CN116341879B - Overdue asset collection intelligent case division method and system - Google Patents
Overdue asset collection intelligent case division method and system Download PDFInfo
- Publication number
- CN116341879B CN116341879B CN202310602327.5A CN202310602327A CN116341879B CN 116341879 B CN116341879 B CN 116341879B CN 202310602327 A CN202310602327 A CN 202310602327A CN 116341879 B CN116341879 B CN 116341879B
- Authority
- CN
- China
- Prior art keywords
- communication
- cases
- case
- successful
- communicated
- 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 33
- 238000004891 communication Methods 0.000 claims abstract description 257
- 230000002159 abnormal effect Effects 0.000 claims abstract description 65
- 238000012545 processing Methods 0.000 claims abstract description 5
- 238000004422 calculation algorithm Methods 0.000 claims description 19
- 230000006870 function Effects 0.000 claims description 16
- 230000008909 emotion recognition Effects 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 10
- 239000013598 vector Substances 0.000 claims description 10
- 238000003860 storage Methods 0.000 claims description 8
- 238000013145 classification model Methods 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 5
- 238000005457 optimization Methods 0.000 claims description 4
- 238000012360 testing method Methods 0.000 claims description 4
- 241000257303 Hymenoptera Species 0.000 claims description 2
- 230000003044 adaptive effect Effects 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 claims description 2
- 238000013527 convolutional neural network Methods 0.000 claims description 2
- 230000008451 emotion Effects 0.000 claims description 2
- 230000002996 emotional effect Effects 0.000 claims description 2
- 230000000977 initiatory effect Effects 0.000 claims description 2
- 238000010606 normalization Methods 0.000 claims description 2
- 239000003016 pheromone Substances 0.000 claims description 2
- 238000000926 separation method Methods 0.000 claims description 2
- 238000003306 harvesting Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000011156 evaluation Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
- G06Q10/063114—Status monitoring or status determination for a person or group
-
- 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- 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
- G06Q40/02—Banking, e.g. interest calculation or account maintenance
-
- 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
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- Educational Administration (AREA)
- General Business, Economics & Management (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Technology Law (AREA)
- Telephonic Communication Services (AREA)
Abstract
The invention provides an overdue asset collection intelligent case division method and system, which belong to the technical field of data processing and specifically comprise the following steps: dividing the existing collection cases into the successful cases and the successful cases which are communicated through the communication conditions of the existing collection cases of the collection staff, dividing the successful cases which are not communicated into abnormal cases and normal cases based on the abnormal telephone communication values of the successful cases which are not communicated, and combining the communication times of other communication modes except telephone communication; dividing the communicated successful cases into matched cases and non-matched cases based on the matched times and the non-matched times of the offending subjects of the communicated successful cases and combining the proportion of the matched times in the latest preset time; the scoring value of the collecting personnel is determined based on the number and the plumpness scores of the different types of cases of the collecting personnel, and the collecting personnel of the cases to be distributed are determined based on the scoring value, so that the accuracy of case division is further improved, and the collecting efficiency is improved.
Description
Technical Field
The invention belongs to the technical field of data processing, and particularly relates to an overdue asset collection intelligent case dividing method and system.
Background
In order to realize intelligent division of consumption finance and other types of overdue assets, in the prior art, for example, CN114626735A "collect case distribution method, device, equipment and computer readable storage medium" performs differential rating on collect cases through the collect amount, collect times, overdue time, corresponding collect person and client credit scores corresponding to the collect cases, and distributes cases according to the rating result and the work efficiency of the collect person, but the following technical problems exist:
1. The indexes such as the corresponding harvest rate, the harvest time, the expiration time, the corresponding harvest person and the credit score of the client and the harvest difficulty of the to-be-harvested cases are not limited by the natural law, classification of the cases cannot be realized, even if the expiration time of the cases is long, the harvest difficulty is not necessarily indicated to be large, and the cases which are communicated with the offending subjects of the to-be-harvested cases cannot be established, the harvest difficulty is obviously greater than those which are already communicated with the offending subjects, after communication is established, the harvest difficulty of the cases which are strong and refused to be not matched are obviously greater than those of the cases with high matching degree, so that the cases cannot be accurately allocated if the case classification mode is adopted.
2. The differential classification of the cashier is carried out based on the working efficiency of the cashier, namely the number of cashier successful cases is not limited by the natural law, even though the cashier is high at present, the cashier is not always high in the day, so the differential classification of the cashier cannot be realized, and meanwhile, the cashier with more working plumpness of the cashier, such as more telephone time and more communication times, is not considered, and the working plumpness of the cashier is obviously higher, so if the working efficiency is adopted for carrying out the differential classification of the cashier, the accurate distribution of the cases cannot be realized.
3. The determination of the distributable collecting staff is not considered by combining the work plumpness of the collecting staff and the number of the different types of cases, if the number of the different types of cases is only considered, the identification of the distributable collecting staff cannot be accurately realized because the work plumpness of the different collecting staff is different.
Based on the technical problems, an overdue asset collection intelligent case division method and system are needed to be designed.
Disclosure of Invention
The invention aims to provide an overdue asset collection intelligent case division method and system.
In order to solve the technical problems, the first aspect of the present invention provides an overdue asset collection intelligent case separation method, which specifically includes:
S11: dividing the existing collection cases into a communicated successful case and an un-communicated successful case by the communication condition of the existing collection cases of the collection staff, determining the abnormal telephone communication value of the un-communicated successful case based on the communication times, the hang-up times and the communication times of different feedback types of telephone communication of the un-communicated successful case, and dividing the un-communicated successful case into an abnormal case and a normal case by combining the communication times of other communication modes except the telephone communication;
S12: determining emotion recognition results and coordination degree recognition results of the offensive body based on recognition results of communication keywords of the offensive body of the communicated successful cases, dividing the communication success times into coordination times and non-coordination times by combining the emotion recognition results and the coordination degree recognition results, and dividing the communicated successful cases into coordination cases and non-coordination cases by combining the proportion of the coordination times in a preset time;
S13: determining the plumpness score of the adductor based on the communication duration, the communication times and the average feedback time of the telephone communication of the adductor in the set time and the communication duration, the communication times and the average feedback time of other communication modes;
S14: and determining the grading value of the collecting personnel based on the number of normal cases, the number of abnormal cases, the number of matched cases, the number of non-matched cases and the plumpness grading of the collecting personnel, and determining the collecting personnel to be distributed based on the grading value of the distributing personnel when the collecting personnel is determined to be the distributing personnel based on the grading value of the distributing personnel.
In another aspect, an embodiment of the present application provides a computer readable storage medium, on which a computer program is stored, where the computer program, when executed in a computer, causes the computer to execute the above-mentioned method for collecting overdue assets.
On the other hand, the embodiment of the application provides a terminal device, which comprises a memory, a processor and a program stored on the memory and capable of running on the processor, wherein the processor realizes the overdue asset collection intelligent case division method when executing the program.
The invention has the beneficial effects that:
The method has the advantages that the existing collection accelerating cases are divided into the communicated successful cases and the non-communicated successful cases, so that the difference determination of the existing collection accelerating cases of the collection accelerating staff is realized, meanwhile, the non-communicated successful cases are divided into abnormal cases and normal cases according to the telephone communication abnormal values of the non-communicated successful cases and the communication times of other communication modes except telephone communication, so that the recognition of the cases with difficult communication in the non-communicated cases is realized, the collection accelerating difficulty of the cases with difficult communication is fully considered, and the reliability and the accuracy of the final case distribution are guaranteed.
The method comprises the steps of dividing the successful communication times into the matched times and the unmatched times based on the emotion recognition result and the matching degree recognition result of the successful communication cases, and dividing the successful communication cases into the matched cases and the unmatched cases by combining the proportion of the matched times in the preset time, so that the matching conditions of different successful communication cases are distinguished, the cases with larger matching difficulty are recognized, and the accuracy of evaluation is ensured.
The establishment of the plumpness score is carried out by the communication duration, the communication times and the average feedback time of telephone communication of the receipts-accelerating personnel in the set time, so that the determination of the activity condition of the receipts-accelerating personnel from multiple angles is realized, and a foundation is laid for further carrying out differential division of the receipts-accelerating personnel.
The screening of the collecting personnel is realized from the five aspects of the number of normal cases, the number of abnormal cases, the matched cases, the non-matched cases and the plumpness by constructing the grading values of the collecting personnel, the number of different cases is considered, and the plumpness of different collecting personnel is also considered, so that the cases to be distributed can be accurately and rapidly processed, and the collecting efficiency of the cases to be distributed is ensured.
Additional features and advantages will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and drawings.
In order to make the above objects, features and advantages of the present invention more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
The above and other features and advantages of the present invention will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings;
FIG. 1 is a flow chart of a method of overdue asset collection intelligent case division;
FIG. 2 is a flowchart showing the specific steps of dividing the successful cases of the communication into abnormal cases and normal cases;
FIG. 3 is a flow chart of dividing the communicated successful cases into matched cases and non-matched cases;
FIG. 4 is a flowchart of specific steps in the establishment of a fullness score;
FIG. 5 is a flowchart showing specific steps for assignable person determination;
FIG. 6 is a diagram of the communication mode of the successful cases without communication;
Fig. 7 is a block diagram of a computer-readable storage medium.
Detailed Description
Example embodiments will now be described more fully with reference to the accompanying drawings. However, the exemplary embodiments can be embodied in many forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the example embodiments to those skilled in the art. The same reference numerals in the drawings denote the same or similar structures, and thus detailed descriptions thereof will be omitted.
First aspect
In order to solve the above problems, according to an aspect of the present invention, as shown in fig. 1, there is provided an overdue asset collection intelligent case division method, which specifically includes:
S11: dividing the existing collection cases into a communicated successful case and an un-communicated successful case through the communication condition of the existing collection cases of collection staff, dividing the un-communicated successful case into an abnormal case and a normal case based on the communication mode of the un-communicated successful case and the communication times of different communication modes, and dividing the communicated successful case into the normal case and the abnormal case based on the recognition result of the communication keywords of the default main body of the communicated successful case;
It should be noted that, the successful cases that have been communicated are cases that have completed communication with the offending body and have been fed back, the successful cases that have not been communicated with the offending body and have not been fed back, and the abnormal cases in the successful cases that have been communicated are cases that have strong attitudes of the offending body, are not matched with or are unwilling to be paid, and can be determined by the recognition results of the communication keywords communicated with the abnormal cases, and the abnormal cases that have not been communicated are cases that have been tried to communicate for many times but have not been responded or hung up by a person, or have no contact way with the offending body, so that the difficulty of prompting is obviously greater compared with other cases like the abnormal cases.
Specifically, as shown in fig. 6, the communication manner of the unsuccessful case includes: mail communication, instant messaging software communication, short message communication, and telephone communication.
Specifically, as shown in fig. 2, the steps for dividing the successful case of the communication failure into the abnormal case and the normal case are as follows:
S21: determining whether the non-communication successful case is an abnormal case based on the communication times of the telephone communication of the non-communication successful case, if so, taking the non-communication successful case as the abnormal case, and if not, entering step S22;
It should be noted that, when the number of times of telephone communication of the un-communicated successful case is greater than the set number of times of communication, the un-communicated successful case is taken as an abnormal case;
S22: determining whether the non-communication successful case is a suspected abnormal case or not based on the communication times of the telephone communication of the non-communication successful case, if so, entering step S23, and if not, entering step S24;
It should be noted that, when the number of communication times of the telephone communication is greater than a certain magnitude, the abnormal case can be identified by combining with other communication times, but the magnitude is also smaller than the original set number of communication times;
S23: determining whether the non-communication successful case is an abnormal case based on the sum of the communication times of other communication modes except telephone communication of the suspected abnormal case, if so, taking the non-communication successful case as the abnormal case, and if not, entering step S24;
s24: determining a telephone communication abnormal value of the un-communicated successful case based on the communication times, the hang-up times and the communication times of different feedback types of the telephone communication of the un-communicated successful case, determining other communication abnormal values of the un-communicated case based on the communication times of other communication modes except the telephone communication of the un-communicated successful case and the communication times in a preset time, obtaining the communication abnormal value based on the telephone communication abnormal value and the other communication abnormal values, and dividing the un-communicated successful case into an abnormal case and a normal case based on the communication abnormal value.
It should be noted that, the abnormal cases are determined by using a classification model based on an SVM algorithm of ACO optimization.
It should be further noted that, the adaptive adjustment coefficient and the random factor are introduced into the initiation function of the ACO algorithm, so that the global searching capability of the algorithm is high, the distance between the current node and the optional node is considered in the construction process of the heuristic function, and the heuristic information of the distance between the current node and the target point is added. In the later stage of algorithm operation, in order to accelerate the algorithm to converge to the global optimal solution, the function of heuristic information is weakened, wherein the calculation formula of the improved heuristic function is as follows:
Wherein Lij is the distance between grid i and grid j, ljt is the Euclidean distance between grid j and grid T, ljt is the Euclidean distance between grid j and grid T, T is a constant greater than 0, T is a dynamic coefficient, rand (0, 1) is a random number with a value range of 0 to 1, U is the current iteration number, and Umax is the maximum iteration number.
By constructing the case characteristic value and the main body characteristic value, the evaluation of the existing collection-promoting case is realized from the two aspects of the case and the offending main body of the case, the accuracy is high, the data dimension required to be processed by the classification model is reduced, and the processing efficiency and accuracy are further improved.
By adopting the classification model based on the SVM algorithm of ACO optimization, the obtained case type result is more accurate, and meanwhile, the problem of lower classification efficiency caused by the situation of local optimum is further avoided by adopting the optimization algorithm.
The specific steps of the classification model construction based on the ACO-SVM algorithm are specifically illustrated as follows:
The ant colony algorithm is utilized to search and optimize punishment parameters and kernel function parameters of the SVM, and the flow of the algorithm is as follows:
Step 1: dividing the acquired data set into a training set and a testing set, respectively extracting the characteristics to form characteristic vectors of the training set and the testing set, and carrying out data normalization processing:
Step 2: the ACO parameters are initialized, a combination of penalty parameters and kernel function parameters is randomly generated within a certain range to be used as the whole solution space set I, and ants are randomly placed in the solution space set I.
Step 3: starting ACO, training and learning the SVM by using a training set, wherein in the training process, the probability of the kth ant selecting the jth parameter combination from the set I at the t moment is updated, along with the iteration process, the pheromone concentration at the j position in the solution space is updated, and the step 3 is repeated, so that the obtained parameter combination is placed in the set FA;
Step 4: and determining optimal parameters of the SVM algorithm based on the set FA, and completing the construction of the classification model.
In this embodiment, the existing collecting-accelerating case is divided into the successful case and the non-successful case, so that the difference determination of the existing collecting-accelerating case of the collecting-accelerating person is realized, meanwhile, the non-successful case is divided into the abnormal case and the normal case according to the telephone communication abnormal value of the non-successful case and the communication times of other communication modes except telephone communication, so that the recognition of the case with difficult communication in the non-successful case is realized, the collecting-accelerating difficulty of the case with difficult communication is fully considered, and the reliability and the accuracy of the final case distribution are ensured.
S12: determining emotion recognition results and coordination degree recognition results of the offensive body based on recognition results of communication keywords of the offensive body of the communicated successful cases, dividing the communication success times into coordination times and non-coordination times by combining the emotion recognition results and the coordination degree recognition results, and dividing the communicated successful cases into coordination cases and non-coordination cases by combining the proportion of the coordination times in a preset time;
as shown in fig. 3, the division of the successful communication cases into the matched cases and the non-matched cases specifically includes:
S31: based on the communication success times of the offender of the successful communication case, judging the communication keywords of the communication success times to determine whether the communication keywords contain non-matching keywords, if not, determining the communication success times as matching times;
S32: acquiring context information of a communication text of successful times of communication by utilizing BiGRU algorithm, learning emotional characteristics of a tag by utilizing a tag encoder, constructing attention weights of the word vectors by utilizing the tag precoder, generating text feature vectors based on the word vectors and the attention weights, and acquiring emotion recognition results of the successful times of communication by taking the obtained text feature vectors as input of a convolutional neural network;
S33: acquiring communication keywords of the communication success times of the offender of the successful communication case, carrying out a matching degree identification result of the communication success times based on the number of matching keywords and the number of non-matching keywords in the communication keywords, and dividing the communication success times into matching times and non-matching times based on the matching degree identification result and the emotion identification result;
S34: determining the matching degree of the communicated successful cases by combining the proportion of matching times in the communication times of the offending subjects in the latest preset time based on the number of matching times and the number of non-matching times of the offending subjects of the communicated successful cases, and dividing the communicated successful cases into matching cases and non-matching cases based on the matching degree.
In this embodiment, the number of successful communication cases is divided into the number of matching times and the number of non-matching times based on the emotion recognition result and the matching degree recognition result of the successful communication cases, and the successful communication cases are divided into the matching cases and the non-matching cases by combining the proportion of the matching times in the preset time, so that the matching conditions of different successful communication cases are distinguished, the cases with higher matching difficulty are recognized, and the accuracy of evaluation is ensured.
S13: determining the plumpness score of the adductor based on the communication duration, the communication times and the average feedback time of the telephone communication of the adductor in the set time and the communication duration, the communication times and the average feedback time of other communication modes;
It should be noted that the average feedback time reflects the feedback time when the receiving person obtains feedback of the default subject in instant messaging software, short messages, mails and the like, generally, the shorter the feedback time, the higher the fullness score for the adductor.
Specifically, as shown in fig. 4, the specific steps of the establishment of the fullness score are as follows:
s41: determining whether the collecting person is a suspected active collecting person or not based on the communication times of telephone communication and the communication times of other communication modes of the collecting person in a set time, if so, entering a step S42, and if not, entering a step S44;
when the number of telephone communication or the number of other communication modes within the set time is larger than the set number, the collecting person is used as a suspected active collecting person;
S42: obtaining a telephone communication value based on the communication duration, the communication times and the average communication times of each existing prompting receipt in a set time of the prompting receipt, determining whether the prompting receipt is an active prompting receipt or not based on the telephone communication value, if so, constructing a plumpness score based on the telephone communication value, and if not, entering step S43;
S43: determining other communication values of the collecting person based on the communication duration, the communication times and the average feedback time of other communication modes of the collecting person in the set time, determining whether the collecting person is an active collecting person or not based on the other communication values, if so, constructing a plumpness score based on the other communication values, and if not, entering step S44;
S44: and constructing total communication times based on the communication times of telephone communication of the receiving personnel in the set time and the communication times of other communication modes, and constructing plumpness scores based on the total communication times, the telephone communication values and other communication.
Specifically, the value of the fullness score ranges from 0 to 1, wherein the greater the value of the fullness score, the higher the fullness of the receiving person.
In this embodiment, the establishment of the plumpness score is performed by the communication duration, the communication times, the communication duration of other communication modes, the communication times and the average feedback time of telephone communication of the receivings in the set time, so that the determination of the activity condition of the receivings from multiple angles is realized, and a foundation is laid for further carrying out differential classification of the receivings.
S14: and determining the grading value of the collecting personnel based on the number of normal cases, the number of abnormal cases, the number of matched cases, the number of non-matched cases and the plumpness grading of the collecting personnel, and determining the collecting personnel to be distributed based on the grading value of the distributing personnel when the collecting personnel is determined to be the distributing personnel based on the grading value of the distributing personnel.
Specifically, as shown in fig. 5, the specific steps of the assignable personnel determination are as follows:
S51: determining whether the collecting person is an unallocated person or not based on the number of abnormal cases and the number of non-matched cases of the collecting person, if so, determining that the collecting person is not an allocable person, and if not, entering step S52;
s52: determining whether further determination of the plumpness score is needed or not based on the number of normal cases and the number of matched cases of the collecting staff, if yes, entering step S53, and if not, entering step S54;
S53: determining whether the collecting person is an assignable person based on the plumpness score of the collecting person, if so, determining that the collecting person is an assignable person, and if not, entering step S54;
S54: determining the score value of the collecting personnel based on the plumpness score of the collecting personnel, the number of normal cases, the number of abnormal cases, the number of matched cases and the number of non-matched cases, and determining whether the collecting personnel is an assignable personnel based on the score value.
In the embodiment, through the construction of the grading values of the collecting staff, the screening of the collecting staff is realized from five aspects of the number of normal cases, the number of abnormal cases, the matched cases, the non-matched cases and the plumpness, not only the number of different cases is considered, but also the plumpness of different collecting staff is considered, so that the cases to be distributed can be accurately and quickly processed, and the collecting efficiency of the cases to be distributed is ensured.
It should be noted that, the determining of the collecting personnel of the case to be distributed based on the scoring value of the allocable personnel specifically includes:
And determining the collecting staff of the case to be distributed based on the scoring value of the allocable staff.
On the other hand:
As shown in fig. 7, an embodiment of the present application provides a computer readable storage medium having a computer program stored thereon, which when executed in a computer, causes the computer to perform an overdue asset collection intelligent classification method as described above.
It should be noted that the method for intelligently sorting overdue asset collection specifically comprises the following steps:
Dividing the existing collection cases into a communicated successful case and an un-communicated successful case through the communication condition of the existing collection cases of collection staff, dividing the un-communicated successful case into an abnormal case and a normal case based on the communication mode of the un-communicated successful case and the communication times of different communication modes, and dividing the communicated successful case into the normal case and the abnormal case based on the recognition result of the communication keywords of the default main body of the communicated successful case;
And determining whether the collecting person is an unallocated person or not based on the number of abnormal cases of the collecting person, if so, determining that the collecting person is not an allocable person, and failing to allocate the case to be allocated.
On the other hand:
The embodiment of the application provides a terminal device which comprises a memory, a processor and a program stored on the memory and capable of running on the processor, wherein the processor realizes the overdue asset collection intelligent case division method when executing the program.
It should be noted that the method for intelligently sorting overdue asset collection specifically comprises the following steps:
Dividing the existing collection cases into a communicated successful case and an un-communicated successful case by the communication condition of the existing collection cases of the collection staff, determining the abnormal telephone communication value of the un-communicated successful case based on the communication times, the hang-up times and the communication times of different feedback types of telephone communication of the un-communicated successful case, and dividing the un-communicated successful case into an abnormal case and a normal case by combining the communication times of other communication modes except the telephone communication;
Determining emotion recognition results and coordination degree recognition results of the offensive body based on recognition results of communication keywords of the offensive body of the communicated successful cases, dividing the communication success times into coordination times and non-coordination times by combining the emotion recognition results and the coordination degree recognition results, and dividing the communicated successful cases into coordination cases and non-coordination cases by combining the proportion of the coordination times in a preset time;
And determining the grading value of the collecting personnel based on the number of normal cases, the number of abnormal cases, the number of matched cases and the number of non-matched cases of the collecting personnel, and determining the collecting personnel to be distributed based on the grading value of the distributing personnel when the collecting personnel is determined to be the distributing personnel based on the grading value of the distributing personnel.
In the several embodiments provided in the present application, it should be understood that the disclosed system and method may be implemented in other manners as well. The system 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 systems, 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 invention 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 on a computer readable storage medium. Based on this understanding, the technical solution of the present invention 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 of the embodiments of the present invention. And the aforementioned storage medium includes: a usb 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.
With the above-described preferred embodiments according to the present invention as an illustration, the above-described descriptions can be used by persons skilled in the relevant art to make various changes and modifications without departing from the scope of the technical idea of the present invention. The technical scope of the present invention is not limited to the description, but must be determined according to the scope of claims.
Claims (7)
1. An overdue asset collection intelligent case separation method specifically comprises the following steps:
Dividing the existing collection cases into a communicated successful case and an un-communicated successful case by the communication condition of the existing collection cases of the collection staff, determining the abnormal telephone communication value of the un-communicated successful case based on the communication times, the hang-up times and the communication times of different feedback types of telephone communication of the un-communicated successful case, and dividing the un-communicated successful case into an abnormal case and a normal case by combining the communication times of other communication modes except the telephone communication;
the specific steps of dividing the un-communicated successful case into an abnormal case and a normal case are as follows:
S21, determining whether the non-communication successful case is an abnormal case or not based on the communication times of telephone communication of the non-communication successful case, if so, taking the non-communication successful case as the abnormal case, and if not, entering step S22;
s22, determining whether the non-communication successful case is a suspected abnormal case or not based on the communication times of the telephone communication of the non-communication successful case, if so, entering a step S23, and if not, entering a step S24;
S23, determining whether the non-communication successful case is an abnormal case or not based on the sum of the communication times of other communication modes except telephone communication of the suspected abnormal case, if so, taking the non-communication successful case as the abnormal case, and if not, entering step S24;
S24, determining a telephone communication abnormal value of the un-communicated successful case based on the communication times, the hang-up times and the communication times of different feedback types of the telephone communication of the un-communicated successful case, determining other communication abnormal values of the un-communicated case based on the communication times of other communication modes except the telephone communication of the un-communicated successful case and the communication times in a preset time, obtaining the communication abnormal value based on the telephone communication abnormal value and the other communication abnormal values, and dividing the un-communicated successful case into an abnormal case and a normal case based on the communication abnormal value;
Determining emotion recognition results and coordination degree recognition results of the offensive body based on the recognition results of the communication keywords of the offensive body of the communicated successful cases, dividing the communication success times into coordination times and non-coordination times by combining the emotion recognition results and the coordination degree recognition results, and dividing the communicated successful cases into coordination cases and non-coordination cases by combining the proportion of the coordination times in the latest preset time;
The abnormal cases are determined by adopting a classification model based on an SVM algorithm of ACO optimization;
it should be further noted that, the adaptive adjustment coefficient and the random factor are introduced into the initiation function of the ACO algorithm, the distance between the current node and the optional node is considered in the construction process of the heuristic function, and heuristic information of the distance between the current node and the optional node is added, so that the function of the heuristic information is weakened at the later stage of the operation of the algorithm, wherein the calculation formula of the improved heuristic function is as follows:
Wherein Lij is the distance between grid i and grid j, ljt is the Euclidean distance between grid j and grid T, ljt is the Euclidean distance between grid j and grid T, which is a constant greater than 0, T is a dynamic coefficient, rand (0, 1) is a random number with a value range of 0 to 1, U is the current iteration number, and Umax is the maximum iteration number;
The ant colony algorithm is utilized to search and optimize punishment parameters and kernel function parameters of the SVM, and the flow of the algorithm is as follows:
Step 1: dividing the acquired data set into a training set and a testing set, respectively extracting the characteristics to form characteristic vectors of the training set and the testing set, and carrying out data normalization processing:
Step 2: initializing each parameter of ACO, randomly generating a combination of penalty parameters and kernel function parameters in a certain range as a whole solution space set I, and randomly placing ants in the solution space I;
step 3: starting ACO, training and learning the SVM by using a training set, wherein in the training process, the probability of the kth ant selecting the jth parameter combination from the set I at the t moment is updated, along with the iteration process, the pheromone concentration at the j position in the solution space is updated, and the step 3 is repeated, so that the obtained parameter combination is placed in the set FA;
Step 4: determining optimal parameters of an SVM algorithm based on the set FA, and completing construction of a classification model;
dividing the successful communication cases into matched cases and non-matched cases, wherein the method specifically comprises the following steps:
Based on the communication success times of the offender of the successful communication case, judging the communication keywords of the communication success times to determine whether the communication keywords contain non-matching keywords, if not, determining the communication success times as matching times;
Acquiring context information of a communication text of successful times of communication by utilizing BiGRU algorithm, learning emotional characteristics of a tag by utilizing a tag encoder, constructing attention weights of the word vectors by utilizing the tag precoder, generating text feature vectors based on the word vectors and the attention weights, and acquiring emotion recognition results of the successful times of communication by taking the obtained text feature vectors as input of a convolutional neural network;
Acquiring communication keywords of the communication success times of the offender of the successful communication case, carrying out a matching degree identification result of the communication success times based on the number of matching keywords and the number of non-matching keywords in the communication keywords, and dividing the communication success times into matching times and non-matching times based on the matching degree identification result and the emotion identification result;
determining the matching degree of the communicated successful cases by combining the proportion of matching times in the communication times of the offending subjects in the latest preset time based on the number of matching times and the number of non-matching times of the offending subjects of the communicated successful cases, and dividing the communicated successful cases into matching cases and non-matching cases based on the matching degree;
Determining the plumpness score of the adductor based on the communication duration, the communication times and the average feedback time of the telephone communication of the adductor in the set time and the communication duration, the communication times and the average feedback time of other communication modes;
the specific steps of the establishment of the plumpness score are as follows:
S41, determining whether the collecting person is a suspected active collecting person or not based on the communication times of telephone communication and other communication modes of the collecting person in a set time, if so, entering a step S42, and if not, entering a step S44;
s42, obtaining a telephone communication value based on the communication duration, the communication times and the average communication times of each existing prompting and receiving case of the prompting and receiving person in a set time, determining whether the prompting and receiving person is an active prompting and receiving person or not based on the telephone communication value, if so, constructing a fullness score based on the telephone communication value, otherwise, entering step S43;
S43, determining other communication values of the collecting staff based on the communication duration, the communication times and the average feedback time of other communication modes of the collecting staff in the set time, determining whether the collecting staff is an active collecting staff or not based on the other communication values, if so, constructing a fullness score based on the other communication values, and if not, entering step S44;
S44, constructing total communication times based on the communication times of telephone communication of the receiving personnel in the set time and the communication times of other communication modes, and constructing fullness scores based on the total communication times, telephone communication values and other communication values;
Determining the grading value of the collecting personnel based on the number of normal cases, the number of abnormal cases, the number of matched cases, the number of non-matched cases and the plumpness grading of the collecting personnel, and determining the collecting personnel to be distributed based on the grading value of the distributing personnel when the collecting personnel is determined to be the distributing personnel based on the grading value of the distributing personnel;
the specific steps of the determination of the allocatable personnel are as follows:
S51, determining whether the collecting person is an unallocated person or not based on the number of abnormal cases and the number of non-matched cases of the collecting person, if so, determining that the collecting person is not an allocable person, and if not, entering step S52;
S52, determining whether further determination of the plumpness score is needed or not based on the number of normal cases and the number of matched cases of the collecting staff, if so, entering a step S53, and if not, entering a step S54;
s53, determining whether the collecting personnel are assignable personnel or not based on the plumpness score of the collecting personnel, if so, determining that the collecting personnel are assignable personnel, and if not, entering step S54;
S54, determining the score value of the collecting personnel based on the fullness score of the collecting personnel, the number of normal cases, the number of abnormal cases, the number of matched cases and the number of non-matched cases, and determining whether the collecting personnel is an assignable personnel based on the score value.
2. The method for intelligent case division for overdue asset collection of claim 1, wherein the communication mode of the successful cases without communication comprises: mail communication, instant messaging software communication, short message communication, and telephone communication.
3. The method for intelligently sorting overdue asset collection according to claim 1, wherein the successful case is regarded as an abnormal case when the number of communication times of telephone communication of the successful case is greater than a set number of communication times.
4. The method of claim 1, wherein the value of the fullness score ranges from 0 to 1, and wherein the greater the value of the fullness score, the higher the fullness of the collector's job.
5. The method for intelligently sorting overdue asset collection according to claim 1, wherein the determination of the collection staff of the case to be distributed is performed based on the scoring value of the allocable staff, specifically comprising:
And determining the collecting staff of the case to be distributed based on the scoring value of the allocable staff.
6. A computer readable storage medium having stored thereon a computer program which, when executed in a computer, causes the computer to perform an overdue asset clearing intelligent case distribution method as claimed in any of claims 1-5.
7. A terminal device comprising a memory, a processor and a program stored on the memory and executable on the processor, the processor implementing the execution of an overdue asset clearing intelligent sort method as claimed in any one of claims 1 to 5 when the program is executed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310602327.5A CN116341879B (en) | 2023-05-26 | 2023-05-26 | Overdue asset collection intelligent case division method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310602327.5A CN116341879B (en) | 2023-05-26 | 2023-05-26 | Overdue asset collection intelligent case division method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116341879A CN116341879A (en) | 2023-06-27 |
| CN116341879B true CN116341879B (en) | 2024-05-31 |
Family
ID=86884360
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310602327.5A Active CN116341879B (en) | 2023-05-26 | 2023-05-26 | Overdue asset collection intelligent case division method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116341879B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116862668B (en) * | 2023-09-05 | 2023-11-24 | 杭州度言软件有限公司 | Intelligent collecting accelerating method for improving collecting accelerating efficiency |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE10322634A1 (en) * | 2003-05-20 | 2004-12-23 | TOS Team für Organisation und Systeme GmbH | Control method for a communication event, especially a telephone call, message transfer or data transmission, whereby communication initiator and target resource profiles are defined and used when generating a communication link |
| CN109064315A (en) * | 2018-08-02 | 2018-12-21 | 平安科技（深圳）有限公司 | Overdue bill intelligence collection method, apparatus, computer equipment and storage medium |
| CN110288464A (en) * | 2019-06-11 | 2019-09-27 | 深圳前海微众银行股份有限公司 | A kind of collection method, system and device |
| CN110728429A (en) * | 2019-09-18 | 2020-01-24 | 平安科技（深圳）有限公司 | Case allocation method, case allocation device, case allocation medium and computer equipment |
| CN110807699A (en) * | 2019-10-12 | 2020-02-18 | 上海上湖信息技术有限公司 | Overdue event payment collection method and device and computer readable storage medium |
| CN111539613A (en) * | 2020-04-20 | 2020-08-14 | 浙江网商银行股份有限公司 | Case distribution method and device |
| CN112365189A (en) * | 2020-11-30 | 2021-02-12 | 支付宝(杭州)信息技术有限公司 | Case distribution method and device |
| CN113642908A (en) * | 2021-08-20 | 2021-11-12 | 上海通联金融服务有限公司 | Multi-dimensional case allocation method based on credit card post-credit collection service |
| CN114626735A (en) * | 2022-03-23 | 2022-06-14 | 平安普惠企业管理有限公司 | Urging case allocation method, urging case allocation device, urging case allocation equipment and computer readable storage medium |
| CN114971284A (en) * | 2022-05-25 | 2022-08-30 | 平安银行股份有限公司 | Case distribution method and device and computer equipment |
| DE202022103268U1 (en) * | 2022-06-10 | 2022-09-03 | Puspanjali Mohapatra | A multi-objective feature selection system based on antlion optimization |
| CN115099531A (en) * | 2022-08-19 | 2022-09-23 | 国网江苏省电力有限公司苏州供电分公司 | Power transmission line lightning stroke early warning method and system based on support vector machine |
| CN115393077A (en) * | 2022-10-28 | 2022-11-25 | 深圳市人马互动科技有限公司 | Data processing method based on loan transaction man-machine conversation system and related device |
| CN115994818A (en) * | 2022-10-31 | 2023-04-21 | 兴业银行股份有限公司 | Intelligent collection system and method for special assets of bank |
| CN116011457A (en) * | 2022-12-08 | 2023-04-25 | 山东大学 | Emotion intelligent recognition method based on data enhancement and cross-modal feature fusion |
-
2023
- 2023-05-26 CN CN202310602327.5A patent/CN116341879B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE10322634A1 (en) * | 2003-05-20 | 2004-12-23 | TOS Team für Organisation und Systeme GmbH | Control method for a communication event, especially a telephone call, message transfer or data transmission, whereby communication initiator and target resource profiles are defined and used when generating a communication link |
| CN109064315A (en) * | 2018-08-02 | 2018-12-21 | 平安科技（深圳）有限公司 | Overdue bill intelligence collection method, apparatus, computer equipment and storage medium |
| CN110288464A (en) * | 2019-06-11 | 2019-09-27 | 深圳前海微众银行股份有限公司 | A kind of collection method, system and device |
| CN110728429A (en) * | 2019-09-18 | 2020-01-24 | 平安科技（深圳）有限公司 | Case allocation method, case allocation device, case allocation medium and computer equipment |
| CN110807699A (en) * | 2019-10-12 | 2020-02-18 | 上海上湖信息技术有限公司 | Overdue event payment collection method and device and computer readable storage medium |
| CN111539613A (en) * | 2020-04-20 | 2020-08-14 | 浙江网商银行股份有限公司 | Case distribution method and device |
| CN112365189A (en) * | 2020-11-30 | 2021-02-12 | 支付宝(杭州)信息技术有限公司 | Case distribution method and device |
| CN113642908A (en) * | 2021-08-20 | 2021-11-12 | 上海通联金融服务有限公司 | Multi-dimensional case allocation method based on credit card post-credit collection service |
| CN114626735A (en) * | 2022-03-23 | 2022-06-14 | 平安普惠企业管理有限公司 | Urging case allocation method, urging case allocation device, urging case allocation equipment and computer readable storage medium |
| CN114971284A (en) * | 2022-05-25 | 2022-08-30 | 平安银行股份有限公司 | Case distribution method and device and computer equipment |
| DE202022103268U1 (en) * | 2022-06-10 | 2022-09-03 | Puspanjali Mohapatra | A multi-objective feature selection system based on antlion optimization |
| CN115099531A (en) * | 2022-08-19 | 2022-09-23 | 国网江苏省电力有限公司苏州供电分公司 | Power transmission line lightning stroke early warning method and system based on support vector machine |
| CN115393077A (en) * | 2022-10-28 | 2022-11-25 | 深圳市人马互动科技有限公司 | Data processing method based on loan transaction man-machine conversation system and related device |
| CN115994818A (en) * | 2022-10-31 | 2023-04-21 | 兴业银行股份有限公司 | Intelligent collection system and method for special assets of bank |
| CN116011457A (en) * | 2022-12-08 | 2023-04-25 | 山东大学 | Emotion intelligent recognition method based on data enhancement and cross-modal feature fusion |
Non-Patent Citations (2)
| Title |
|---|
| 人工智能分案机制探析;金昌伟;;中国政法大学学报(第02期);全文 * |
| 基于信用管理策略下应收账款管理案例分析;艾建银;;科技经济导刊;20180125(第03期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116341879A (en) | 2023-06-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110334737B (en) | Customer risk index screening method and system based on random forest | |
| Özgür et al. | Adaptive anti-spam filtering for agglutinative languages: a special case for Turkish | |
| CN116341879B (en) | Overdue asset collection intelligent case division method and system | |
| CN106453033A (en) | Multilevel Email classification method based on Email content | |
| CN110245783B (en) | Short-term load prediction method based on C-means clustering fuzzy rough set | |
| CN110807699B (en) | Overdue event payment collection method and device and computer readable storage medium | |
| KR20200075120A (en) | Business default prediction system and operation method thereof | |
| WO2020024444A1 (en) | Group performance grade recognition method and apparatus, and storage medium and computer device | |
| CN113516340A (en) | Intelligent work order pushing method and device | |
| Mazumder et al. | Network intrusion detection using hybrid machine learning model | |
| Dada et al. | Random forests machine learning technique for email spam filtering | |
| Hebbar et al. | Comparison of machine learning techniques to predict the attrition rate of the employees | |
| US20240154975A1 (en) | Systems and methods for accelerating a disposition of digital dispute events in a machine learning-based digital threat mitigation platform | |
| Cui et al. | On effective e-mail classification via neural networks | |
| Sheu | An Efficient Two-phase Spam Filtering Method Based on E-mails Categorization. | |
| Zhao et al. | Debt detection in social security by sequence classification using both positive and negative patterns | |
| Özgür et al. | Spam mail detection using artificial neural network and Bayesian filter | |
| Sahithi et al. | Credit card fraud detection using ensemble methods in machine learning | |
| Soonthornphisaj et al. | Anti-spam filtering: a centroid-based classification approach | |
| CN116595486A (en) | Risk identification method, risk identification model training method and corresponding device | |
| CN113450207A (en) | Intelligent collection accelerating method, device, equipment and storage medium | |
| Ryu et al. | Ensemble classifier based on misclassified streaming data | |
| CN113721770A (en) | Method for providing voice help in intelligent household equipment and intelligent household equipment | |
| CN113850483A (en) | Enterprise credit risk rating system | |
| WO2020024448A1 (en) | Group performance grade identification method, device, storage medium, and computer apparatus |
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 |