CN116595190A - Method and device for mining diving guests - Google Patents
Method and device for mining diving guests Download PDFInfo
- Publication number
- CN116595190A CN116595190A CN202310554312.6A CN202310554312A CN116595190A CN 116595190 A CN116595190 A CN 116595190A CN 202310554312 A CN202310554312 A CN 202310554312A CN 116595190 A CN116595190 A CN 116595190A
- Authority
- CN
- China
- Prior art keywords
- data
- scene
- community
- customer
- model
- 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 84
- 238000005065 mining Methods 0.000 title claims abstract description 58
- 230000009189 diving Effects 0.000 title description 5
- 238000012545 processing Methods 0.000 claims abstract description 50
- 238000013145 classification model Methods 0.000 claims abstract description 41
- 238000007781 pre-processing Methods 0.000 claims abstract description 19
- 238000004140 cleaning Methods 0.000 claims description 21
- 238000012549 training Methods 0.000 claims description 17
- 238000004590 computer program Methods 0.000 claims description 10
- 238000009412 basement excavation Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 6
- 230000010354 integration Effects 0.000 claims description 2
- 238000006243 chemical reaction Methods 0.000 abstract description 6
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000000873 masking effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000002829 reductive effect Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000004148 unit process Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Biomedical Technology (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Animal Behavior & Ethology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application provides a method and a device for mining a diver, wherein the method comprises the following steps: acquiring internal and external data of an enterprise operation scene; preprocessing the internal and external data to obtain preprocessed data; processing the pre-processed data through a pre-constructed text classification model to obtain model processing data; identifying a scene where a customer is located according to a preset map community algorithm and model processing data; identifying a community scene where the potential customer is located according to the scene where the customer is located; and determining the mining result of the potential passenger according to the community scene of the potential passenger. Therefore, the method and the device can utilize limited customer data to mine the diver, can determine social scenes of the diver, are good in applicability, and are further beneficial to realizing follow-up accurate conversion marketing.
Description
Technical Field
The application relates to the technical field of data processing, in particular to a method and a device for mining a diving guest.
Background
Currently, with the development of internet technology, a large number of users are accumulated in an internet platform, and when the users use services in the internet platform, a large amount of data is generated. The existing method for mining the potential customers generally uses known case customers as a benchmark, and uses customer characteristic data to find customers similar to the case customers as the most common mining mode, so that characteristics used when the customers are judged to be similar in a mining model directly determine the quality of the final mined potential customers. However, in practice, the existing method is poor in applicability, and the social scene of the diver cannot be determined, so that the follow-up accurate conversion marketing cannot be realized.
Disclosure of Invention
The embodiment of the application aims to provide a method and a device for mining a potential passenger, which can be used for mining the potential passenger by using limited client data, can determine the social scene of the potential passenger, have good applicability and are further beneficial to realizing the follow-up accurate conversion marketing.
The first aspect of the embodiment of the application provides a method for mining a diver, which comprises the following steps:
acquiring internal and external data of an enterprise operation scene;
preprocessing the internal and external data to obtain preprocessed data;
processing the preprocessing data through a pre-constructed text classification model to obtain model processing data;
identifying a scene where a customer is located according to a preset map community algorithm and the model processing data;
identifying a community scene where the potential customer is located according to the scene where the customer is located;
and determining a potential passenger mining result according to the community scene of the potential passenger.
In the implementation process, the method can obtain the internal and external data of the enterprise operation scene preferentially; preprocessing the internal and external data to obtain preprocessed data; then, processing the preprocessed data through a pre-constructed text classification model to obtain model processing data; then, identifying a scene where the client is located according to a preset map community algorithm and model processing data; when the scene of the client is identified, identifying a community scene of the potential client according to the scene of the client; and finally, determining the mining result of the potential passenger according to the community scene of the potential passenger. Therefore, the method can utilize limited client data to mine the potential customers, can determine social scenes of the potential customers, has good applicability, and is further beneficial to realizing subsequent accurate conversion marketing.
Further, before the acquiring the internal and external data of the enterprise operation scene, the method further includes:
constructing an original classification model;
pre-training the original classification model by RoBERTa based on a transducer frame to obtain a text classification model;
the model super-parameters of the text classification model comprise a learning rate parameter, a batch size parameter and a training period parameter; the tuning mode of the text classification model is grid search.
Further, the internal and external data includes one or more of an enterprise customer name, an industrial and commercial operation scope, an industry to which an enterprise belongs, or an industry chain to which an enterprise belongs.
Further, the identifying the scene of the client according to the preset map community algorithm and the model processing data includes:
acquiring relationship type data from the model processing data; wherein the relational data at least comprises bill data, dynamic account data and bidding data;
integrating the relationship type data to obtain integrated data;
performing data cleaning on the integrated data to obtain cleaning data;
constructing a relationship graph according to the cleaning data;
carrying out community division according to a preset louvain algorithm and the relation graph to obtain a community grouping result;
and determining the scene of the client according to the community grouping result.
Further, the identifying the social scene where the potential customer is located according to the scene where the customer is located includes:
determining a target diver according to the model processing data;
and determining a community scene where the target potential customer is located according to the scene where the customer is located.
A second aspect of an embodiment of the present application provides a diver excavation apparatus, including:
the acquisition unit is used for acquiring the internal and external data of the enterprise operation scene;
the preprocessing unit is used for preprocessing the internal and external data to obtain preprocessed data;
the model processing unit is used for processing the preprocessing data through a pre-constructed text classification model to obtain model processing data;
the first identification unit is used for identifying a scene where a client is located according to a preset map community algorithm and the model processing data;
the second identification unit is used for identifying a community scene where the potential customer is located according to the scene where the customer is located;
and the determining unit is used for determining the mining result of the potential passenger according to the community scene of the potential passenger.
In the implementation process, the device can acquire the internal and external data of the enterprise operation scene through the acquisition unit; preprocessing the internal and external data through a preprocessing unit to obtain preprocessed data; the method comprises the steps that a model processing unit processes pretreatment data through a pre-constructed text classification model to obtain model processing data; the scene of the customer is identified according to a preset map community algorithm and model processing data through a first identification unit; identifying a community scene where the potential customer is located according to the scene where the customer is located through a second identification unit; and determining the mining result of the potential passenger according to the community scene of the potential passenger by a determining unit. Therefore, the device can utilize limited customer data to carry out the diving and mining, and can confirm the social scene of diving, and the suitability is good, and then is favorable to realizing subsequent accurate conversion marketing.
Further, the diver excavation apparatus further includes:
the model construction unit is used for constructing an original classification model before the internal and external data of the enterprise operation scene are acquired;
the pre-training unit is used for pre-training the original classification model by adopting RoBERTa based on a transducer frame to obtain a text classification model; the model super-parameters of the text classification model comprise a learning rate parameter, a batch size parameter and a training period parameter; the tuning mode of the text classification model is grid search.
Further, the internal and external data includes one or more of an enterprise customer name, an industrial and commercial operation scope, an industry to which an enterprise belongs, or an industry chain to which an enterprise belongs.
Further, the first identifying unit includes:
an acquisition subunit, configured to acquire relationship type data from the model processing data; wherein the relational data at least comprises bill data, dynamic account data and bidding data;
the integration subunit is used for integrating the relationship type data to obtain integrated data;
the cleaning subunit is used for carrying out data cleaning on the integrated data to obtain cleaning data;
a construction subunit, configured to construct a relationship graph according to the cleaning data;
the division subunit is used for carrying out community division according to a preset louvain algorithm and the relation graph to obtain a community grouping result;
and the first determination subunit is used for determining the scene of the client according to the community grouping result.
Further, the second identifying unit includes:
a second determining subunit, configured to determine a target diver according to the model processing data;
and the identification subunit is used for determining a community scene where the target potential customer is located according to the scene where the customer is located.
A third aspect of the embodiment of the present application provides an electronic device, including a memory and a processor, where the memory is configured to store a computer program, and the processor is configured to execute the computer program to cause the electronic device to execute the method for mining a potential passenger according to any one of the first aspect of the embodiment of the present application.
A fourth aspect of the embodiments of the present application provides a computer readable storage medium storing computer program instructions which, when read and executed by a processor, perform the method of mining a potential passenger according to any one of the first aspect of the embodiments of the present application.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments of the present application will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and should not be considered as limiting the scope, and other related drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of a method for mining a diver according to an embodiment of the present application;
FIG. 2 is a schematic flow chart of another method for mining a diver according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of a device for mining a diver according to an embodiment of the present application;
fig. 4 is a schematic structural diagram of another underwater excavation apparatus according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be described below with reference to the accompanying drawings in the embodiments of the present application.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures. Meanwhile, in the description of the present application, the terms "first", "second", and the like are used only to distinguish the description, and are not to be construed as indicating or implying relative importance.
Example 1
Referring to fig. 1, fig. 1 is a schematic flow chart of a method for mining a diver according to the present embodiment. The method for mining the potential guests comprises the following steps:
s101, acquiring internal and external data of an enterprise operation scene.
S102, preprocessing the internal and external data to obtain preprocessed data.
S103, processing the preprocessed data through a pre-constructed text classification model to obtain model processing data.
S104, identifying the scene where the client is located according to a preset map community algorithm and model processing data.
S105, identifying a community scene where the potential customer is located according to the scene where the customer is located.
S106, determining a potential passenger mining result according to a community scene where the potential passenger is located.
In this embodiment, the method may be applied to the financial field. Specifically, the method can acquire client information of a specified client in a banking system, and then acquire 6 data corresponding to acquiring bills, dynamic accounts, bidding and the like. Clients having the same type of data or the same data relationship are then mined based on this, and the business content and business scope of the clients are further combined to determine whether the mined clients are truly potential clients. After the determination is completed, the method can obtain high-quality potential clients, so that the subsequent accurate conversion and marketing are facilitated.
In the embodiment, the method can search similar clients by utilizing a natural language processing related technology, select communities by combining with a graph technology, and finally judge the concentrations of related industrial clients in different industries in communities to perform the recognition of the potential scene.
In this embodiment, the execution subject of the method may be a computing device such as a computer or a server, which is not limited in this embodiment.
In this embodiment, the execution body of the method may be an intelligent device such as a smart phone or a tablet computer, which is not limited in this embodiment.
Therefore, by implementing the method for mining the client described in this embodiment, unstructured text data and relationship data reserved by the client, which cannot be effectively utilized in conventional client mining, can be fully utilized, so that the method can complete client mining under the condition of less client data by using the latest AI technology in combination with actual application scenes. Meanwhile, the potential customer mining logic of the method is stronger in interpretability, and based on a similar concept, the method for judging the enterprise similarity by using enterprise customer operation data can be closer to the actual production condition of an enterprise; in addition, based on the concept of the category, the method of identifying the affiliated business scene after the client relationship data are clustered through communities can be closer to the actual business situation of the enterprise client.
Example 2
Referring to fig. 2, fig. 2 is a schematic flow chart of a method for mining a diver according to the present embodiment. The method for mining the potential guests comprises the following steps:
s201, constructing an original classification model.
S202, a RoBERTa pre-training original classification model based on a transducer frame is adopted, and a text classification model is obtained.
In this embodiment, the model super-parameters of the text classification model include a learning rate parameter, a batch size parameter, and a training period parameter; the tuning way of the text classification model is grid search.
In this embodiment, the method may use a RoBERTa pre-training model based on a transducer framework, where model parameters 230M are partially optimized compared to BERT. Such as: the batch size is changed from 256 to 2k, the training deployment is reduced, static masking is changed into dynamic masking when Mask is used, and a larger byte-level BPE dictionary is used when encoding.
In this embodiment, since the case client is currently available as a positive sample and there is no negative sample reference, the method uses the client of the conventional first industry as a negative sample according to the business experience, thereby defining the target value.
In this embodiment, since there are many model super parameters, tuning of the model parameters is required. Specifically, the model hyper-parameters relate to 3 model hyper-parameters including learning rate, batch size and training period number, and the adopted tuning mode is grid search.
In this embodiment, the method may evaluate the effect of the model application. Specifically, the method can group model prediction probability distribution on a test set, so that the industry probability of the top three of target clients is 2.5 times of the other probabilities, and the model accuracy rate reaches more than 85.
S203, acquiring the internal and external data of the enterprise operation scene.
In this embodiment, the internal and external data includes one or more of an enterprise client name, an industrial and commercial operation scope, an industry to which an enterprise belongs, or an industry chain to which an enterprise belongs.
S204, preprocessing the internal and external data to obtain preprocessed data.
In this embodiment, the method may perform feature preprocessing on unstructured data of an enterprise. The method can identify the internal and external data of the enterprise business scenario, wherein the internal and external data comprises an enterprise client name, an industrial and commercial business scope, an enterprise belonging industry or an industry chain. Specifically, the data are in unstructured text form, the text needs to be preprocessed before splicing, and the processing modes include, but are not limited to, special character replacement, stop word processing, text splicing symbol unification and the like; according to the method, the text length is not limited due to model selection used later, so that the spliced text does not need to be subjected to target length truncation.
S205, processing the preprocessed data through a pre-constructed text classification model to obtain model processing data.
S206, obtaining relation type data from the model processing data.
In this embodiment, the relationship data includes at least ticket data, ledger data, and bid and ask data.
S207, integrating the relationship type data to obtain integrated data.
And S208, data cleaning is carried out on the integrated data to obtain cleaning data.
S209, constructing a relation map according to the cleaning data.
S210, community division is carried out according to a preset louvain algorithm and a relation graph, and a community grouping result is obtained.
S211, determining the scene of the client according to the community grouping result.
In this embodiment, the method may identify a scene in which the client is located based on a graph community algorithm.
In this embodiment, the method may select 6 kinds of relationship type data related to bills, dynamic accounts, bidding, etc., integrate the data, clean the data, and reject part of non-target enterprises to form a relationship network of 350 tens of thousands of nodes and 1300 tens of thousands of edges. It can be seen that the method can build a map as such.
In this embodiment, in the community grouping process, a louvain algorithm may be selected for community division. The louvain is a greedy algorithm based on multi-level optimization modularity, so that the problem of a large community is solved through multiple iterations. Meanwhile, the method can further iterate the preliminarily divided communities until the number of nodes in the community with the maximum volume is less than 5 ten thousand, and finally, 4000 communities are formed through 3 rounds of division.
S212, determining the target diver according to the model processing data.
S213, determining a community scene where the target potential customer is located according to the scene where the customer is located.
In this embodiment, the method may select clients with a probability of 0.9 or more as the potential guests through the above steps, and locate a community in which the target potential guests are located. Based on keyword matching of business content of enterprises, the method can classify business ranges of other enterprises in the community of the potential customers, divide the potential customers into 4 scenes, and respectively conduct further marketing of corresponding scenes. The scene keywords can adopt business or industry experience, and the potential passenger probability threshold value can be set according to the number of potential passengers planned to be marketed.
S214, determining a potential passenger mining result according to the community scene where the potential passenger is located.
In this embodiment, the execution subject of the method may be a computing device such as a computer or a server, which is not limited in this embodiment.
In this embodiment, the execution body of the method may be an intelligent device such as a smart phone or a tablet computer, which is not limited in this embodiment.
Therefore, by implementing the method for mining the client described in this embodiment, unstructured text data and relationship data reserved by the client, which cannot be effectively utilized in conventional client mining, can be fully utilized, so that the method can complete client mining under the condition of less client data by using the latest AI technology in combination with actual application scenes. Meanwhile, the potential customer mining logic of the method is stronger in interpretability, and based on a similar concept, the method for judging the enterprise similarity by using enterprise customer operation data can be closer to the actual production condition of an enterprise; in addition, based on the concept of the category, the method of identifying the affiliated business scene after the client relationship data are clustered through communities can be closer to the actual business situation of the enterprise client.
Example 3
Referring to fig. 3, fig. 3 is a schematic structural diagram of a device for mining a submarine according to the present embodiment. As shown in fig. 3, the diver excavation apparatus includes:
an obtaining unit 310, configured to obtain internal and external data of an enterprise operation scenario;
a preprocessing unit 320, configured to preprocess the internal and external data to obtain preprocessed data;
the model processing unit 330 is configured to process the pre-processed data through a pre-constructed text classification model to obtain model processing data;
the first identifying unit 340 is configured to identify a scene where the client is located according to a preset graph community algorithm and model processing data;
a second identifying unit 350, configured to identify a community scene where the potential customer is located according to a scene where the customer is located;
the determining unit 360 is configured to determine a mining result of the potential passenger according to a community scene where the potential passenger is located.
In this embodiment, the explanation of the device for mining the diver may refer to the description in embodiment 1 or embodiment 2, and the description is not repeated in this embodiment.
Therefore, by implementing the device for mining the potential customers described in this embodiment, unstructured text data and relationship data reserved by customers, which cannot be effectively utilized in conventional potential customers mining, can be fully utilized, so that the method can complete customer mining under the condition of less customer data by using the latest AI technology in combination with actual application scenes. Meanwhile, the potential customer mining logic of the method is stronger in interpretability, and based on a similar concept, the method for judging the enterprise similarity by using enterprise customer operation data can be closer to the actual production condition of an enterprise; in addition, based on the concept of the category, the method of identifying the affiliated business scene after the client relationship data are clustered through communities can be closer to the actual business situation of the enterprise client.
Example 4
Referring to fig. 4, fig. 4 is a schematic structural diagram of a device for mining a submarine according to the present embodiment. As shown in fig. 4, the diver excavation apparatus includes:
an obtaining unit 310, configured to obtain internal and external data of an enterprise operation scenario;
a preprocessing unit 320, configured to preprocess the internal and external data to obtain preprocessed data;
the model processing unit 330 is configured to process the pre-processed data through a pre-constructed text classification model to obtain model processing data;
the first identifying unit 340 is configured to identify a scene where the client is located according to a preset graph community algorithm and model processing data;
a second identifying unit 350, configured to identify a community scene where the potential customer is located according to a scene where the customer is located;
the determining unit 360 is configured to determine a mining result of the potential passenger according to a community scene where the potential passenger is located.
As an alternative embodiment, the diver's excavating device further comprises:
a model construction unit 370 for constructing an original classification model before acquiring the internal and external data of the enterprise business scenario;
a pre-training unit 380, configured to pre-train the original classification model by using a RoBERTa based on a transducer framework, to obtain a text classification model; the model super-parameters of the text classification model comprise a learning rate parameter, a batch size parameter and a training period parameter; the tuning way of the text classification model is grid search.
In this embodiment, the internal and external data includes one or more of an enterprise client name, an industrial and commercial operation scope, an industry to which an enterprise belongs, or an industry chain to which an enterprise belongs.
As an alternative embodiment, the first recognition unit 340 includes:
an obtaining subunit 341, configured to obtain relationship type data from the model processing data; wherein the relational data at least comprises bill data, dynamic account data and bidding data;
an integrating subunit 342, configured to integrate the relationship type data to obtain integrated data;
a cleaning subunit 343, configured to perform data cleaning on the integrated data to obtain cleaning data;
a construction subunit 344 for constructing a relationship graph from the cleaning data;
a dividing sub-unit 345, configured to divide communities according to a preset louvain algorithm and a relationship graph, so as to obtain community grouping results;
the first determining subunit 346 is configured to determine, according to the community grouping result, a scene in which the client is located.
As an alternative embodiment, the second recognition unit 350 includes:
a second determining subunit 351 configured to determine a target diver according to the model processing data;
and the identifying subunit 352 is configured to determine a social scene where the target potential passenger is located according to the scene where the client is located.
In this embodiment, the explanation of the device for mining the diver may refer to the description in embodiment 1 or embodiment 2, and the description is not repeated in this embodiment.
Therefore, by implementing the device for mining the potential customers described in this embodiment, unstructured text data and relationship data reserved by customers, which cannot be effectively utilized in conventional potential customers mining, can be fully utilized, so that the method can complete customer mining under the condition of less customer data by using the latest AI technology in combination with actual application scenes. Meanwhile, the potential customer mining logic of the method is stronger in interpretability, and based on a similar concept, the method for judging the enterprise similarity by using enterprise customer operation data can be closer to the actual production condition of an enterprise; in addition, based on the concept of the category, the method of identifying the affiliated business scene after the client relationship data are clustered through communities can be closer to the actual business situation of the enterprise client.
An embodiment of the present application provides an electronic device, including a memory and a processor, where the memory is configured to store a computer program, and the processor is configured to execute the computer program to cause the electronic device to execute a method for mining a diver in embodiment 1 or embodiment 2 of the present application.
An embodiment of the present application provides a computer readable storage medium storing computer program instructions that, when read and executed by a processor, perform the method of mining a potential customer of embodiment 1 or embodiment 2 of the present application.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. The apparatus embodiments described above are merely illustrative, for example, of the flowcharts and block diagrams in the figures that illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. 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 and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present application may be integrated together to form a single part, or each module may exist alone, or two or more modules may be integrated to form a single part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The above description is only an example of the present application and is not intended to limit the scope of the present application, and various modifications and variations will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application. It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures.
The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
Claims (10)
1. A method of mining a diver comprising:
acquiring internal and external data of an enterprise operation scene;
preprocessing the internal and external data to obtain preprocessed data;
processing the preprocessing data through a pre-constructed text classification model to obtain model processing data;
identifying a scene where a customer is located according to a preset map community algorithm and the model processing data;
identifying a community scene where the potential customer is located according to the scene where the customer is located;
and determining a potential passenger mining result according to the community scene of the potential passenger.
2. The method of claim 1, further comprising, prior to said obtaining the internal and external data of the enterprise business scenario:
constructing an original classification model;
pre-training the original classification model by RoBERTa based on a transducer frame to obtain a text classification model;
the model super-parameters of the text classification model comprise a learning rate parameter, a batch size parameter and a training period parameter; the tuning mode of the text classification model is grid search.
3. The method of claim 1, wherein the internal and external data comprises one or more of an enterprise customer name, an industrial and commercial business scope, an enterprise industry, or an industry chain to which an enterprise belongs.
4. The method of claim 1, wherein the identifying the scene of the customer according to the preset graph community algorithm and the model processing data comprises:
acquiring relationship type data from the model processing data; wherein the relational data at least comprises bill data, dynamic account data and bidding data;
integrating the relationship type data to obtain integrated data;
performing data cleaning on the integrated data to obtain cleaning data;
constructing a relationship graph according to the cleaning data;
carrying out community division according to a preset louvain algorithm and the relation graph to obtain a community grouping result;
and determining the scene of the client according to the community grouping result.
5. The method of claim 1, wherein the identifying a community scenario in which the potential customer is located according to the scenario in which the customer is located comprises:
determining a target diver according to the model processing data;
and determining a community scene where the target potential customer is located according to the scene where the customer is located.
6. A diver excavation device, comprising:
the acquisition unit is used for acquiring the internal and external data of the enterprise operation scene;
the preprocessing unit is used for preprocessing the internal and external data to obtain preprocessed data;
the model processing unit is used for processing the preprocessing data through a pre-constructed text classification model to obtain model processing data;
the first identification unit is used for identifying a scene where a client is located according to a preset map community algorithm and the model processing data;
the second identification unit is used for identifying a community scene where the potential customer is located according to the scene where the customer is located;
and the determining unit is used for determining the mining result of the potential passenger according to the community scene of the potential passenger.
7. The diver mining device of claim 6, further comprising:
the model construction unit is used for constructing an original classification model before the internal and external data of the enterprise operation scene are acquired;
the pre-training unit is used for pre-training the original classification model by adopting RoBERTa based on a transducer frame to obtain a text classification model; the model super-parameters of the text classification model comprise a learning rate parameter, a batch size parameter and a training period parameter; the tuning mode of the text classification model is grid search.
8. The diver mining device of claim 6, wherein the first recognition unit comprises:
an acquisition subunit, configured to acquire relationship type data from the model processing data; wherein the relational data at least comprises bill data, dynamic account data and bidding data;
the integration subunit is used for integrating the relationship type data to obtain integrated data;
the cleaning subunit is used for carrying out data cleaning on the integrated data to obtain cleaning data;
a construction subunit, configured to construct a relationship graph according to the cleaning data;
the division subunit is used for carrying out community division according to a preset louvain algorithm and the relation graph to obtain a community grouping result;
and the first determination subunit is used for determining the scene of the client according to the community grouping result.
9. An electronic device comprising a memory for storing a computer program and a processor that runs the computer program to cause the electronic device to perform the method of diver mining of any of claims 1 to 5.
10. A readable storage medium having stored therein computer program instructions which, when read and executed by a processor, perform the method of diver mining of any of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310554312.6A CN116595190A (en) | 2023-05-16 | 2023-05-16 | Method and device for mining diving guests |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310554312.6A CN116595190A (en) | 2023-05-16 | 2023-05-16 | Method and device for mining diving guests |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116595190A true CN116595190A (en) | 2023-08-15 |
Family
ID=87598584
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310554312.6A Pending CN116595190A (en) | 2023-05-16 | 2023-05-16 | Method and device for mining diving guests |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116595190A (en) |
-
2023
- 2023-05-16 CN CN202310554312.6A patent/CN116595190A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111859960B (en) | Semantic matching method, device, computer equipment and medium based on knowledge distillation | |
| CN110598206B (en) | Text semantic recognition method and device, computer equipment and storage medium | |
| CN112434535B (en) | Element extraction method, device, equipment and storage medium based on multiple models | |
| CN110310114B (en) | Object classification method, device, server and storage medium | |
| CN114357197B (en) | Event reasoning method and device | |
| CN112907358A (en) | Loan user credit scoring method, loan user credit scoring device, computer equipment and storage medium | |
| CN114742016A (en) | Chapter-level event extraction method and device based on multi-granularity entity differential composition | |
| CN113988195A (en) | Private domain traffic clue mining method and device, vehicle and readable medium | |
| CN113011156A (en) | Quality inspection method, device and medium for audit text and electronic equipment | |
| CN112926308A (en) | Method, apparatus, device, storage medium and program product for matching text | |
| CN112241458A (en) | Text knowledge structuring processing method, device, equipment and readable storage medium | |
| CN109635289B (en) | Entry classification method and audit information extraction method | |
| CN116702765A (en) | Event extraction method and device and electronic equipment | |
| CN116089584A (en) | Event extraction method, device, electronic equipment and medium | |
| CN116304014A (en) | Method for training entity type recognition model, entity type recognition method and device | |
| CN114969253A (en) | Market subject and policy matching method and device, computing device and medium | |
| US11880394B2 (en) | System and method for machine learning architecture for interdependence detection | |
| CN115409011A (en) | Small sample nesting relation extraction method based on dynamic prompt learning | |
| CN116595190A (en) | Method and device for mining diving guests | |
| CN114328841A (en) | Question-answer model training method and device, question-answer method and device | |
| CN114610576A (en) | Log generation monitoring method and device | |
| CN111046934B (en) | SWIFT message soft clause recognition method and device | |
| CN114254622A (en) | Intention identification method and device | |
| CN115186096A (en) | Recognition method, device, medium and electronic equipment for specific type word segmentation | |
| CN112115258A (en) | User credit evaluation method, device, server 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 |