the business features of the sales king include: the area is as follows: shanghai; age: 25; sex: male; job level: 3; the age limit is as follows: 4 years; historical clue processing rate: 90 percent; the conversion rate is 80 percent; the contract signing rate is 70%; the rejection rate is 30 percent.

Specifically, n business features of the salesperson can be subjected to matrix multiplication with vectors of 1 row and m columns to be converted into a feature matrix of n rows and m columns. The method aims to expand the dimensionality of the business features of the salesman and improve the robustness of model matching. Wherein n and m are each an integer greater than 0.

In one embodiment of the invention, the thread characteristics include one or more of the following parameters: company information, lead business scope, type of goods sold, business area and area where the company is located. As one example, the company information includes a judicial and/or company name. Company information is obtained under relevant laws and regulations.

In an embodiment of the invention, there are a number of parameters that take into account the business characteristics of the sales force. In the actual matching process of the data, the importance degrees of the parameters are different, and the importance of each parameter needs to be considered.

Referring to fig. 2, fig. 2 is a schematic flow chart of obtaining the cue joint characteristics of the cue features according to the embodiment of the present invention. The method specifically comprises the following steps:

s201, identifying the weight of the business characteristics of the salesman through an index matrix of an encoder in a Transformer model.

The Transformer model includes an encoder and a decoder. The encoder in the Transformer model is used for extracting deep dimensionality of business features of sales personnel, learning and expanding the characteristic dimensionality by combining an attention mechanism in a feature matrix mode, and improving the stability and the anti-interference capability of the Transformer model.

Specifically, 6 sub-encoders and 6 sub-decoders are included in the encoder. The input to each sub-encoder is the output of the previous sub-encoder. The structure of each sub-encoder comprises a self-attention mechanism module and a feed-forward neural network. The autoflight mechanism is a variant of the attentiveness mechanism that reduces reliance on external information and is more adept at capturing internal correlations of data or features.

A pooling index operation is added between each sub-encoder. As an example, a pooling indexing operation is implemented between each sub-encoder using an index matrix by which weights of traffic characteristics are identified.

The index matrix is used for the characteristic up-sampling process of the corresponding decoder, the operation can effectively distinguish the business characteristics of the salespersons, and meanwhile, the prediction on the importance of the business characteristics of the salespersons is obviously improved. That is, the weights of the traffic characteristics are identified by the index matrix of the encoder in the transform model. For the service characteristics with higher importance, the weight of the service characteristics is improved; for the service features with lower importance, the weight of the service features is reduced.

S202, inputting the weight of the business characteristics and the clue characteristics of the sales clues into a deep FM model to obtain clue joint characteristics of the clue characteristics.

DeepFM is based on wide and Deep models, the FM layer can enable the models to have strong memory capacity, the DNN layer can enable the models to have good generalization capacity, and meanwhile, the FM and DNN share a Feature Embedding module, so that the method has the advantages of simple structure, low complexity and more accurate shared input and learning.

The depfm model includes a depfm encoder and a depfm decoder. The weight of the business features and the lead features of the sales leads are input into the deep fm model. And the deep FM encoder learns the clue characteristics according to the weight of the service characteristics, and the capability of extracting information from the wide side of the model is improved. The deep FM decoder fuses clue characteristics based on the weight of the service characteristics, namely, the characteristic dimension is reduced while main clue characteristics are kept. Finally, a clue union characteristic of the clue characteristics is obtained.

In the embodiment of fig. 2, the clue union characteristics of the clue characteristics are obtained in the deep fm model by using the weight of the service characteristics.

S102, inputting the business characteristics of the salespeople into a Transformer model to obtain the business combination characteristics of the business characteristics.

The Transformer model improves the training speed of the model through an attention mechanism, is suitable for parallelization calculation, and has higher precision and performance than the previous popular Recurrent Neural Network (RNN) due to the complexity of the model structure. The Transformer improves the training speed of the model through an attention mechanism, is suitable for parallelization calculation, and the complexity of the model structure causes the model to be higher than the previously popular RNN in precision and performance.

The role of the decoder in the Transformer model is to characterize the "value" that may occur next, based on the results of the encoder and the results of the last prediction. And simultaneously, the main characteristics are extracted and the dimension is reduced.

Inputting the business characteristics of the salespersons into an encoder of a Transformer model, and then acquiring the business combination characteristics of the business characteristics in a decoder of the Transformer model. It should be noted that the service combination feature of the service features is characterized from the service perspective. The service combination feature is the main feature highlighted on the basis of the service feature.

S103, extracting a clue service combination characteristic value by utilizing a residual error network in a characteristic set constructed by the line clue combination characteristic and the service combination characteristic, and outputting the matching degree of the salesman and the sales clue based on the clue service combination characteristic value so as to match the salesman and the sales clue.

And after the deep FM model is adopted to obtain the clue joint characteristics of the clue characteristics and the Transformer model is adopted to obtain the service comprehensive characteristics of the service characteristics, the matching degree of the salesperson and the sales clue can be determined. If the matching degree is high, the contract signing feasibility of the salesperson for the sales lead is higher; if the matching degree is low, the contract signing feasibility of the salesperson for the sales lead is smaller.

Referring to fig. 3, fig. 3 is a schematic flow chart of extracting characteristic values of a thread service combination according to an embodiment of the present invention. The method specifically comprises the following steps:

s301, splicing the clue joint characteristics and the service combination characteristics to establish a characteristic set.

The clue union characteristics are used for characterizing the characteristics of the sales clue; the business portfolio characteristics are used to characterize the sales force. And directly splicing the clue joint characteristics and the service combination characteristics, and taking a matrix obtained after splicing as a characteristic set.

S302, inputting the feature set into a residual error network, and extracting a clue service combination feature value.

In order to know the features of the feature set, the feature set may be input to a residual network. The residual network comprises a plurality of convolutional layers, and the convolutional layers extract depth characteristics, namely cue service combination characteristic values, of input data of the residual network.

Meanwhile, the information in the feature set is skipped over the convolutional layers and is directly transmitted to the following layers, and finally the information in the feature set is used as input to activate the function so as to obtain the clue service combination feature value output by the residual error network.

In the embodiment of fig. 3, the values of the thread service combination features are extracted on the basis of the construction of the thread union features and the service combination features. The thread business combination feature value fully characterizes the relevance of the thread feature to the salesperson.

The clue traffic combination characteristic value is a floating point number between 0 and 1. And taking the characteristic value of the lead service combination as the matching degree of the salesperson and the sales lead. The salespersons and sales leads are then matched based on the degree of match. As an example, if the matching degree is greater than or equal to the preset threshold value of the matching degree, the sales leads are matched to the salespersons; and if the matching degree is smaller than the preset matching degree threshold value, the technical scheme from the S101 to the S103 is executed again to determine the salespersons.

The embodiment of the invention relates to a Transformer model, a deep FM model and a residual error network, wherein the models are obtained by training historical data.

Specifically, a Transformer model is obtained by training by adopting historical business features and historical clue features. And training to obtain the DeepFM model through the historical service characteristics and the historical clue characteristics. And training to obtain a residual error network by utilizing the historical service characteristics and the historical clue characteristics.

In the process of using the Transformer model, in order to improve the speed of matching data, the business combination characteristics of the salesperson can be stored in advance.

Referring to fig. 4, fig. 4 is a schematic flow chart of extracting service combination features according to an embodiment of the present invention. The method specifically comprises the following steps:

s401, in the process of training the Transformer model, inputting the business characteristics of the trained salespersons into the Transformer model to obtain and store the training business combination characteristics.

In the practical application process, the mobility of the salespersons is small, and the salespersons are the same as the trained salespersons. In the training of the Transformer model, the business characteristics of the training salesman can be input into the Transformer model to acquire and store the training business combination characteristics.

S402, the training service combination characteristic is used as the service combination characteristic of the service characteristic.

The salespersons are the same as the trained salespersons, so that the trained business combination features can be used as the business combination features of the business features.

In the embodiment of fig. 4, the training data is used to predict the service combination feature of the service feature, so as to improve the speed of matching the data.

In the embodiment of the invention, after the business characteristics of the salesperson are coded by a Transformer model, the business characteristics are input into a deep fm model by combining the thread characteristics of the sales thread so as to obtain the thread joint characteristics of the thread characteristics; inputting the business characteristics of the salespersons into a Transformer model to obtain the business combination characteristics of the business characteristics; and extracting a clue business combination characteristic value from the characteristic set constructed by the clue joint characteristic and the business combination characteristic by using a residual error network, and outputting the matching degree of the salesperson and the sales thread based on the clue business combination characteristic value so as to match the salesperson and the sales thread. The clue service combination characteristic value can represent the correlation between the salesperson and the sales clue, and then the salesperson and the sales clue are matched based on the matching degree, so that the matching degree between the sales clue and the salesperson is improved, and the order number is further improved.

The technical scheme in the embodiment of the invention integrates the characteristics of the deep FM model, the transform model and the residual error network, and has the characteristics of high training speed, accurate learning result, strong robustness, low complexity and low data volume requirement.

Referring to fig. 5, fig. 5 is a schematic diagram of a method of matching data using a model according to an embodiment of the present invention.

In fig. 5, after the business characteristics of the salesperson are encoded by the Transformer model, the business characteristics and the cue characteristics are output by the deep fm encoder and the deep fm decoder.

And (4) outputting the service combination characteristics after the service characteristics of the salespersons pass through a Transformer encoder and a Transformer decoder.

And inputting the clue joint characteristic and the service combination characteristic into a residual error network, and outputting the matching degree of the salesperson and the sales clue by the residual error network.

Referring to fig. 6, fig. 6 is a schematic diagram of a main structure of a data matching apparatus according to an embodiment of the present invention, where the data matching apparatus may implement a data matching method, as shown in fig. 6, the data matching apparatus specifically includes:

the clue module 601 is used for coding the service characteristics of the salespersons through a Transformer model and inputting the business characteristics of the salespersons into the deep FM model by combining the clue characteristics of the sales clues so as to obtain clue joint characteristics of the clue characteristics;

a service module 602, configured to input the service characteristics of the salesperson into a transform model to obtain service combination characteristics of the service characteristics;

a matching module 603, configured to extract a clue business combination feature value in a feature set constructed by the clue combined feature and the business combination feature by using a residual network, and output a matching degree between the salesperson and the sales lead based on the clue business combination feature value, so as to match the salesperson and the sales lead.

In one embodiment of the invention, the cue features include one or more of the following parameters: company information, lead business scope, type of goods sold, business area and area where the company is located.

In one embodiment of the invention, the service characteristics include one or more of the following parameters: the region, the work information, the historical clue processing rate, the conversion rate, the contract signing rate and the abandonment rate.

In an embodiment of the present invention, the thread module 601 is specifically configured to identify, by using an index matrix of an encoder in the transform model, a weight of the business feature of the salesperson;

In one embodiment of the invention, the sales personnel are the same as the trained sales personnel,

a service module 602, configured to, in the process of training the Transformer model, input the service features of the training salespersons into the Transformer model to obtain and store training service combination features;

In an embodiment of the present invention, the matching module 603 is specifically configured to splice the cue joint feature and the service combination feature to establish the feature set;

In an embodiment of the present invention, the transform model, the deep fm model, and the residual network are obtained by training using historical service features and historical cue features.

Fig. 7 shows an exemplary system architecture 700 to which the method of matching data or the apparatus for matching data of an embodiment of the present invention may be applied.

As shown in fig. 7, the system architecture 700 may include

terminal devices

701, 702, 703, a network 704, and a server 705. The network 704 is the medium used to provide communications links between

terminal devices

701, 702, 703 and the server 705. Network 704 may include various types of connections, such as wire, wireless communication links, or fiber optic cables, to name a few.

A user may use the

terminal devices

701, 702, 703 to interact with a server 705 over a network 704, to receive or send messages or the like. The

terminal devices

701, 702, 703 may have installed thereon various communication client applications, such as a shopping application, a web browser application, a search application, an instant messaging tool, a mailbox client, social platform software, etc. (by way of example only).

The

terminal devices

701, 702, 703 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.

The server 705 may be a server providing various services, such as a background management server (for example only) providing support for shopping websites browsed by users using the

terminal devices

701, 702, 703. The backend management server may analyze and process the received data such as the product information query request, and feed back a processing result (for example, target push information and product information — just an example) to the terminal device.

It should be noted that the method for matching data provided by the embodiment of the present invention is generally executed by the server 705, and accordingly, the apparatus for matching data is generally disposed in the server 705.

It should be understood that the number of terminal devices, networks, and servers in fig. 7 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

Referring now to FIG. 8, shown is a block diagram of a computer system 800 suitable for use with a terminal device implementing an embodiment of the present invention. The terminal device shown in fig. 8 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present invention.

As shown in fig. 8, a computer system 800 includes a Central Processing Unit (CPU) 801 which can perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM) 802 or a program loaded from a storage section 808 into a Random Access Memory (RAM) 803. In the RAM 803, various programs and data necessary for the operation of the system 800 are also stored. The CPU 801, ROM 802, and RAM 803 are connected to each other via a bus 804. An input/output (I/O) interface 805 is also connected to bus 804.

The following components are connected to the I/O interface 805: an input portion 806 including a keyboard, a mouse, and the like; an output section 807 including a signal such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage portion 808 including a hard disk and the like; and a communication section 809 including a network interface card such as a LAN card, a modem, or the like. The communication section 809 performs communication processing via a network such as the internet. A drive 810 is also connected to the I/O interface 805 as needed. A removable medium 811 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 810 as necessary, so that a computer program read out therefrom is mounted on the storage section 808 as necessary.

In particular, according to the embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer-readable medium, the computer program comprising program code for performing the method illustrated by the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication section 809 and/or installed from the removable medium 811. The computer program executes the above-described functions defined in the system of the present invention when executed by the Central Processing Unit (CPU) 801.

It should be noted that the computer readable medium shown in the present invention can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present invention, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present invention, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The modules described in the embodiments of the present invention may be implemented by software or hardware. The described modules may also be provided in a processor, which may be described as: a processor includes a thread module, a business module, and a matching module. The names of these modules do not form a limitation on the module itself in some cases, for example, the thread module may also be described as "used to encode the business features of the salesperson through a Transformer model, and then input the thread features of the sales thread into the deep fm model in combination to obtain the thread joint features of the thread features".

As another aspect, the present invention also provides a computer-readable medium that may be contained in the apparatus described in the above embodiments; or may be separate and not incorporated into the device. The computer readable medium carries one or more programs which, when executed by a device, cause the device to comprise:

after the business characteristics of the salespeople are coded by a Transformer model, the business characteristics are input into a deep FM model in combination with the thread characteristics of the sales thread so as to obtain the thread joint characteristics of the thread characteristics;

inputting the business characteristics of the salespeople into a Transformer model to obtain the business combination characteristics of the business characteristics;

According to the technical scheme of the embodiment of the invention, after the business characteristics of the salespersons are coded by a Transformer model, the business characteristics are input into a deep FM model by combining the clue characteristics of sales clues so as to obtain the clue joint characteristics of the clue characteristics; inputting the business characteristics of the salespeople into a Transformer model to obtain the business combination characteristics of the business characteristics; and extracting a clue business combination characteristic value from the characteristic set constructed by the clue joint characteristic and the business combination characteristic by using a residual error network, and outputting the matching degree of the salesperson and the sales thread based on the clue business combination characteristic value so as to match the salesperson and the sales thread. The clue service combination characteristic value can represent the relevance of the salesperson and the sales clue, so that the salesperson and the sales clue are matched based on the matching degree, the matching degree of the sales clue and the salesperson is improved, and the order number is further improved.

The above-described embodiments should not be construed as limiting the scope of the invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions can occur, depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method of matching data, comprising:

2. The method of matching data according to claim 1, wherein the cue features comprise one or more of the following parameters: company information, lead business scope, type of goods sold, business area and area where the company is located.

3. The method of matching data according to claim 1, wherein the traffic characteristics include one or more of the following parameters: the region, the work information, the historical clue processing rate, the conversion rate, the contract signing rate and the abandonment rate.

4. The method for matching data according to claim 1, wherein the step of inputting the thread characteristics of the sales leads into a deep fm model after coding the business characteristics of the sales leads by a transform model to obtain the thread joint characteristics of the thread characteristics comprises:

5. The method of matching data of claim 1, wherein the salesperson is the same as a trained salesperson,

the step of inputting the business characteristics of the salespersons into a Transformer model to obtain the business combination characteristics of the business characteristics comprises the following steps:

6. The method of claim 1, wherein the extracting the cue business combination feature value from the feature set constructed by the cue joint feature and the business combination feature by using a residual network comprises:

7. The method of matching data according to claim 1, wherein the fransformer model, the deep fm model, and the residual network are trained using historical traffic features and historical cue features.

8. An apparatus for matching data, comprising:

the business module is used for inputting the business characteristics of the salespersons into a Transformer model so as to obtain the business combination characteristics of the business characteristics;

and the matching module is used for extracting a clue service combination characteristic value in a characteristic set constructed by the clue joint characteristic and the service combination characteristic by using a residual error network, and outputting the matching degree of the salesman and the sales clue based on the clue service combination characteristic value so as to match the salesman and the sales clue.

9. An electronic device that matches data, comprising:

one or more processors;

a storage device to store one or more programs,

when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-7.

10. A computer-readable medium, on which a computer program is stored, which, when being executed by a processor, carries out the method according to any one of claims 1-7.