Extraction algorithm for government administrative judgment and case handling personnel
Through the algorithm of multi-dimensional scoring and weighted average value sorting, government administrative ruling case handlers are scientifically and reasonably allocated, solving the problem of unreasonable workload allocation in the existing technology, and improving administrative ruling efficiency and user satisfaction.
Patent Information
- Application Number
- CN202411291763.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-14
- Publication Date
- 2024-12-17
AI Technical Summary
In the existing technology, the distribution method of government administrative ruling case handlers is unreasonable, especially in government procurement projects, some case handlers are responsible for a large number of cases while others are idle, and with the establishment of the administrative ruling center , the traditional distribution method can no longer meet the needs of standardization and institutionalization.
A government administrative ruling case handler extraction algorithm is used to calculate the weighted score of each case handler by assigning weighted values to the case handler, based on multi-dimensional data such as project matching degree, party characteristics, current and annual case volume, and pass the weighted average Value sorting determines the optimal case handler to ensure the scientificity and rationality of case allocation.
It improves the efficiency of the administrative ruling process, reduces the probability of administrative reconsideration and litigation, improves user satisfaction, and ensures the reasonable allocation of cases and the reasonable load of case handlers.
Smart Images
Figure CN119151229A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of information processing technology, and in particular to an algorithm for extracting government administrative adjudication case handlers. Background Art
[0002] At present, the way to assign complaint case handlers is generally that the personnel of the internal finance department responsible for the project procurement plan handle the complaint cases, and the basis for the assignment is mainly the subject items of the procurement project. Since the occurrence of complaints about government procurement projects is a random event, from a statistical perspective, some subject items also have a higher probability of complaints, which may result in some case handlers being responsible for a large number of cases while other case handlers are relatively idle. Therefore, the general method of assigning case handlers has the situation of unreasonable workload distribution.
[0003] In addition, in recent years, provincial finance departments in many parts of the country have successively established government procurement administrative adjudication centers, aiming to promote the standardization, regularization and institutionalization of government procurement administrative adjudication work across the country. In administrative adjudication centers, case handling personnel generally tend to be full-time, and the previous method of assigning complaint case handling personnel is no longer feasible. Summary of the invention
[0004] The purpose of the present invention is to provide an algorithm for extracting government administrative adjudication case handlers, which can provide a reasonable solution for the link of assigning case handlers for administrative adjudication and provide a scientific basis for the extraction results, thereby enhancing the case handling efficiency of the financial department.
[0005] The purpose of the present invention is achieved through the following technical solutions:
[0006] An algorithm for extracting government administrative adjudication case handlers includes the following steps:
[0007] S1. Obtain the current case handlers who are not on leave from the full list of case handlers, form an on-the-job list table A, and assign a weight value AW to the on-the-job list table A;
[0008] S2. Associate the full list of case handlers with the project type to generate a project matching table. The project matching table includes project types, case handlers and their corresponding matching scores. There are multiple project types in the project matching table. Weight values are assigned to each project type according to the type of complaint case. The weighted score of each case handler is calculated, and the project matching score table B is formed according to the score arrangement. The weight value BW is assigned to the project matching score table B.
[0009] S3. Associate the full list of case handlers with the characteristics of the parties to generate a party matching table. The party matching table includes the characteristics of the parties, the case handlers and their corresponding matching scores. There are multiple party characteristics in the party matching table. A weight value is assigned to each party characteristic according to the type of complaint case. The weighted score of each case handler is calculated, and the party matching score table C is formed according to the score arrangement, and a weight value CW is assigned to the party matching score table C;
[0010] S4. Associate the full list of case handlers with the current number of cases being handled, and generate a current case handler table. The current case handler table includes the case handlers and their corresponding current number of cases being handled and scores. The case handlers with higher number of cases being handled have lower scores. The current case handler score table D is formed according to the score arrangement, and a weight value DW is assigned to the current case handler score table D.
[0011] S5. Associate the full list of case handlers with the annual caseload to generate an annual case handler table, which includes case handlers and their corresponding annual caseloads and scores. Case handlers with higher annual caseloads have lower scores. The annual case handler score table E is formed based on the score arrangement, and a weight value EW is assigned to the annual case handler score table E;
[0012] S6. Perform an intersection operation on the in-service list table A, the project matching score table B, the client matching score table C, the current case handling staff score table D, and the annual case handling staff score table E to obtain the final staff list S;
[0013] S7. Calculate the weighted average of each case handler in the final personnel list S, sort them from high to low according to the weighted average, and select the case handler with the highest weighted average as the final case handler;
[0014] The weighted average is calculated as follows:
[0015]
[0016] Among them, the weight value is the weight value of each dimension in steps S1-S5, and the score is the score of the case handler in the table of each dimension.
[0017] Furthermore, the calculation formula of the weighted score in step S2 is as follows:
[0018] Weighted score = ∑ (project type weight value × matching score).
[0019] Furthermore, there are three project types in the project matching table, namely service projects, engineering projects, and procurement projects.
[0020] Furthermore, the calculation formula of the weighted score in step S3 is as follows:
[0021] Weighted score = ∑ (parties’ characteristic weight value × matching score).
[0022] Furthermore, there are three party characteristics in the party matching table, namely, the type of case handling for complaints about procurement documents, the type of case handling for complaints about procurement processes, and the type of case handling for complaints about procurement results.
[0023] Furthermore, the scoring rule in step S4 is:
[0024] The current number of cases in progress falls into the range of 0-5, with a corresponding weight of 55%; the current number of cases in progress falls into the range of 6-30, with a corresponding weight of 20%; the current number of cases in progress falls into the range of 31-50, with a corresponding weight of 15%; the current number of cases in progress is greater than 50, with a corresponding weight of 10%;
[0025] Rating = full score × weight value.
[0026] Compared with the prior art, the advantages of the present invention are: the algorithm of the present invention aims to comprehensively analyze the adaptability of case handlers and complaint cases in the administrative adjudication information system, based on the case handling volume, historical case volume, historical case types, project information of complaint cases, complaint types, relevant party information and other data of case handlers, combined with a certain random probability, to determine the optimal case handler for the current complaint case, thereby improving the timeliness of responses to inquiries, reducing the probability of subsequent administrative reconsideration and administrative litigation, and improving user satisfaction of all parties. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 It is a structural schematic diagram of an embodiment of the present invention. DETAILED DESCRIPTION
[0028] The present invention is described in detail below with reference to the accompanying drawings and embodiments:
[0029] like Figure 1 Shown is a schematic diagram of an implementation example of an algorithm for extracting government administrative adjudication case handlers provided by the present invention.
[0030] An algorithm for extracting government administrative adjudication case handlers includes the following steps:
[0031] Step 1
[0032] The current case handlers who are not on leave are obtained from the full list of case handlers to form an on-the-job list table A, and a weight value AW is assigned to the on-the-job list table A; the weight value AW is 0.
[0033] Step 2
[0034] The full list of case handlers is associated with the project type to generate a project matching table, which includes project types, case handlers and their corresponding matching scores. There are multiple project types in the project matching table. Weight values are assigned to each project type according to the type of complaint case. The weighted score of each case handler is calculated, and the scores are arranged to form a project matching score table B, and a weight value BW is assigned to the project matching score table B.
[0035] The program match table can be a spreadsheet or database where each row represents a caseworker and each column represents a different program type and match score.
[0036] The weighted score is calculated as follows:
[0037] Weighted score = ∑ (project type weight value × matching score).
[0038] Project type weight values can be determined based on the project's urgency, complexity, or other criteria.
[0039] In one embodiment, there are three types of projects in the project matching table, namely, service projects, engineering projects, and procurement projects.
[0040] Assume a project matching table as follows:
[0041] Case Investigators Project Type A Project Type B Project Type C Zhang San 90 80 70 Li Si 85 85 90 Wang Wu 75 95 80
[0042] If we assign weights of 3, 2, and 1 to project types A, B, and C respectively, we can calculate the weighted score of each case handler:
[0043] Zhang San's weighted score = 3 × 90 + 2 × 80 + 1 × 70 = 490
[0044] Li Si's weighted score = 3 × 85 + 2 × 85 + 1 × 90 = 450
[0045] Wang Wu’s weighted score = 3×75+2×95+1×80=475
[0046] According to the weighted score, Zhang San is the case handler with the highest matching degree in this dimension.
[0047] Step 3
[0048] The full list of case handlers is associated with the characteristics of the parties to generate a party matching table. The party matching table includes the characteristics of the parties, the case handlers and their corresponding matching scores. There are multiple party characteristics in the party matching table. A weight value is assigned to each party characteristic according to the type of complaint case. The weighted score of each case handler is calculated, and the party matching score table C is formed according to the score arrangement, and the weight value CW is assigned to the party matching score table C;
[0049] The client match table can be a spreadsheet or database where each row represents a case officer and each column represents a different client characteristic and match score.
[0050] The weighted score is calculated as follows:
[0051] Weighted score = ∑ (parties’ characteristic weight value × matching score).
[0052] The weights of party characteristics may be determined based on the relevance, importance, or other criteria of the characteristics.
[0053] In one embodiment, there are three party characteristics in the party matching table, namely, the type of case handling for complaints about procurement documents, the type of case handling for complaints about procurement processes, and the type of case handling for complaints about procurement results.
[0054] Assume a party matching table as follows:
[0055] Case Investigators Characteristics of the parties Characteristics of the parties Characteristics of the parties Zhang San 90 80 70 Li Si 85 85 90 Wang Wu 75 95 80
[0056] If the client characteristics X, Y, and Z are assigned weights of 3, 2, and 1 respectively, the matching score of each case handler can be calculated:
[0057] Zhang San's matching score = 3 × 90 + 2 × 80 + 1 × 70 = 490
[0058] Li Si's matching score = 3×85+2×85+1×90=450
[0059] Wang Wu's matching score = 3 × 75 + 2 × 95 + 1 × 80 = 475
[0060] According to the matching score, Zhang San is the case handler with the highest matching degree in this dimension.
[0061] Step 4
[0062] The full list of case handlers is associated with the current number of cases being handled, and a current case handler table is generated. The current case handler table includes the case handlers and their corresponding current number of cases being handled and scores. The case handlers with higher number of cases being handled have lower scores. The current case handler score table D is formed according to the score arrangement, and a weight value DW is assigned to the current case handler score table D;
[0063] Scoring Rules:
[0064] The current number of cases in progress falls into the range of 0-5, with a corresponding weight of 55%; the current number of cases in progress falls into the range of 6-30, with a corresponding weight of 20%; the current number of cases in progress falls into the range of 31-50, with a corresponding weight of 15%; the current number of cases in progress is greater than 50, with a corresponding weight of 10%;
[0065] Rating = full score × weight value.
[0066] The weight values of the above scoring rules can be configured dynamically.
[0067] Step 5
[0068] The full list of case handlers is associated with the annual caseload to generate an annual case handler table, which includes case handlers and their corresponding annual caseloads and scores. Case handlers with higher annual caseloads have lower scores. The annual case handler score table E is formed according to the score arrangement, and a weight value EW is assigned to the annual case handler score table E;
[0069] The scoring rules for step 5 are the same as those for step 4.
[0070] Step 6
[0071] Perform intersection operations on the in-service list table A, the project matching score table B, the client matching score table C, the current case handler score table D, and the annual case handler score table E to find the case handlers who meet all the dimensional conditions at the same time and obtain the final personnel list S. In programming, this can be achieved through the intersection operation of sets.
[0072] Step 7
[0073] Calculate the weighted average of each case handler in the final personnel list S, sort them from high to low according to the weighted average, and select the case handler with the highest weighted average as the final case handler;
[0074] The weighted average is calculated as follows:
[0075]
[0076] Among them, the weight value is the weight value of each dimension (AW, BW, CW, DW, EW) in steps S1-S5, and the score is the score of the case handler in the table of each dimension.
[0077] Algorithm example:
[0078] Assume that there are the following five dimensions of weights and scores:
[0079] AW: The weight is 3, and the score list is [Zhang San: 90, Li Si: 85, Wang Wu: 80]
[0080] BW: weight is 2, the score list is [Zhang San: 85, Li Si: 90, Wang Wu: 75]
[0081] CW: weight is 1, the score list is [Zhang San: 70, Li Si: 80, Wang Wu: 90]
[0082] DW: The weight is 4, and the score list is [Zhang San: 80, Li Si: 75, Wang Wu: 95]
[0083] EW: The weight is 5, and the score list is [Zhang San: 95, Li Si: 90, Wang Wu: 85]
[0084] The weighted average for each case officer can be calculated:
[0085]
[0086] Based on the weighted average, we can select the case handler with the highest weighted average as the final user.
[0087] Supplementary explanation: When the number of case handlers obtained according to the calculation rules is empty, the compensation mechanism will randomly obtain the case handlers who are not on vacation to proceed with the case direction.
[0088] Based on the above dimensions, horizontal expansion, vertical expansion and extension, and operation rules can be carried out;
[0089] Horizontal expansion: You can increase or decrease the corresponding dimensions according to your business scenarios, or adjust the weight of a certain dimension;
[0090] Vertical expansion: adjustment of the weight values of each dimension and change of the calculation method in rule calculation;
[0091] Calculation rules: built-in: weighted calculation, weighted average calculation, custom rules.
[0092] In practical applications, in addition to the project matching score, other factors may need to be considered, such as the availability, historical performance, personal preferences, etc. of the case handlers. The above factors can also be integrated into the final score in a weighted manner.
Claims
1. An algorithm for extracting government administrative adjudication case handlers, characterized in that: The following steps are involved: S1. Obtain the current case handlers who are not on leave from the full list of case handlers, form an on-the-job list table A, and assign a weight value AW to the on-the-job list table A; S2. Associate the full list of case handlers with the project type to generate a project matching table. The project matching table includes project types, case handlers and their corresponding matching scores. There are multiple project types in the project matching table. Weight values are assigned to each project type according to the type of complaint case. The weighted score of each case handler is calculated, and the project matching score table B is formed according to the score arrangement. The weight value BW is assigned to the project matching score table B. S3. Associate the full list of case handlers with the characteristics of the parties to generate a party matching table. The party matching table includes the characteristics of the parties, the case handlers and their corresponding matching scores. There are multiple party characteristics in the party matching table. A weight value is assigned to each party characteristic according to the type of complaint case. The weighted score of each case handler is calculated, and the party matching score table C is formed according to the score arrangement, and a weight value CW is assigned to the party matching score table C; S4. Associate the full list of case handlers with the current number of cases being handled, and generate a current case handler table. The current case handler table includes the case handlers and their corresponding current number of cases being handled and scores. The case handlers with higher number of cases being handled have lower scores. The current case handler score table D is formed according to the score arrangement, and a weight value DW is assigned to the current case handler score table D. S5. Associate the full list of case handlers with the annual caseload to generate an annual case handler table, which includes case handlers and their corresponding annual caseloads and scores. Case handlers with higher annual caseloads have lower scores. The annual case handler score table E is formed based on the score arrangement, and a weight value EW is assigned to the annual case handler score table E; S6. Perform an intersection operation on the in-service list table A, the project matching score table B, the client matching score table C, the current case handling staff score table D, and the annual case handling staff score table E to obtain the final staff list S; S7. Calculate the weighted average of each case handler in the final personnel list S, sort them from high to low according to the weighted average, and select the case handler with the highest weighted average as the final case handler; The weighted average is calculated as follows: Among them, the weight value is the weight value of each dimension in steps S1-S5, and the score is the score of the case handler in the table of each dimension.
2. The algorithm for extracting government administrative adjudication case handlers according to claim 1 is characterized by: The calculation formula of the weighted score in step S2 is as follows: Weighted score = ∑ (project type weight value × matching score).
3. The algorithm for extracting government administrative adjudication case handlers according to claim 2 is characterized by: There are three project types in the project matching table, namely service projects, engineering projects, and procurement projects.
4. The algorithm for extracting government administrative adjudication case handlers according to claim 1 is characterized by: The calculation formula of the weighted score in step S3 is as follows: Weighted score = ∑ (parties’ characteristic weight value × matching score).
5. The algorithm for extracting government administrative adjudication case handlers according to claim 4 is characterized by: There are three party characteristics in the party matching table, namely, the type of case handling for complaints about procurement documents, the type of case handling for complaints about procurement processes, and the type of case handling for complaints about procurement results.
6. The algorithm for extracting government administrative adjudication case handlers according to claim 1 is characterized by: Scoring rules in step S4: The current number of cases in progress falls into the range of 0-5, with a corresponding weight of 55%; the current number of cases in progress falls into the range of 6-30, with a corresponding weight of 20%; the current number of cases in progress falls into the range of 31-50, with a corresponding weight of 15%; the current number of cases in progress is greater than 50, with a corresponding weight of 10%; Rating = full score × weight value.