CN109544100B - Case processing reminding method, device, equipment and medium based on deep learning - Google Patents
Case processing reminding method, device, equipment and medium based on deep learning Download PDFInfo
- Publication number
- CN109544100B CN109544100B CN201811252691.9A CN201811252691A CN109544100B CN 109544100 B CN109544100 B CN 109544100B CN 201811252691 A CN201811252691 A CN 201811252691A CN 109544100 B CN109544100 B CN 109544100B
- Authority
- CN
- China
- Prior art keywords
- case
- attribute information
- deep learning
- learning model
- target
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000013135 deep learning Methods 0.000 title abstract description 3
- 238000013136 deep learning model Methods 0.000 claims abstract description 125
- 238000004590 computer program Methods 0.000 claims description 17
- 238000009795 derivation Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 230000007246 mechanism Effects 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000008520 organization Effects 0.000 description 3
- 238000000605 extraction Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/109—Time management, e.g. calendars, reminders, meetings or time accounting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/107—Computer-aided management of electronic mailing [e-mailing]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Data Mining & Analysis (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Technology Law (AREA)
- Computer Hardware Design (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a case processing reminding method, a device, computer equipment and a storage medium based on deep learning. Because the deep learning model is obtained by training historical attribute information and historical case processing types as samples, the target case processing types can be accurately obtained, and after customer service personnel of a designated terminal receive reminding mails, complaint cases with the target case processing types of emergency types can be processed at the first time, so that the efficiency of processing the emergency complaint cases is improved.
Description
Technical Field
The invention relates to the field of big data processing, in particular to a case processing reminding method, device, equipment and medium based on a deep learning model.
Background
At present, in the security company, along with the increasing of cases, if complaints are increasing, the efficiency of case processing is required to be higher and higher, and especially the efficiency of case processing is required to be higher and higher.
In general, an insurance company sets a special database to store attribute information of complaints, a flow of processing the complaints involves a plurality of links and a plurality of customer service staff, after a previous customer service staff finishes a first work task, the first work task is transferred to a next customer service staff to process a second work task, and the second work task is sequentially carried out, if one of the links has a problem, the efficiency of processing the complaints is affected. In general, customer service personnel process complaint cases according to the obtained sequence of complaint cases, and because the emergency degree of the emergency complaint cases is higher than that of non-emergency complaint cases, the emergency complaint cases are required to be processed preferentially. However, the number of complaints handled by the customer service staff is huge, and the customer service staff easily forgets to handle the emergency complaints before handling the non-emergency complaints when the customer service staff is in tension to handle a large number of complaints, and in the traditional operation, only when the customer service staff recall that the emergency complaints need to be handled by the customer service staff, the customer service staff can handle the emergency complaints, so that the emergency complaints cannot be handled in time.
Disclosure of Invention
The embodiment of the invention provides a case processing reminding method and device based on a deep learning model, computer equipment and a storage medium, which are used for solving the problem that emergency complaint cases cannot be processed in time.
A case processing reminding method based on a deep learning model comprises the following steps:
acquiring a complaint case to be processed, sender information and recipient information;
the target attribute information of the complaint case to be processed is derived from a specified database;
inputting the target attribute information into a pre-trained deep learning model to obtain a target case processing type output by the deep learning model, wherein the deep learning model is obtained by training historical attribute information and historical case processing type as samples;
judging whether the target case processing type is an emergency type or not;
when the target case processing type is the emergency type, generating a case reminding mail according to the target attribute information, the sender information, the receiver information and a preset mail template;
and sending the case reminding mail to a designated terminal.
Case processing reminding device based on deep learning model includes:
The first acquisition module is used for acquiring the complaint case to be processed, the sender information and the receiver information;
the export module is used for exporting the target attribute information of the complaint case to be processed from the appointed database;
the input module is used for inputting the target attribute information into a pre-trained deep learning model to obtain a target case processing type output by the deep learning model, wherein the deep learning model is obtained by training historical attribute information and a historical case processing type as samples;
the first judging module is used for judging whether the target case processing type is an emergency type or not;
the generation module is used for generating a case reminding mail according to the target attribute information, the sender information, the receiver information and a preset mail template when the target case processing type is the emergency type;
and the sending module is used for sending the case reminding mail to the appointed terminal.
A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the case processing reminding method based on the deep learning model when executing the computer program.
A computer readable storage medium storing a computer program which when executed by a processor implements the steps of the case processing reminding method based on the deep learning model.
According to the case processing reminding method, the device, the computer equipment and the storage medium based on the deep learning model, the complaint case to be processed, the sender information and the receiver information are firstly obtained, then the attribute information of the complaint case to be processed is derived from the appointed database, the attribute information is input into the deep learning model trained in advance to obtain the target case processing type output by the deep learning model, then whether the target case processing type is an emergency type is judged, and when the target case processing type is the emergency type, case reminding mails are generated according to the attribute information, the sender information, the receiver information and the preset mail template, and finally the case reminding mails are sent to the appointed terminal. Because the deep learning model is obtained by training historical attribute information and historical case processing types as samples, the target case processing types can be accurately obtained, when the target case processing types are the emergency types, case reminding mails are generated according to the acquired attribute information, sender information, recipient information and preset mail templates of the complaints, and the case reminding mails are sent to the appointed terminal, so that customer service personnel of the appointed terminal can process the complaints of the emergency types of the target case processing types for the first time after receiving the reminding mails, and the efficiency of processing the emergency complaints is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments of the present invention will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of an application environment of a case processing reminding method based on a deep learning model according to an embodiment of the invention;
FIG. 2 is a flow chart of a case processing reminding method based on a deep learning model according to an embodiment of the invention;
FIG. 3 is a flowchart of training a deep learning model in a case processing reminding method based on the deep learning model according to an embodiment of the invention;
FIG. 4 is a flow chart of determining positive or negative samples in a case processing reminding method based on a deep learning model according to an embodiment of the invention;
FIG. 5 is a flowchart of step S50 in a case processing reminding method based on a deep learning model according to an embodiment of the invention;
FIG. 6 is a flowchart of step S20 in a case processing reminding method based on a deep learning model according to an embodiment of the invention;
FIG. 7 is a schematic diagram of a case handling reminder based on a deep learning model according to an embodiment of the present invention;
FIG. 8 is a schematic diagram of a computer device in accordance with an embodiment of the invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The case processing reminding method based on the deep learning model can be applied to an application environment as shown in fig. 1, wherein computer equipment communicates with a server through a network. The computer device may be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices, among others. The server may be implemented as a stand-alone server or as a server cluster composed of a plurality of servers.
In an embodiment, as shown in fig. 2, a case processing reminding method based on a deep learning model is provided, and the case processing reminding method based on the deep learning model is applied in the financial industry, and is described by taking a server in fig. 1 as an example, and includes the following steps:
S10, acquiring complaint cases to be processed, sender information and recipient information;
in this embodiment, the complaint case to be processed is a complaint case to be processed by an insurance company, such as a complaint case of claim settlement, and the complaint case is stored in a preset specified database and is in a state that can be invoked at any time. The sender information is information of a sender in the mail, the receiver information is information of a receiver in the mail, and the sender information and the receiver information are stored in a preset mail database and are in a state capable of being called at any time.
Specifically, a first storage path of a complaint case to be processed in a preset appointed database is obtained, then the complaint case is extracted from the preset appointed database according to the first storage path, meanwhile, a second storage path of sender information in a preset mail database is obtained, then the sender information is extracted from the preset mail database according to the second storage path, a third storage path of recipient information in the preset mail database is obtained, and then the recipient information is extracted from the preset mail database according to the third storage path.
It should be noted that, the preset specified database and the preset mail database may be an SQL database or an available database, and specific contents of the preset specified database and the preset mail database may be set according to practical applications, which is not limited herein.
S20, deriving target attribute information of the complaint case to be processed from a specified database;
in the present embodiment, the target attribute information is information representing an attribute of a complaint case, such as a complaint case type. The target attribute information includes a reception start time, an underwriting organization, a reception worker number, a case state, and the like.
Specifically, the target attribute information of the complaint case to be processed is stored in a target table, such as an excel table, and it can be understood that the target attribute information of the complaint case to be processed is exported from a preset specified database by adopting a preset export table method, such as exporting the excel table by adopting a POI export method, wherein the POI is named as Poor Obfuscation Implementation in english, the chinese name is a public operation class, and the function of exporting the excel table is provided.
It should be noted that, the specified database in the step S20 is the specified database in the step S10, and in order to avoid repetition, specific contents of the preset table deriving method are not described here, and may be set according to practical applications, which is not limited herein.
S30, inputting target attribute information into a pre-trained deep learning model to obtain a target case processing type output by the deep learning model;
In this embodiment, the target case processing types include an emergency type, a normal type, and the like, where specific contents of the target case processing types may be set according to actual applications, and are not limited herein.
Specifically, a pre-trained deep learning model, such as a pre-trained convolutional neural network model, is firstly obtained, then the target attribute information derived in the step S20 is input into the deep learning model to obtain a target case processing type output by the deep learning model, such as "the acceptance start time is 2018.7.20, the underwriting organization is a safe life insurance company, the acceptance staff work number is pa00121 and the case state is a high-risk state" is input into the deep learning model to obtain the target case processing type is an emergency type.
It should be noted that the deep learning model is obtained by training historical attribute information and historical case processing types as samples.
S40, judging whether the processing type of the target case is an emergency type or not;
specifically, it is determined whether the target case processing type output in step S30 is an urgent type, when the target case processing type output in step S30 is an urgent type, step S50 is executed, and when the target case processing type output in step S30 is not an urgent type, step S70 is executed.
S50, generating a case reminding mail according to the target attribute information, the sender information, the receiver information and a preset mail template;
in this embodiment, the preset mail template is a preset mail template, and the mail can be generated only by importing sender information and recipient information into the preset mail template, and the preset mail template is specially stored in a preset template database and is in a state that can be called at any time, wherein the preset template database can be an SQL database or an available database, and specific contents of the preset template database and the preset mail template can be set according to practical applications, and the preset mail template is not limited herein.
Specifically, when the target case processing type is an emergency type, a fourth storage path of a preset mail template in a preset template database is acquired first, then the preset mail template is extracted from the preset template database according to the fourth storage path, and then a case reminding mail is generated according to the derived target attribute information, the acquired sender information, the receiver information and the preset mail template, for example, according to the conditions that the acceptance start time is 2018.7.20, the underwriting mechanism is a safe life insurance company, the acceptance staff work number is pa0121, the case state is a high-risk state, the sender information is zhangsan@163.com, the receiver information lisi@163.com and the preset mail template, the case reminding mail is generated.
S60, sending the case reminding mail to a designated terminal;
in this embodiment, the specified terminal refers to a terminal, such as a smart phone, that a user can browse a case reminding email, where the specified terminal may also be a tablet computer, etc., and specific content of the specified terminal may be set according to practical application, which is not limited herein.
Specifically, the case reminding mail generated in the step S50 is sent to the designated terminal by adopting a preset sending network, for example, the case reminding mail generated in the step S50 is sent to the smart phone by adopting a 4G network, so that after customer service personnel browse the case reminding mail, emergency complaint cases are processed in the first time.
It should be noted that, the preset sending network may be 2G or wifi, and the specific content of the preset sending network may be set according to the actual application, which is not limited herein.
S70, storing the complaint cases into a standby database according to the sequence of receiving the complaint cases.
Specifically, when the target case processing type is not an urgent type, the complaint cases are stored in the standby database according to the sequence of receiving the complaint cases, so that after the urgent type of complaint cases are processed, the complaint cases which are not urgent type are processed according to the sequence.
It should be noted that, the standby database may be an SQL database or an available database, and the specific content of the standby database may be set according to the actual application, which is not limited herein.
In the embodiment corresponding to fig. 2, firstly, acquiring complaint case to be processed, sender information and receiver information, then, deriving attribute information of the complaint case to be processed from a designated database, inputting the attribute information into a pre-trained deep learning model to obtain a target case processing type output by the deep learning model, then, judging whether the target case processing type is an emergency type, and when the target case processing type is the emergency type, generating a case reminding mail according to the attribute information, the sender information, the receiver information and a preset mail template, and finally, transmitting the case reminding mail to a designated terminal. Because the deep learning model is obtained by training historical attribute information and historical case processing types as samples, the target case processing types can be accurately obtained, when the target case processing types are the emergency types, case reminding mails are generated according to the acquired attribute information, sender information, recipient information and preset mail templates of the complaints, and the case reminding mails are sent to the appointed terminal, so that customer service personnel of the appointed terminal can process the complaints of the emergency types of the target case processing types for the first time after receiving the reminding mails, and the efficiency of processing the emergency complaints is improved.
In an embodiment, the case processing reminding method based on the deep learning model is applied in the financial industry, and as shown in fig. 3, in the case processing reminding method based on the deep learning model in the corresponding embodiment of fig. 2, a flowchart of training the deep learning model in an application scenario specifically includes the following steps:
s301, acquiring historical attribute information and a historical case processing type of a historical complaint case as samples;
in this embodiment, the history complaint case is a complaint case in which a case processing type has been identified, such as an emergency type complaint case. The history case processing type is a case processing type of a history complaint case, and the history complaint case and the history case processing type are specially stored in a preset history database and are stored in a state which can be called at any time.
Specifically, a fifth storage path of history attribute information of a history complaint case in a preset history database is obtained, then the history attribute information of the complaint case is extracted from the preset history database according to the fifth storage path to be used as a sample, for example, a "reception start time is 2018.7.20, a underwriting mechanism is a safe life insurance company, a reception staff work number is pa00121 and a case state is a high-risk state" is used as a sample, and meanwhile, a sixth storage path of a history case processing type in the preset history database is obtained, and then the history case processing type is extracted from the preset history database according to the sixth storage path to be used as a sample, for example, an emergency type or a common type is used as a sample.
It should be noted that, the preset history database may be an SQL database or an available database, and the specific content of the preset history database may be set according to the actual application, which is not limited herein.
S302, inputting the history attribute information in the sample into a deep learning model to obtain an output result;
specifically, the history attribute information in the sample acquired in step S301 is input to the deep learning model, and an output result is obtained.
S303, adjusting hidden layer parameters of the deep learning model to minimize errors between the output result and the historical attribute information in the sample;
in this embodiment, hidden layer parameters include the number of neural nodes, the step constant per improvement, target accuracy, etc.
Specifically, the output result is taken as an output target, and hidden layer parameters of the deep learning model are adjusted, so that the error between the output result and the historical case processing type in the sample is minimized.
It should be noted that, in adjusting the hidden layer parameters, by adjusting the step constant first, then adjusting the number of hidden layer nodes after adjusting appropriately, gradually increasing, the accuracy should theoretically be first increasing and then decreasing, and after finding the appropriate number of nodes, finally, gradually increasing the target accuracy, that is, gradually increasing the accuracy of the output target, so that the error between the output result and the processing type of the historical case in the sample is minimized.
Further, it is determined whether the error between the output result and the history case processing type in the sample satisfies the preset condition, if yes, step S304 is executed, and if not, steps S301 to S303 are executed again until the error satisfies the preset condition.
It should be noted that, the preset condition may be "four words including the urgent type, the total number of words of the output result cannot exceed 20", and the specific content of the preset condition may be set according to the actual application, which is not limited herein.
For a better understanding of step S302 and step S303, the following description is given by way of an example, and the details are as follows:
assuming that the sample is a history complaint case, the history attribute information of the history complaint case is "reception start time is 2018.7.20, the underwriting mechanism is a safe life insurance company, the reception staff is pa00121, and the case state is a high-risk state", the deep learning model is a convolutional neural network model, the history case processing type of the history complaint case is an "emergency type", the "reception start time is 2018.7.20, the underwriting mechanism is a safe life insurance company, the reception staff is pa00121, and the case state is a high-risk state" is input to the convolutional neural network model to obtain an output result, the output result is used as an output adjustment target, and hidden layer parameters of the deep learning model, such as a stepping constant, are continuously adjusted so as to minimize errors between the output result and the "emergency type" in the sample, if the output result is the "emergency type", the errors between the output result and the "emergency type" are considered small, and if the output result is the "case ordinary", the errors between the output result and the "emergency type" are considered large.
S304, determining the current deep learning model as a trained deep learning model.
Specifically, when the error between the output result and the history case processing type in the sample meets a preset condition, determining that the current deep learning model is a trained deep learning model, and when the error between the output result and the history case processing type in the sample does not meet the preset condition, determining that the current deep learning model is an untrained deep learning model.
In the embodiment corresponding to fig. 3, the historical attribute information of the historical complaint case and the historical case processing type are obtained as samples, then the historical attribute information in the samples is input into the deep learning model to obtain an output result, then hidden layer parameters of the deep learning model are adjusted to minimize errors between the output result and the historical attribute information in the samples, and finally if the errors meet preset conditions, the current deep learning model is determined to be a trained deep learning model. Because a large amount of historical attribute information and historical case processing types are adopted as samples, the historical attribute information serving as the samples is accurate complaint cases with the case processing types already identified, the historical case processing types serving as the samples are accurate case processing types of the historical complaint cases, a historical cleaning field in the samples is input into the deep learning model to obtain an output result, the output result is used as an output target for adjusting the deep learning model, and the hidden layer parameters of the deep learning model are continuously adjusted by adopting the accurate historical attribute information and the historical case processing types, so that the error between the output result of the deep learning model and the historical case processing types in the samples is minimized, and the accuracy of the historical case processing types output by the deep learning model is ensured.
In an embodiment, the case processing reminding method based on the deep learning model is applied in the financial industry, as shown in fig. 4, in the case processing reminding method based on the deep learning model in the corresponding embodiment of fig. 2, a flowchart of determining a positive sample or a negative sample in an application scene specifically includes the following steps:
s801, judging whether the target case processing type is a preset case type;
in the embodiment of the invention, the preset case types are all preset complaint cases, such as emergency types, common types and the like, wherein the specific content of the preset case types can be set according to actual requirements, and the preset case types are not limited herein.
Specifically, it is determined whether the target case processing type is a preset case type, when the target case processing type is not the preset case type, step S802 is performed, and when the target case processing type is the preset case type, step S803 is performed.
S802, determining target attribute information and a target case processing type as negative samples;
specifically, when the target case processing type is not the preset case type, the target attribute information and the target case processing type are determined as negative samples.
S803, determining the target attribute information and the target case processing type as positive samples, wherein the positive samples are used for updating the deep learning model.
When the target case processing type meets the preset condition, the target attribute information and the target case processing type are determined to be positive samples, and then the deep learning model is updated by adopting the positive samples.
In the embodiment corresponding to fig. 4, by judging whether the target case processing type meets the preset condition, if the target case processing type meets the preset condition, determining the target case processing type as the determined target case processing type, and determining the target attribute information and the target case processing type as positive samples, wherein the positive samples are used for updating the deep learning model. Because the output target case processing type does not belong to the preset case type, whether the target case processing type is the preset case type needs to be judged, if the target case processing type meets the preset condition, the target attribute information and the target case processing type are determined to be positive samples, so that the accurate output result of the deep learning model analysis is ensured, and the accuracy of the deep learning model analysis capability is improved.
In an embodiment, the case processing reminding method based on the deep learning model is applied in the financial industry, as shown in fig. 5, in the case processing reminding method based on the deep learning model in the corresponding embodiment of fig. 2, step S50 is a flowchart in an application scenario, and specifically includes the following steps:
s501, acquiring a preset mail template;
specifically, the content of the acquiring preset mail template in step S501 is consistent with the content of the acquiring preset mail template in step S50, and will not be described here to avoid repetition.
S502, extracting a sender mailbox address in sender information;
specifically, the sender mailbox address is extracted from the sender information extracted in step S10, for example, the sender mailbox address is "zhangsan@163.com"
It should be noted that, the specific content of the mailbox address of the sender may be set according to the practical application, which is not limited herein.
S503, extracting a receiver mailbox address in the receiver information;
specifically, the recipient mailbox address is extracted from the recipient information extracted in step S10, e.g., the recipient mailbox address is "lisi@163.com"
It should be noted that, the specific content of the mailbox address of the recipient may be set according to the practical application, which is not limited herein.
S504, generating a mail text according to the target attribute information and preset characters;
specifically, the mail text is generated according to the target attribute information and the preset text derived in step S20, for example, according to "the reception start time is 2018.7.20, the underwriting organization is a safe life insurance company, the work number of the reception staff is pa00121, and the case status is a high-risk status" and "the case is an emergency case, please preferentially process" to generate the mail text.
The specific content of the preset text may be set according to the practical application, and is not limited herein.
S505, importing the sender mailbox address, the receiver mailbox address and the mail text into a preset mail template to obtain a case reminding mail.
Specifically, the sender mailbox address extracted in step S502, the recipient mailbox address extracted in step S503, and the mail text generated in step S504 are imported into a preset mail template, so as to obtain a case reminding mail.
In the embodiment corresponding to fig. 5, the preset mail template is obtained, the mail address of the sender in the sender information is extracted, the mail address of the receiver in the receiver information is extracted, the mail text is generated according to the target attribute information and the preset text, and the mail address of the sender, the mail address of the receiver and the mail text are imported into the preset mail template to obtain the case reminding mail.
In an embodiment, the case processing reminding method based on the deep learning model is applied in the financial industry, and as shown in fig. 6, in any one of fig. 2 to 5, a flowchart of step S20 in an application scenario in the case processing reminding method based on the deep learning model specifically includes the following steps:
s201, exporting the first case attribute information from a specified database by adopting a preset export form method to obtain second case attribute information;
specifically, a seventh storage path of the specified table in the specified database is obtained first, then the specified table is extracted from the specified database according to the seventh storage path, then the first case attribute information of the specified table is derived from the specified database by adopting a preset derived table method to obtain the second case attribute information of the target specified table, and the specified database and the preset derived table method in the step S201 are the specified database and the preset derived table method in the step S10, so that repetition is avoided and no description is made here.
It should be noted that, the designated table may be an attribute table, the target designated table may be a case table, such as an excel table, and specific contents of the designated table and the target designated table may be set according to practical applications, which is not limited herein.
S202, determining the second case attribute information as target attribute information of the complaint case to be processed.
Specifically, the second case attribute information derived in step S201 is determined as target attribute information of the complaint case to be processed, such as "reception start time is 2018.7.20, the underwriting institution is a safe life insurance company, the reception staff is pa00121, and the case state is a high risk state" is determined as target attribute information of the complaint case to be processed.
In the embodiment corresponding to fig. 6, the first case attribute information is derived from the specified database by adopting a preset derived table method to obtain the second case attribute information, where the first case attribute information is the attribute information to be derived in the specified table, the second case attribute information is the derived attribute information in the target specified table, and the second case attribute information is determined as the target attribute information of the complaint case to be processed. Because the preset export form method has the function of quickly exporting forms, the first case attribute information is exported by adopting the preset export form method, so that the second case attribute information can be quickly obtained, namely the target attribute information of the complaint case to be processed can be quickly obtained, and the acquisition efficiency of the target attribute information is improved.
It should be understood that the sequence number of each step in the foregoing embodiment does not mean that the execution sequence of each process should be determined by the function and the internal logic, and should not limit the implementation process of the embodiment of the present invention.
In an embodiment, a case processing reminding device based on a deep learning model is provided, where the case processing reminding device based on the deep learning model corresponds to the case processing reminding method based on the deep learning model in the above embodiment one by one. As shown in fig. 7, the case processing reminding device based on the deep learning model includes a first acquisition module 71, a derivation module 72, an input module 73, a first judgment module 74, a generation module 75 and a transmission module 76. The functional modules are described in detail as follows:
a first obtaining module 71, configured to obtain complaint case to be processed, sender information, and recipient information;
a deriving module 72 for deriving target attribute information of the complaint case to be processed from the specified database;
the input module 73 is configured to input target attribute information into a pre-trained deep learning model, and obtain a target case processing type output by the deep learning model, where the deep learning model is obtained by training historical attribute information and a historical case processing type as samples;
A first judging module 74, configured to judge whether the target case processing type is an emergency type;
a generating module 75, configured to generate a case alert mail according to the target attribute information, the sender information, the recipient information, and a preset mail template when the target case processing type is an urgent type;
and a sending module 76, configured to send the case reminding mail to the specified terminal.
Further, the case processing reminder based on the deep learning model further comprises:
the second acquisition module is used for acquiring historical attribute information of the historical complaint cases and the historical case processing types as samples;
the second input module is used for inputting the historical attribute information in the sample into the deep learning model to obtain an output result;
the adjusting module is used for adjusting hidden layer parameters of the deep learning model so as to minimize errors between the output result and the historical attribute information in the sample;
the first determining module is used for determining that the current deep learning model is a trained deep learning model if the error meets a preset condition.
Further, the case processing reminder based on the deep learning model further comprises:
the second judging module is used for judging that the target case processing type is a preset case type;
And the second determining module is used for determining the target attribute information and the target case processing type as positive samples if the target case processing type is a preset case type, wherein the positive samples are used for updating the deep learning model.
Further, the generating module 75 includes:
the obtaining sub-module is used for obtaining a preset mail template;
the first extraction submodule is used for extracting the mail address of the sender in the sender information;
the second extraction sub-module is used for extracting the addressee mailbox address in the addressee information;
the generation sub-module is used for generating a mail text according to the target attribute information and preset characters;
and the importing sub-module is used for importing the sender mailbox address, the recipient mailbox address and the mail text into a preset mail template to obtain a case reminding mail.
Further, the deriving module 72 includes:
the guiding sub-module is used for guiding the first case attribute information from the appointed database by adopting a preset guiding table method to obtain second case attribute information, wherein the first case attribute information is the attribute information to be guided out in the appointed table, and the second case attribute information is the guided out attribute information in the target appointed table;
And the breaking sub-module is used for determining the second case attribute information as target attribute information of the complaint case to be processed.
For specific limitation of the case processing reminding device based on the deep learning model, reference may be made to the limitation of the case processing reminding method based on the deep learning model, and the description thereof is omitted here. All or part of each module in the case processing reminding device based on the deep learning model can be realized by software, hardware and a combination 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. 8. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device 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 the computer equipment is used for storing data related to a case processing reminding method based on the deep learning model. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program, when executed by a processor, implements a case processing reminding method based on a deep learning model.
In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the computer program to implement the steps of the case processing reminding method based on the deep learning model of the above embodiment, such as steps S10 to S60 shown in fig. 2. Alternatively, the processor, when executing the computer program, implements the functions of each module/unit of the case processing reminding device based on the deep learning model in the above embodiment, for example, the functions of the first acquisition module 71 to the sending module 76 shown in fig. 6. In order to avoid repetition, a description thereof is omitted.
In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, where the computer program when executed by a processor implements a case processing reminding method based on a deep learning model in the above method embodiment, or where the computer program when executed by a processor implements functions of each module/unit in a case processing reminding device based on a deep learning model in the above device embodiment. In order to avoid repetition, a description thereof is omitted. 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, storage, database, or other medium used in the various embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions.
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention, and are intended to be included in the scope of the present invention.
Claims (5)
1. The case processing reminding method based on the deep learning model is characterized by comprising the following steps of:
Acquiring a complaint case to be processed, sender information and recipient information, wherein the complaint case and the information are stored in a preset appointed database;
the target attribute information of the complaint case to be processed is derived from a specified database, wherein the target attribute information comprises the type of the complaint case, the acceptance starting time and the case state;
inputting the target attribute information into a pre-trained deep learning model to obtain a target case processing type output by the deep learning model, wherein the deep learning model is obtained by training historical attribute information and historical case processing type as samples;
judging whether the target case processing type is an emergency type or not;
when the target case processing type is the emergency type, generating a case reminding mail according to the target attribute information, the sender information, the receiver information and a preset mail template;
sending the case reminding mail to a designated terminal;
the deep learning model is obtained by training historical attribute information and historical case processing types as samples, and the training of the deep learning model specifically comprises the following steps:
Inputting the historical attribute information in the sample into the deep learning model to obtain an output result;
adjusting hidden layer parameters of the deep learning model to minimize errors between the output result and historical attribute information in the sample;
if the error meets a preset condition, determining that the current deep learning model is a trained deep learning model;
after the target attribute information is input to a pre-trained deep learning model to obtain a target case processing type output by the deep learning model, the case processing reminding method of the deep learning model further comprises the following steps:
judging whether the target case processing type is a preset case type or not;
if the target case processing type is the preset case type, determining the target attribute information and the target case processing type as positive samples, wherein the positive samples are used for updating the deep learning model;
the attribute information to be exported is stored in a designated table, and the target attribute information of the complaint case to be processed is exported from a designated database, which comprises the following steps:
deriving first case attribute information from the appointed database by adopting a preset derivation form method to obtain second case attribute information, wherein the first case attribute information is attribute information to be derived in the appointed form, and the second case attribute information is derived attribute information in a target appointed form;
And determining the second case attribute information as target attribute information of the complaint case to be processed.
2. The case processing reminding method of the deep learning model according to claim 1, wherein the generating a case reminding mail according to the target attribute information, the sender information, the recipient information and a preset mail template comprises:
acquiring the preset mail template;
extracting a sender mailbox address in the sender information;
extracting a receiver mailbox address in the receiver information;
generating a mail text according to the target attribute information and preset characters;
and importing the sender mailbox address, the receiver mailbox address and the mail text into the preset mail template to obtain a case reminding mail.
3. Case processing reminding device based on deep learning model, its characterized in that, case processing reminding device based on deep learning model includes:
the first acquisition module is used for acquiring the complaint case to be processed, the sender information and the receiver information, wherein the complaint case and the information are stored in a preset appointed database;
the export module is used for exporting target attribute information of the complaint case to be processed from the appointed database, wherein the target attribute information comprises a complaint case type, a reception start time and a case state;
The input module is used for inputting the target attribute information into a pre-trained deep learning model to obtain a target case processing type output by the deep learning model, wherein the deep learning model is obtained by training historical attribute information and a historical case processing type as samples;
the first judging module is used for judging whether the target case processing type is an emergency type or not;
the generation module is used for generating a case reminding mail according to the target attribute information, the sender information, the receiver information and a preset mail template when the target case processing type is the emergency type;
the sending module is used for sending the case reminding mail to the appointed terminal;
the case processing reminding device based on the deep learning model further comprises:
the second acquisition module is used for acquiring historical attribute information of the historical complaint cases and the historical case processing types as samples;
the second input module is used for inputting the history attribute information in the sample into the deep learning model to obtain an output result;
the adjusting module is used for adjusting hidden layer parameters of the deep learning model so as to minimize errors between the output result and the historical attribute information in the sample;
The first determining module is used for determining that the current deep learning model is a trained deep learning model if the error meets a preset condition;
the case processing reminding device based on the deep learning model further comprises:
the second judging module is used for judging whether the target case processing type is a preset case type or not;
the second determining module is configured to determine the target attribute information and the target case processing type as positive samples if the target case processing type is the preset case type, where the positive samples are used for updating the deep learning model;
the export module comprises:
the guiding sub-module is used for guiding the first case attribute information from the appointed database by adopting a preset guiding table method to obtain second case attribute information, wherein the first case attribute information is the attribute information to be guided out in the appointed table, and the second case attribute information is the guided out attribute information in the target appointed table;
and the breaking sub-module is used for determining the second case attribute information as target attribute information of the complaint case to be processed.
4. Computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the case processing reminder method based on the deep learning model according to any one of claims 1 to 2 when the computer program is executed.
5. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the steps of the deep learning model-based case processing reminding method according to any one of claims 1 to 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811252691.9A CN109544100B (en) | 2018-10-25 | 2018-10-25 | Case processing reminding method, device, equipment and medium based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811252691.9A CN109544100B (en) | 2018-10-25 | 2018-10-25 | Case processing reminding method, device, equipment and medium based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109544100A CN109544100A (en) | 2019-03-29 |
| CN109544100B true CN109544100B (en) | 2024-03-08 |
Family
ID=65845531
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811252691.9A Active CN109544100B (en) | 2018-10-25 | 2018-10-25 | Case processing reminding method, device, equipment and medium based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109544100B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112598386A (en) * | 2020-12-24 | 2021-04-02 | 中国农业银行股份有限公司 | Method, device and equipment for task reminding and storage medium |
| CN114202341A (en) * | 2021-11-16 | 2022-03-18 | 北京云迹科技有限公司 | Complaint processing method and device, electronic device and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004054664A (en) * | 2002-07-22 | 2004-02-19 | Ricoh Co Ltd | Method, system and program for managing electronic mail, and recording medium |
| CN103309948A (en) * | 2013-05-20 | 2013-09-18 | 携程计算机技术（上海）有限公司 | System and method for public opinion monitoring analysis and intelligent distribution processing of coordination center |
| CN107798031A (en) * | 2017-02-17 | 2018-03-13 | 平安科技（深圳）有限公司 | Policy information querying method and device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160262128A1 (en) * | 2013-09-27 | 2016-09-08 | Hewlett Packard Enterprise Development Lp | Notifying a user of critical emails via text messages |
-
2018
- 2018-10-25 CN CN201811252691.9A patent/CN109544100B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004054664A (en) * | 2002-07-22 | 2004-02-19 | Ricoh Co Ltd | Method, system and program for managing electronic mail, and recording medium |
| CN103309948A (en) * | 2013-05-20 | 2013-09-18 | 携程计算机技术（上海）有限公司 | System and method for public opinion monitoring analysis and intelligent distribution processing of coordination center |
| CN107798031A (en) * | 2017-02-17 | 2018-03-13 | 平安科技（深圳）有限公司 | Policy information querying method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109544100A (en) | 2019-03-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108933993B (en) | Short message cache queue selection method and device, computer equipment and storage medium | |
| US10887268B2 (en) | Systems and methods for prioritizing messages for conversion from text to speech based on predictive user behavior | |
| US10193850B2 (en) | Discovering questions, directives, and requests from, and prioritizing replies to, a preferred message sender method and apparatus | |
| WO2020233091A1 (en) | Method and apparatus for service data rollback, computer device and storage medium | |
| CN103854168A (en) | Heterogeneous process to-do task centralized processing method and device | |
| WO2021068348A1 (en) | Task deployment method and system, and storage medium | |
| US11153249B2 (en) | Methods and apparatuses for managing limited engagement by external email resource entity within a group-based communication system | |
| US20200167429A1 (en) | Efficient use of word embeddings for text classification | |
| CN109544100B (en) | Case processing reminding method, device, equipment and medium based on deep learning | |
| WO2020258650A1 (en) | Page information display method and apparatus, computer device, and storage medium | |
| CN108234549B (en) | Policy information checking method and device | |
| CN112437000A (en) | Message queue pushing method and device, computer equipment and storage medium | |
| EP3360095A1 (en) | Identifying search results from local and remote search of communications in parallel | |
| CN110458710B (en) | Request processing method, apparatus, computer device and storage medium | |
| US20170243166A1 (en) | System and method for employee placement prediction using machine learning | |
| CN110069755A (en) | Method, apparatus, computer equipment and the storage medium that data are sent | |
| US20150052451A1 (en) | Real time text message system | |
| US20190303948A1 (en) | Providing customer relationship management data via a messaging interface | |
| CN113761858A (en) | Form data processing method and device, computer equipment and storage medium | |
| WO2020215542A1 (en) | Information notification method and device, computer apparatus, and storage medium | |
| WO2021139480A1 (en) | Gis service aggregation method and apparatus, and computer device and storage medium | |
| CN113542506A (en) | Outbound method, device, equipment and medium | |
| WO2022231593A1 (en) | Automated extraction and standardization of financial time-series data from semi-structured tabular input | |
| US11874823B1 (en) | Agnostic image digitizer to detect fraud | |
| US10176248B2 (en) | Performing a dynamic search of electronically stored records based on a search term format |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |