CN117709968A - Personnel allocation method, apparatus, computer device and storage medium - Google Patents
Personnel allocation method, apparatus, computer device and storage medium Download PDFInfo
- Publication number
- CN117709968A CN117709968A CN202311731675.9A CN202311731675A CN117709968A CN 117709968 A CN117709968 A CN 117709968A CN 202311731675 A CN202311731675 A CN 202311731675A CN 117709968 A CN117709968 A CN 117709968A
- Authority
- CN
- China
- Prior art keywords
- customer
- client
- distributed
- target
- semantic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 59
- 239000013598 vector Substances 0.000 claims abstract description 217
- 238000004590 computer program Methods 0.000 claims description 25
- 230000015654 memory Effects 0.000 claims description 24
- 230000006399 behavior Effects 0.000 claims description 11
- 230000008569 process Effects 0.000 claims description 10
- 238000013473 artificial intelligence Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 9
- 238000003062 neural network model Methods 0.000 description 5
- 101100481876 Danio rerio pbk gene Proteins 0.000 description 4
- 101100481878 Mus musculus Pbk gene Proteins 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 3
- 238000013500 data storage Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000003068 static effect Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 210000001503 joint Anatomy 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 230000006403 short-term memory Effects 0.000 description 2
- 101100008044 Caenorhabditis elegans cut-1 gene Proteins 0.000 description 1
- 101100008046 Caenorhabditis elegans cut-2 gene Proteins 0.000 description 1
- 101100008047 Caenorhabditis elegans cut-3 gene Proteins 0.000 description 1
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 102100029469 WD repeat and HMG-box DNA-binding protein 1 Human genes 0.000 description 1
- 101710097421 WD repeat and HMG-box DNA-binding protein 1 Proteins 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 229910021389 graphene Inorganic materials 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to a personnel allocation method, a personnel allocation device, computer equipment and a storage medium, and relates to the technical field of artificial intelligence. The method comprises the following steps: inputting the client characteristic information of the clients to be distributed into a trained semantic representation network to obtain client semantic vectors of the clients to be distributed; searching in a pre-constructed customer semantic vector library based on the customer semantic vector of the customer to be distributed, and determining a plurality of target customer data similar to the customer semantic vector of the customer to be distributed; and determining target sales personnel from sales personnel corresponding to the target customer data based on the similarity scores of the customer semantic vectors of the customers to be distributed, the corresponding sales personnel and the listing labels in the target customer data, and distributing the target sales personnel to the customers to be distributed. By adopting the method, the matching degree between the distributed target sales personnel and the clients to be distributed can be improved, and more accurate distribution results can be obtained.
Description
Technical Field
The present application relates to the field of artificial intelligence, and in particular, to a personnel allocation method, apparatus, computer device, storage medium and computer program product.
Background
In the pre-sale link, a large number of clients are distributed, namely, for a new client or a client needing secondary distribution, the system precisely matches the most suitable sales personnel for the client based on an artificial intelligence technology, so that the probability of forming a bill is maximum or the client experience is optimal, the transaction is promoted to the greatest extent, and the client satisfaction is improved.
In the related art, a customer to be distributed is generally directly matched with sales personnel, and sales personnel which are most matched with the customer are screened out. Specifically, the customer characteristics and the salesperson characteristics are simultaneously input into a neural network model, the neural network fully interacts, extracts, converts and expresses the information of the customer characteristics and the salesperson characteristics, and the salesperson with the highest score is the system distribution result. However, since the customer features and the salesperson features cannot be in one-to-one correspondence, the two features have a difference in semantic representation, so that the respective optimal semantic representations cannot be extracted, the matching degree score of the customer and the salesperson is inaccurate, and the problem that the matching degree of the salesperson distributed to the customer is lower is solved.
Disclosure of Invention
In view of the foregoing, it is desirable to provide a personnel allocation method, apparatus, computer device, computer readable storage medium, and computer program product.
In a first aspect, the present application provides a method of personnel allocation. The method comprises the following steps:
inputting the client characteristic information of the clients to be distributed into a trained semantic representation network to obtain client semantic vectors of the clients to be distributed;
searching in a pre-constructed customer semantic vector library based on the customer semantic vector of the customer to be distributed, and determining a plurality of target customer data similar to the customer semantic vector of the customer to be distributed; the customer semantic vector library is determined based on historical customer data, each of the historical customer data being generated based on a customer, a corresponding sales person and a listing tag;
and determining target sales personnel from sales personnel corresponding to a plurality of target customer data based on similarity scores of the target customer data and the customer semantic vectors of the customers to be distributed, corresponding sales personnel and the single labels, and distributing the target sales personnel to the customers to be distributed.
In one embodiment, the inputting the client feature information of the client to be allocated into the trained semantic representation network to obtain the client semantic vector of the client to be allocated includes:
Acquiring client characteristic information corresponding to the client to be distributed in a client characteristic library based on the client identifier of the client to be distributed; the customer feature library is used for storing at least one data of a customer identifier, a customer portrait, customer behavior data, a customer relationship network and a customer source channel; the client characteristic information comprises a plurality of characteristic key value pairs, wherein the characteristic key value pairs comprise one-to-one corresponding characteristic names and characteristic values; the client characteristic information of the clients to be distributed is stored in the client characteristic library in advance;
inputting the client characteristic information corresponding to the client to be distributed into a trained semantic representation network to obtain a client semantic vector of the client to be distributed; the semantic representation network receives the characteristic value vectors of a plurality of characteristic key value pairs in the customer characteristic information, and processes each characteristic value vector based on the trained multi-layer perceptron network to obtain the customer semantic vector of the customer to be distributed.
In one embodiment, the searching in a pre-constructed customer semantic vector library based on the customer semantic vector of the customer to be allocated, determining a plurality of target customer data similar to the customer semantic vector of the customer to be allocated, includes:
Determining similarity scores between the client semantic vectors of the clients to be distributed and the client semantic vectors corresponding to the client semantic vector library according to the client semantic vectors contained in the client semantic vector library, so as to obtain a plurality of similarity scores;
and acquiring clients, corresponding salespersons and single labels between the clients and the salespersons, which correspond to the preset number of similarity scores, in order from large to small, and taking the single labels as a plurality of target client data.
In one embodiment, the determining, based on the similarity score of the client semantic vector of the client to be allocated in each target client data, the corresponding sales person and the ordered label, the target sales person from sales persons corresponding to a plurality of target client data includes:
determining sales scores of sales personnel corresponding to the target client data for each target client data; the sales score is determined based on the similarity score corresponding to the target customer data and the label value of the singulation label;
and summing a plurality of sales scores corresponding to the same sales person to obtain updated sales scores of all sales persons, and taking the sales person corresponding to the highest sales score as a target sales person.
In one embodiment, the method further comprises:
acquiring client characteristic information of a plurality of historical clients from a client characteristic library based on client identifiers of the historical clients;
inputting the customer characteristic information of a plurality of history customers into a trained semantic representation network to obtain customer semantic vectors of the history customers, and constructing the customer semantic vector library based on the customer semantic vectors of the history customers
In one embodiment, the method further comprises:
after the target sales person finishes the service of the to-be-allocated clients, updating the to-be-allocated clients, the target sales person and the list forming labels corresponding to the to-be-allocated clients to the client characteristic information of the to-be-allocated clients to obtain updated client characteristic information of the to-be-allocated clients;
and inputting the updated customer characteristic information of the customers to be distributed into a trained semantic representation network to obtain customer semantic vectors of the customers to be distributed, and storing the customer semantic vectors of the customers to be distributed into a customer semantic vector library.
In a second aspect, the present application also provides a personnel dispensing apparatus. The device comprises:
The semantic vector determining module is used for inputting the client characteristic information of the clients to be distributed into the trained semantic representation network to obtain the client semantic vectors of the clients to be distributed;
the client determining module is used for searching in a pre-constructed client semantic vector library based on the client semantic vector of the client to be distributed, and determining a plurality of target client data similar to the client semantic vector of the client to be distributed; the customer semantic vector library is determined based on historical customer data, each of the historical customer data being generated based on a customer, a corresponding sales person and a listing tag;
and the personnel allocation module is used for determining target sales personnel from sales personnel corresponding to a plurality of target customer data based on the similarity scores of the customer semantic vectors of the customers to be allocated in the target customer data, corresponding sales personnel and the single labels, and allocating the target sales personnel to the customers to be allocated.
In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor implementing the steps of the method according to the first aspect when the processor executes the computer program.
In a fourth aspect, the present application also provides a computer-readable storage medium. The computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method according to the first aspect.
In a fifth aspect, the present application also provides a computer program product. The computer program product comprising a computer program which, when executed by a processor, implements the steps of the method according to the first aspect.
The personnel allocation method, the personnel allocation device, the computer equipment, the storage medium and the computer program product are used for obtaining the customer semantic vector of the customer to be allocated by inputting the customer characteristic information of the customer to be allocated into the semantic representation network, and determining a plurality of target customers similar to the customer semantic vector and data corresponding to the target customers in a pre-constructed customer semantic vector library. Because the semantic vectors of the target client and the client to be distributed are similar, the similarity between the clients is larger, sales personnel corresponding to the target client are distributed to the client to be distributed to obtain better matching degree, the characteristic information of the target client and the characteristic information of the client to be distributed can be in one-to-one correspondence, and the characteristic information of the target client and the characteristic information of the client to be distributed are completely identical in semantic representation, so that the server can obtain the matching degree of the target client and the client to be distributed more accurately. Based on this, the server determines an optimal target sales person from the plurality of target customer data by the similarity score, the corresponding sales person, and the corresponding diagonalization label, and distributes the target sales person to the customer to be distributed. The similarity score may represent a degree of matching between the target customer and the customer to be distributed, and the listing label of each sales person may represent a listing situation of the sales person and the target customer. The server can synthesize data such as matching degree, ordering condition and the like to obtain the matching degree of each sales person and the clients to be distributed, and then the target sales person is preferably determined in the sales persons, so that the matching degree between the distributed target sales person and the clients to be distributed is improved, and a more accurate distribution result is obtained.
Drawings
FIG. 1 is a diagram of an application environment for a human distribution method in one embodiment;
FIG. 2 is a flow diagram of a human assignment method in one embodiment;
FIG. 3 is a flow diagram of the step of determining a customer semantic vector for a customer to be assigned in one embodiment;
FIG. 4 is a flow diagram of the steps of determining a similar plurality of target customer data in one embodiment;
FIG. 5 is a flow diagram of the steps for determining a targeted sales person in one embodiment;
FIG. 6 is a flow diagram of the steps for building a customer semantic vector library in one embodiment;
FIG. 7 is a flow diagram of the update customer semantic vector library step in one embodiment;
FIG. 8 is a flow chart of a human assignment method in another embodiment;
FIG. 9 is a block diagram of an exemplary embodiment of a human dispensing device;
fig. 10 is an internal structural view of a computer device in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
The personnel allocation method provided by the embodiment of the application can be applied to an application environment shown in fig. 1. The terminal 102 communicates with the server 104 through a network, the terminal 102 may upload the client feature information of the client to be allocated to the server 104, and the server 104 processes the client feature information through a semantic representation network to obtain a client semantic vector. The server 104 compares the obtained customer semantic vectors in a customer semantic vector library in the data storage system, determines target customer data corresponding to a plurality of similar target customers, determines target sales personnel based on the target customer data, and finally sends information of the target sales personnel to the terminal 102 so as to complete personnel distribution. The data storage system may store data that the server 104 needs to process, such as customer semantic vectors for each customer, customer characteristic information for each customer. The data storage system may be integrated on the server 104 or may be located on a cloud or other network server. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, internet of things devices, and portable wearable devices, where the internet of things devices may be smart speakers, smart televisions, smart air conditioners, smart vehicle devices, and the like. The portable wearable device may be a smart watch, smart bracelet, headset, or the like. The server 104 may be implemented as a stand-alone server or as a server cluster of multiple servers.
In one embodiment, as shown in fig. 2, a personnel allocation method is provided, and the method is applied to the server in fig. 1 for illustration, and includes the following steps:
step S202, inputting the client characteristic information of the clients to be distributed into a trained semantic representation network to obtain client semantic vectors of the clients to be distributed.
The customer to be distributed can be a new customer or a customer secondarily distributed to sales personnel, and the customer characteristic information is used for representing basic data information of multiple dimensions of the customer, such as customer ID, character figure information of the customer, customer behavior data, customer relationship network and the like. The semantic representation network is a pre-trained neural network model and is used for converting the customer characteristic information into customer semantic vectors of customers. The neural network model may be a common convolutional neural network CNN (Convolutional Neural Network), a recurrent neural network RNN (Recurrent Neural Network), a long and short term memory network LSTM (Long Short Term Memory), and the like, which is not particularly limited in this application.
Specifically, the server acquires the client characteristic information of the clients to be distributed from the policy database or the user database of each service system, namely, acquires the basic information of the clients to be distributed, the behavior data of the clients to be distributed, the relationship network of the clients to be distributed and the source channel of the clients to be distributed. And inputting the customer characteristic information into a pre-trained semantic representation network, and extracting the characteristics of all data in the customer characteristic information by the semantic representation network to obtain a customer semantic vector corresponding to the customer to be distributed.
Step S204, based on the customer semantic vector of the customer to be distributed, searching in a pre-constructed customer semantic vector library, and determining a plurality of target customer data similar to the customer semantic vector of the customer to be distributed.
Wherein the customer semantic vector library is determined based on historical customer data, each generated based on the customer, the corresponding sales person, and the listing tag. The history client data is data corresponding to the history client assigned with the sales person, and thus each history client data includes the sales person corresponding to the history client and the ticket label between the history client and the sales person. The order tag is used to indicate whether sales personnel and historic customers have fulfilled a trade completion order. The historical customer data may include customer characteristic information corresponding to the historical customer, and the historical customer data may be obtained from a policy database or a user database of each business system. Accordingly, since the customer to be distributed may be a new customer, the customer characteristic information corresponding to the customer to be distributed does not include the corresponding sales person and the listing label.
Specifically, the server takes the client semantic vector of the client to be allocated as a reference, and retrieves the client semantic vector of each history client in a pre-constructed client semantic vector library. For the client semantic vector of each historical client, the server determines a plurality of target clients similar to the client semantic vector of the client to be distributed through a preset similarity determination strategy, and acquires target client data corresponding to the plurality of target clients.
Step S206, determining target sales personnel from sales personnel corresponding to a plurality of target customer data based on similarity scores of customer semantic vectors of customers to be distributed, corresponding sales personnel and the formed single labels in the target customer data, and distributing the target sales personnel to the customers to be distributed.
Specifically, the server traverses the plurality of target client data, and determines sales scores for respective sales persons based on the similarity scores corresponding to the respective target clients and the diagnostically ordered labels of the sales persons, and determines matched target sales persons from the respective sales persons based on the magnitudes of the sales scores. And the server returns the information of the target sales person to the terminal, and establishes communication connection between the client corresponding to the client to be distributed and the client of the target sales person, thereby completing distribution of the target sales person.
In the personnel allocation method, the customer semantic vector of the customer to be allocated is obtained by inputting the customer characteristic information of the customer to be allocated into the semantic representation network, and a plurality of target customers similar to the customer semantic vector and data corresponding to the target customers are determined in a pre-constructed customer semantic vector library. Because the semantic vectors of the target client and the client to be distributed are similar, the similarity between the clients is larger, sales personnel corresponding to the target client are distributed to the client to be distributed to obtain better matching degree, the characteristic information of the target client and the characteristic information of the client to be distributed can be in one-to-one correspondence, and the characteristic information of the target client and the characteristic information of the client to be distributed are completely identical in semantic representation, so that the server can obtain the matching degree of the target client and the client to be distributed more accurately. Based on this, the server determines an optimal target sales person from the plurality of target customer data by the similarity score, the corresponding sales person, and the corresponding diagonalization label, and distributes the target sales person to the customer to be distributed. The similarity score may represent a degree of matching between the target customer and the customer to be distributed, and the listing label of each sales person may represent a listing situation of the sales person and the target customer. The server can synthesize data such as matching degree, ordering condition and the like to obtain the matching degree of each sales person and the clients to be distributed, and then the target sales person is preferably determined in the sales persons, so that the matching degree between the distributed target sales person and the clients to be distributed is improved, and a more accurate distribution result is obtained.
In one embodiment, as shown in fig. 3, the specific implementation process of step of inputting the client feature information of the client to be allocated into the trained semantic representation network to obtain the client semantic vector of the client to be allocated includes:
step S302, based on the client identifier of the client to be allocated, the client characteristic information corresponding to the client to be allocated in the client characteristic library is obtained.
Wherein the customer feature library is used for storing at least one data of a customer identifier, a customer portrait, customer behavior data, a customer relationship network and a customer source channel; the client characteristic information comprises a plurality of characteristic key value pairs, wherein the characteristic key value pairs comprise one-to-one corresponding characteristic names and characteristic values; customer characteristic information of customers to be distributed is stored in a customer characteristic library in advance.
Specifically, the server may obtain a client identifier of the client to be allocated, and obtain, through the client identifier, client feature information of the client to be allocated in a client feature library constructed in advance, thereby obtaining a feature key value pair of the client to be allocated.
In one example, when constructing the customer feature library, the server uses the customer identifier as a primary key, copies customer feature information of each customer from a policy database or a user database of each service system to the customer feature library, obtains a pre-constructed customer feature library, and updates the customer feature library in real time.
In one example, the customer feature library may contain data such as customer identifiers, customer portraits, customer behavior data, customer relationship networks, and customer source channels, where the customer identifiers may be customer IDs; the customer image may contain information of the customer's gender, age, academic, occupation, birth place, address, etc.; the client behavior data can be information such as historical application records, claim settlement records, application registration records and the like; the customer relationship network can be information such as family relationship, social relationship, policy relationship and the like; the customer source channel may be a channel for employee recommendations, client web pages, applications, phones, etc.
In one example, the customer characteristic information contains a plurality of characteristic key value pairs, the keys in the characteristic key value pairs being characteristic names, e.g., gender, academy, customer relationship, the values in the characteristic key value pairs being characteristic values, e.g., male, university, family, customer mother ID.
Step S304, inputting the client characteristic information corresponding to the client to be distributed into a trained semantic representation network to obtain the client semantic vector of the client to be distributed.
The semantic representation network receives feature value vectors of a plurality of feature key value pairs in the client feature information, and processes the feature value vectors based on the trained multi-layer perceptron network to obtain client semantic vectors of clients to be distributed.
Specifically, the server performs feature extraction on the client feature information corresponding to the client to be allocated, determines a feature value vector of each feature key value pair based on a pre-trained feature vector determination strategy according to feature names and feature values of each feature key value pair, and obtains a plurality of feature value vectors. After the server splices the plurality of eigenvalue vectors, inputting the spliced eigenvalue vectors into a multi-layer perceptron network, and obtaining the characteristic representation of the clients to be distributed through the operation of the multi-layer perceptron network, namely obtaining the client semantic vectors of the clients to be distributed.
In this embodiment, the client feature information is obtained from the client feature library through the client identifier, and a plurality of feature key value pairs in the client feature information are input into the multi-layer perceptron network, so that a plurality of feature vectors are converted into client semantic vectors of clients to be allocated, and the client semantic vectors contain client features of a plurality of dimensions and can accurately represent the semantic information of the clients to be allocated.
In one embodiment, as shown in fig. 4, the specific implementation process of "searching in a pre-constructed customer semantic vector library based on the customer semantic vector of the customer to be allocated and determining a plurality of target customer data similar to the customer semantic vector of the customer to be allocated" includes:
Step S402, for each customer semantic vector contained in the customer semantic vector library, determining similarity scores between the customer semantic vector of the customer to be allocated and the customer semantic vector corresponding to the customer semantic vector library, and obtaining a plurality of similarity scores.
Specifically, the server traverses each client semantic vector in the client semantic vector library, calculates a similarity score between the client semantic vector of the client to be distributed and the currently traversed client semantic vector through a vector distance measurement strategy, and accordingly obtains a plurality of similarity scores. The vector distance metric strategy may be a euclidean distance algorithm, a cosine distance algorithm, a hamming distance algorithm, a manhattan distance algorithm, or the like.
Step S404, obtaining clients, corresponding salespersons and single labels between the clients and the salespersons, which correspond to the preset number of similarity scores, according to the sequence from large to small, and taking the single labels as a plurality of target client data.
Specifically, the server screens a preset number of similarity scores from among a plurality of similarity scores in order from large to small. The server traverses the preset number of similarity scores, obtains target client data of target clients corresponding to the similarity scores based on client semantic vectors corresponding to the similarity scores, namely, determines a plurality of target clients similar to the clients to be distributed through the similarity scores, obtains a data set containing the target clients, sales personnel and the single labels from a policy system, and further obtains a plurality of target client data.
In this embodiment, the similarity score between the client to be allocated and other clients is determined through the vector distance measurement policy, so that a plurality of target clients with highest scores are screened out, target client data of the target clients are obtained, sales personnel and single labels corresponding to the target clients are obtained, and the similarity score can represent the matching degree between every two clients, so that the target clients similar to the client to be allocated can be accurately determined through the method.
In one embodiment, as shown in fig. 5, the specific implementation process of determining a target sales person from sales persons corresponding to a plurality of target customer data based on similarity scores of customer semantic vectors of customers to be allocated, corresponding sales persons and a list forming label in each target customer data includes:
step S502, for each target customer data, determines a sales score of a sales person corresponding to the target customer data.
Wherein the sales score is determined based on the similarity score corresponding to the target customer data and the tag value of the singulated tag.
Specifically, the server traverses each target client data, calculates a product value between the similarity score in the current target client data and the tag value of the single tag, and takes the product value as a sales score corresponding to the current sales person. The label value of the singulation label is a preset value, for example, if the label value of the singulation label can be 1 or-1, 1 indicates that the sales person is in a bill with the target client, and-1 indicates that the sales person is not in a bill with the target client. And after the traversal of the target client data is completed, obtaining the sales score of each sales person.
Step S504, a plurality of sales scores corresponding to the same sales person are summed to obtain updated sales scores of all sales persons, and the sales person corresponding to the highest sales score is taken as a target sales person.
Specifically, the target client data may have a single label between the same sales person and multiple target clients, so that the sales scores corresponding to the same sales person are summed up for the same sales person to obtain the final sales score of the sales person, and the sales scores of all sales persons are obtained. The server takes the sales person with the highest score as the target sales person. If a plurality of sales persons with the same sales scores and the highest sales scores exist, one sales person is randomly selected from the sales persons as a target sales person.
In this embodiment, the sales score corresponding to each sales person is determined through the similarity score and the label value of the formed label, and the sales total score of the same sales person is combined, so that the sales score of each sales person can be obtained, and the sales person with the highest score is selected as the target sales person, so that the target sales person obtained through the method has higher matching degree with the clients to be distributed.
In one embodiment, as shown in fig. 6, the personnel allocation method further includes:
step S602, based on the client identifiers of the history clients, client feature information of a plurality of history clients is obtained from the client feature library.
Wherein the customer feature library is used for storing at least one data of a customer identifier, a customer portrait, customer behavior data, a customer relationship network and a customer source channel; the client characteristic information comprises a plurality of characteristic key value pairs, wherein the characteristic key value pairs comprise one-to-one corresponding characteristic names and characteristic values; customer characteristic information of the history customers is stored in the customer characteristic library in advance.
Specifically, the server may obtain the client identifier of the history client, and obtain the client feature information of the history client in the client feature library constructed in advance through the client identifier, so as to obtain the feature key value pair of the history client, and may also obtain the history client data corresponding to the history client, that is, obtain the sales personnel and the ticket label corresponding to the history client, and update the history client data to the client feature information.
Step S604, inputting the customer characteristic information of a plurality of history customers into a trained semantic representation network to obtain customer semantic vectors of the plurality of history customers, and constructing a customer semantic vector library based on the customer semantic vectors of the plurality of history customers.
Specifically, the server performs feature extraction on client feature information corresponding to a history client, determines a feature value vector of each feature key value pair based on a pre-trained feature vector determination strategy according to feature names and feature values of each feature key value pair, and obtains a plurality of feature value vectors. After the server splices the plurality of eigenvalue vectors, inputting the spliced eigenvalue vectors into a multi-layer perceptron network, and obtaining the customer semantic vectors of the historical customers through the operation of the multi-layer perceptron network. After the customer semantic vectors of the plurality of historic customers are obtained, they are stored to a customer semantic vector library.
In this embodiment, by converting the client feature information and the history client data of the history clients into the client semantic vectors and establishing the client semantic vector library, the feature representation of each history client can be obtained, thereby improving the accuracy of the client semantic vectors of each history client.
In one embodiment, as shown in fig. 7, the personnel allocation method further includes:
step S702, after the target sales personnel completes the service of the customer to be allocated, updating the customer to be allocated, the target sales personnel and the formed label corresponding to the customer to be allocated to the customer characteristic information of the customer to be allocated, and obtaining the updated customer characteristic information of the customer to be allocated.
Specifically, after the server distributes the target sales personnel to the clients to be distributed, the target sales personnel and the clients to be distributed finish the butt joint, and based on the butt joint result, the list forming labels corresponding to the clients to be distributed are updated, namely whether the target sales personnel and the clients to be distributed form a list or not is determined. And updating the target sales personnel and the formed ticket label to the customer characteristic information of the customer to be distributed. At this point, the customer to be allocated is already a history customer.
Step S704, inputting the updated customer characteristic information of the customers to be distributed into a trained semantic representation network to obtain customer semantic vectors of the customers to be distributed, and storing the customer semantic vectors of the customers to be distributed into a customer semantic vector library.
Specifically, the server performs feature extraction on the client feature information corresponding to the client to be allocated, determines a feature value vector of each feature key value pair based on a pre-trained feature vector determination strategy according to feature names and feature values of each feature key value pair, and obtains a plurality of feature value vectors. After the server splices the plurality of eigenvalue vectors, inputting the spliced eigenvalue vectors into a multi-layer perceptron network, obtaining customer semantic vectors of customers to be distributed through the operation of the multi-layer perceptron network, and updating the customer semantic vectors to a customer semantic vector library.
In this embodiment, the target sales personnel and the formed labels corresponding to the clients to be allocated are updated to the client semantic vectors and stored in the client semantic vector library, so that a dynamic data stream of the clients can be formed, and the newly added client semantic vectors are provided for subsequent clients to be allocated to search, thereby continuously improving timeliness and accuracy of allocation.
As shown in fig. 8, the following describes in detail a specific implementation procedure of the personnel allocation method, including the following steps:
step 1, constructing a customer semantic vector library: the server queries all features of the served clients (history clients) from the client feature library by taking the client ID as a key (primary key), inputs all the features into a semantic representation network to obtain semantic vector representations of the served clients, and adds the semantic vector representations into a client semantic vector library for indexing to form a client semantic vector library.
Step 2, customer semantic vector retrieval: for a new customer or a customer needing secondary distribution sales personnel, firstly, a server inquires all the characteristics of the customer from a customer characteristic library (the characteristics are consistent with those of the customer served in the step 1), then semantic representation is carried out on the characteristics of the customer by utilizing a semantic representation network (the semantic network is the same model as the network in the step 1) to obtain a semantic vector of the customer to be distributed, and finally, the semantic vector of the customer to be distributed is taken as query input, and the topK customer and similarity score similar to the semantic of the customer to be distributed are searched.
Step 3, screening optimal sales personnel: and (3) based on the topK clients and corresponding salespersons thereof searched in the step (2) and whether the clients are in a single label (namely, clients, salespersons or not) data set, voting and scoring are carried out on each candidate salesperson by utilizing a voting algorithm, and the salesperson with the highest score is obtained, namely, the best salesperson of the clients to be distributed.
Step 4, the served client reflows: and (3) extracting corresponding salespersons and labels of whether the salespersons are in a list or not from the clients who finish the service, calculating semantic vector representations of the served clients by using the method of the step (1), and adding the semantic vector representations into a client semantic vector library in time to form a dynamic data stream. The newly added customer semantic vector is used for the next customer semantic vector retrieval (i.e. step 2).
In one example, the technical details of the specific embodiment described above are as follows:
1. customer feature library: the database storing the characteristics of the client includes client id, client basic portrait (sex, age, academic, occupation, birth place, address, etc.), client behavior data (whether history is applied for insurance, whether claim is settled, whether life insurance APP is registered, etc.), client relationship network (family relationship, social relationship, policy relationship, etc.), client source channel, etc.
2. Semantic representation network: and a neural network model which is trained in advance. The model input is a pair of characteristic key values, the keys are characteristic names (such as sex, academic, history application, client relationship (mother), client source channels and the like), the values are characteristic values (such as men, university family, history non-application, client mother id, natural visit and the like), each characteristic value of each characteristic name is provided with a vector trained in advance in one-to-one correspondence with the vector, and the model input is the concatenation of all characteristic value vectors. The model structure is a multi-layer perceptron network. The neurons on the top layer of the model serve as characteristic representation of the client, namely the output of the semantic representation network.
3. Customer semantic vector library: the vector database is a database for storing, managing and retrieving vector data, and has the advantages of high vector retrieval efficiency and strong analysis capability. The invention represents the customer features as semantic vectors, stored in a vector library.
4. Vector retrieval: the vector is used as query input, and one or more vectors closest to the vector library are matched as search results. Common vector distance measurement methods include euclidean distance, cosine distance, hamming distance, manhattan distance, and the like. The invention takes vector distance as a customer similarity score. (vector libraries on the general market all provide vector retrieval functionality, return results with vector distance fields.)
Voting algorithm: assuming that the customer, similarity score, corresponding salesman, and singulation labels of similarity topK are (cut_1, score_1, samples_1, and Succ_1), (cut_2, score_2, samples_2, and Succ_2), (cut_3, score_3, and score_3), …, (cut_k, score_k, samples_k, and score_k), where samples_1, samples_2, samples_3, …, samples_k are salesman ids (the same salesman may be included in the topK), and that the samples_1, the samples_2, the samples_3, …, and the samples_k are singulation labels (1 or-1), where 1 represents a single item, -1 represents a non-singulation item), and the score of each salesman is Σscore, i.e., the salesman is a weighted sum of the salesman and the score in the salesman. The voting algorithm selects the sales personnel with the highest score, and if the sales personnel with the same score exist, one sales personnel is randomly selected.
In the embodiment, the semantic vector representations of the clients and the salespersons are calculated independently and in advance, so that the client distribution efficiency can be improved, the condition that the client features and the salespersons features are input to the neural network model simultaneously in the prior art is avoided, the clients and the salespersons are not required to be matched one by one, and the calculated amount is reduced.
In addition, the embodiment can directly match the clients to be distributed with the served clients, wherein the clients to be distributed are the same type of characteristic data, and corresponding semantic representation exists, so that the problem that the characteristics cannot be corresponding when the clients are matched with sales personnel is solved, and the matching accuracy is higher.
Finally, the embodiment can dynamically update the capability portraits of sales personnel through the process of the back flow of the served clients to obtain the sales score, thereby improving the matching accuracy. The situation that the most suitable salesperson cannot be screened out by matching static capacity images corresponding to salespersons with client characteristics in the related technology is avoided.
It should be understood that, although the steps in the flowcharts related to the above embodiments are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.
Based on the same inventive concept, the embodiment of the application also provides a personnel allocation device for realizing the personnel allocation method. The implementation of the solution provided by the device is similar to the implementation described in the above method, so the specific limitation in the embodiment of one or more personnel allocation devices provided below may be referred to the limitation of the personnel allocation method hereinabove, and will not be repeated here.
In one embodiment, as shown in FIG. 9, there is provided a people dispensing device 900 comprising: a semantic vector determination module 901, a customer determination module 902, and a person assignment module 903, wherein:
the semantic vector determining module 901 is configured to input the client feature information of the client to be allocated into a trained semantic representation network, so as to obtain a client semantic vector of the client to be allocated.
The client determining module 902 is configured to determine, based on a client semantic vector of a client to be allocated, a plurality of target client data similar to the client semantic vector of the client to be allocated, by searching in a pre-constructed client semantic vector library; the customer semantic vector library is determined based on historical customer data, each generated based on the customer, the corresponding sales person, and the listing labels.
The personnel allocation module 903 is configured to determine a target sales person from sales persons corresponding to the plurality of target customer data based on the similarity score, the corresponding sales person, and the ordered label of the customer semantic vector of the customer to be allocated in each target customer data, and allocate the target sales person to the customer to be allocated.
Further, the semantic vector determination module 901 is specifically configured to: based on a client identifier of a client to be allocated, acquiring client characteristic information corresponding to the client to be allocated in a client characteristic library; the customer feature library is used for storing at least one data of customer identifiers, customer portraits, customer behavior data, customer relationship networks and customer source channels; the client characteristic information comprises a plurality of characteristic key value pairs, wherein the characteristic key value pairs comprise one-to-one corresponding characteristic names and characteristic values; customer characteristic information of customers to be distributed is pre-stored in a customer characteristic library;
inputting the client characteristic information corresponding to the client to be distributed into a trained semantic representation network to obtain a client semantic vector of the client to be distributed; the semantic representation network receives feature value vectors of a plurality of feature key value pairs in the client feature information, and processes the feature value vectors based on the trained multi-layer perceptron network to obtain client semantic vectors of clients to be distributed.
Further, the client determination module 902 is specifically configured to: for each customer semantic vector contained in a customer semantic vector library, determining similarity scores between the customer semantic vector of the customer to be distributed and the customer semantic vector corresponding to the customer semantic vector library, and obtaining a plurality of similarity scores;
and acquiring clients, corresponding salespersons and single labels between the clients and the salespersons, which correspond to the preset number of similarity scores, in order from large to small, and taking the single labels as a plurality of target client data.
Further, the personnel allocation module 903 is specifically configured to: determining sales scores of sales personnel corresponding to the target client data aiming at each target client data; the sales score is determined based on the similarity score corresponding to the target customer data and the tag value of the singulated tag;
and summing a plurality of sales scores corresponding to the same sales person to obtain updated sales scores of all sales persons, and taking the sales person corresponding to the highest sales score as a target sales person.
Further, the device also comprises a vector library construction module, which is used for acquiring the client characteristic information of a plurality of history clients from the client characteristic library based on the client identifiers of the history clients;
Inputting the client characteristic information of the plurality of historical clients into a trained semantic representation network to obtain client semantic vectors of the plurality of historical clients, and constructing a client semantic vector library based on the client semantic vectors of the plurality of historical clients.
Further, the device also comprises a vector library updating module, which is used for updating the clients to be distributed, the target sales personnel and the formed labels corresponding to the clients to be distributed to the client characteristic information of the clients to be distributed after the target sales personnel finish the service of the clients to be distributed, so as to obtain the updated client characteristic information of the clients to be distributed;
and inputting the updated customer characteristic information of the customers to be distributed into a trained semantic representation network to obtain customer semantic vectors of the customers to be distributed, and storing the customer semantic vectors of the customers to be distributed into a customer semantic vector library.
The individual modules in the above-described people distribution arrangement can be realized in whole or in part by software, hardware and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a server, and the internal structure of which may be as shown in fig. 10. The computer device includes a processor, a memory, an Input/Output interface (I/O) and a communication interface. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface is connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of computer devices is used for customer characteristic information and customer semantic vectors, the customer characteristic information may include at least one of customer identifiers, customer portraits, customer behavior data, customer relationship networks, and customer source channels. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a personnel allocation method.
It will be appreciated by those skilled in the art that the structure shown in fig. 10 is merely a block diagram of some of the structures associated with the present application and is not limiting of the computer device to which the present application may be applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
In an embodiment, there is also provided a computer device comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the steps of the method embodiments described above when the computer program is executed.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored which, when executed by a processor, carries out the steps of the method embodiments described above.
In an embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the steps of the method embodiments described above.
It should be noted that, the user information (including, but not limited to, user equipment information, user personal information, etc.) and the data (including, but not limited to, data for analysis, stored data, presented data, etc.) referred to in the present application are information and data authorized by the user or sufficiently authorized by each party, and the collection, use and processing of the related data are required to comply with the related laws and regulations and standards of the related countries and regions.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the various embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magnetic random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (Phase Change Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like. The databases referred to in the various embodiments provided herein may include at least one of relational databases and non-relational databases. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic units, quantum computing-based data processing logic units, etc., without being limited thereto.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples only represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the present application. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application shall be subject to the appended claims.
Claims (10)
1. A method of personnel allocation, the method comprising:
inputting the client characteristic information of the clients to be distributed into a trained semantic representation network to obtain client semantic vectors of the clients to be distributed;
searching in a pre-constructed customer semantic vector library based on the customer semantic vector of the customer to be distributed, and determining a plurality of target customer data similar to the customer semantic vector of the customer to be distributed; the customer semantic vector library is determined based on historical customer data, each of the historical customer data being generated based on a customer, a corresponding sales person and a listing tag;
And determining target sales personnel from sales personnel corresponding to a plurality of target customer data based on similarity scores of the target customer data and the customer semantic vectors of the customers to be distributed, corresponding sales personnel and the single labels, and distributing the target sales personnel to the customers to be distributed.
2. The method according to claim 1, wherein said inputting the client characteristic information of the client to be distributed into the trained semantic representation network to obtain the client semantic vector of the client to be distributed comprises:
acquiring client characteristic information corresponding to the client to be distributed in a client characteristic library based on the client identifier of the client to be distributed; the customer feature library is used for storing at least one data of a customer identifier, a customer portrait, customer behavior data, a customer relationship network and a customer source channel; the client characteristic information comprises a plurality of characteristic key value pairs, wherein the characteristic key value pairs comprise one-to-one corresponding characteristic names and characteristic values; the client characteristic information of the clients to be distributed is stored in the client characteristic library in advance;
inputting the client characteristic information corresponding to the client to be distributed into a trained semantic representation network to obtain a client semantic vector of the client to be distributed; the semantic representation network receives the characteristic value vectors of a plurality of characteristic key value pairs in the customer characteristic information, and processes each characteristic value vector based on the trained multi-layer perceptron network to obtain the customer semantic vector of the customer to be distributed.
3. The method of claim 1, wherein the retrieving, based on the customer semantic vector of the customer to be assigned, in a pre-constructed customer semantic vector library, determines a plurality of target customer data that are similar to the customer semantic vector of the customer to be assigned, comprises:
determining similarity scores between the client semantic vectors of the clients to be distributed and the client semantic vectors corresponding to the client semantic vector library according to the client semantic vectors contained in the client semantic vector library, so as to obtain a plurality of similarity scores;
and acquiring clients, corresponding salespersons and single labels between the clients and the salespersons, which correspond to the preset number of similarity scores, in order from large to small, and taking the single labels as a plurality of target client data.
4. The method of claim 1, wherein the determining the target sales person from sales persons corresponding to the plurality of target customer data based on the similarity score, the corresponding sales person, and the listing label of the customer semantic vector of the customer to be assigned in each of the target customer data comprises:
determining sales scores of sales personnel corresponding to the target client data for each target client data; the sales score is determined based on the similarity score corresponding to the target customer data and the label value of the singulation label;
And summing a plurality of sales scores corresponding to the same sales person to obtain updated sales scores of all sales persons, and taking the sales person corresponding to the highest sales score as a target sales person.
5. The method according to claim 1, wherein the method further comprises:
acquiring client characteristic information of a plurality of historical clients from a client characteristic library based on client identifiers of the historical clients;
inputting the customer characteristic information of a plurality of history customers into a trained semantic representation network to obtain customer semantic vectors of the history customers, and constructing the customer semantic vector library based on the customer semantic vectors of the history customers.
6. The method according to any one of claims 1 to 5, further comprising:
after the target sales person finishes the service of the to-be-allocated clients, updating the to-be-allocated clients, the target sales person and the list forming labels corresponding to the to-be-allocated clients to the client characteristic information of the to-be-allocated clients to obtain updated client characteristic information of the to-be-allocated clients;
and inputting the updated customer characteristic information of the customers to be distributed into a trained semantic representation network to obtain customer semantic vectors of the customers to be distributed, and storing the customer semantic vectors of the customers to be distributed into a customer semantic vector library.
7. A personnel dispensing apparatus, the apparatus comprising:
the semantic vector determining module is used for inputting the client characteristic information of the clients to be distributed into the trained semantic representation network to obtain the client semantic vectors of the clients to be distributed;
the client determining module is used for searching in a pre-constructed client semantic vector library based on the client semantic vector of the client to be distributed, and determining a plurality of target client data similar to the client semantic vector of the client to be distributed; the customer semantic vector library is determined based on historical customer data, each of the historical customer data being generated based on a customer, a corresponding sales person and a listing tag;
and the personnel allocation module is used for determining target sales personnel from sales personnel corresponding to a plurality of target customer data based on the similarity scores of the customer semantic vectors of the customers to be allocated in the target customer data, corresponding sales personnel and the single labels, and allocating the target sales personnel to the customers to be allocated.
8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 6 when the computer program is executed.
9. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.
10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311731675.9A CN117709968A (en) | 2023-12-15 | 2023-12-15 | Personnel allocation method, apparatus, computer device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311731675.9A CN117709968A (en) | 2023-12-15 | 2023-12-15 | Personnel allocation method, apparatus, computer device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117709968A true CN117709968A (en) | 2024-03-15 |
Family
ID=90163376
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311731675.9A Pending CN117709968A (en) | 2023-12-15 | 2023-12-15 | Personnel allocation method, apparatus, computer device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117709968A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117973818A (en) * | 2024-04-01 | 2024-05-03 | 四川三瑞智慧建设工程有限公司 | Internet of things data processing method based on artificial intelligence |
-
2023
- 2023-12-15 CN CN202311731675.9A patent/CN117709968A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117973818A (en) * | 2024-04-01 | 2024-05-03 | 四川三瑞智慧建设工程有限公司 | Internet of things data processing method based on artificial intelligence |
| CN117973818B (en) * | 2024-04-01 | 2024-06-07 | 四川三瑞智慧建设工程有限公司 | Internet of things data processing method based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11968105B2 (en) | Systems and methods for social graph data analytics to determine connectivity within a community | |
| KR101009830B1 (en) | Compatibility scoring of users in a social network | |
| CN111177473B (en) | Personnel relationship analysis method, device and readable storage medium | |
| CN113722611B (en) | Recommendation method, device and equipment for government affair service and computer readable storage medium | |
| CN111506820B (en) | Recommendation model, recommendation method, recommendation device, recommendation equipment and recommendation storage medium | |
| CN117709968A (en) | Personnel allocation method, apparatus, computer device and storage medium | |
| CN112711713B (en) | Interest point recommendation and display method and device, computer equipment and storage medium | |
| CN112131261A (en) | Community query method and device based on community network and computer equipment | |
| CN113705698B (en) | Information pushing method and device based on click behavior prediction | |
| CN113656699B (en) | User feature vector determining method, related equipment and medium | |
| CN115758271A (en) | Data processing method, data processing device, computer equipment and storage medium | |
| CN113448876B (en) | Service testing method, device, computer equipment and storage medium | |
| CN113688314B (en) | Physical therapy store recommending method and device | |
| CN115391666A (en) | Hospital online recommendation method and device, computer equipment and storage medium | |
| CN115169637A (en) | Social relationship prediction method, device, equipment and medium | |
| CN111291196A (en) | Method and device for improving knowledge graph and method and device for processing data | |
| Hou et al. | BFRecSys: A Blockchain-based Federated Matrix Factorization for Recommendation Systems | |
| CN116701972B (en) | Service data processing method, device, equipment and medium | |
| US20230177581A1 (en) | Product metadata suggestion using embeddings | |
| Al-Kerboly et al. | Clustering Algorithms Comparison for University of Anbar Researchers’ Google Scholar Profiles | |
| CN116401414A (en) | Layout and display method, device, equipment and medium of page modules in page | |
| CN115422214A (en) | Corpus updating method, apparatus, computer device, storage medium and product | |
| CN116955429A (en) | Service recommendation method, device, equipment and storage medium | |
| CN117556474A (en) | Data processing method, device, computer equipment and storage medium | |
| CN115659005A (en) | Product pushing method and device, computer equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |