CN113177165A - Improved KNN-based personalized inspection officer recommendation method - Google Patents

Improved KNN-based personalized inspection officer recommendation method Download PDF

Info

Publication number
CN113177165A
CN113177165A CN202110558100.6A CN202110558100A CN113177165A CN 113177165 A CN113177165 A CN 113177165A CN 202110558100 A CN202110558100 A CN 202110558100A CN 113177165 A CN113177165 A CN 113177165A
Authority
CN
China
Prior art keywords
case
workload
inspector
cases
expected workload
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.)
Granted
Application number
CN202110558100.6A
Other languages
Chinese (zh)
Other versions
CN113177165B (en
Inventor
栗伟
闵新�
杨金钊
王子晴
赵大哲
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Northeastern University China
Original Assignee
Northeastern University China
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Northeastern University China filed Critical Northeastern University China
Priority to CN202110558100.6A priority Critical patent/CN113177165B/en
Publication of CN113177165A publication Critical patent/CN113177165A/en
Application granted granted Critical
Publication of CN113177165B publication Critical patent/CN113177165B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9536—Search customisation based on social or collaborative filtering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
    • G06F18/24147—Distances to closest patterns, e.g. nearest neighbour classification
    • 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
    • G06Q10/063112—Skill-based matching of a person or a group to a task
    • 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
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/18—Legal services

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • Tourism & Hospitality (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Marketing (AREA)
  • Educational Administration (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Business, Economics & Management (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Biology (AREA)
  • Artificial Intelligence (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Technology Law (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Development Economics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention provides an improved KNN-based personalized recommendation method for a scout organ, and relates to the technical field of personalized recommendation. According to the method, the expected workload of the inspector is analyzed on the list of the inspector to be recommended, the expected workload of the current case handling of the inspector is predicted, the individualized inspector is recommended, and the newly arrived case is distributed. The distribution rationality of the cases is improved, and the case handling efficiency is improved.

Description

Improved KNN-based personalized inspection officer recommendation method
Technical Field
The invention relates to the technical field of personalized recommendation, in particular to a personalized recommendation method for a scout organ based on improved KNN.
Background
The KNN (K nearest neighbor) algorithm is one of the common classification algorithms in data mining, that is, K nearest neighbor nodes are selected from the target points to represent the target points by calculating euclidean distances between the target points and the neighbor nodes.
In the case handling process of the inspection institution, the case distribution problem is directly related to the case handling efficiency, the case handling enthusiasm of the inspection officer and the improvement of the incentive system. Since judicial reform, many scholars have studied how to achieve the optimal matching of cases and inspectors, but the current research is only in a qualitative analysis stage, and except for listing specific targets required to be completed by intelligent case division, factors influencing the intelligent case division result are not specifically given. Because different cases vary greatly in their type, the appropriate inspectors to handle will also vary. By analyzing the case workload and the capability of examining, managing and managing cases, the cases of different types are distributed, and the optimal matching between the cases and the examiners can be effectively realized by combining the current load of the examiners.
Disclosure of Invention
In order to solve the technical problems, the invention provides an improved KNN-based personalized recommendation method for an inspector, which is characterized in that the expected workload of the inspector is analyzed on a list of the inspectors to be recommended, the expected workload of the inspector handling cases currently is predicted, the personalized inspector is recommended, and the newly arrived cases are distributed.
A personalized recommendation method for a scout organ based on improved KNN comprises the following steps:
step 1: preprocessing the case handling data of the inspection hall, extracting the data of all inspectors to obtain an inspection operator list, and obtaining a handling history case set of each inspector and a current handling case set of the inspectors;
step 2: calculating expected workload of each case born by each inspector according to the cases historically handled by each inspector; then dividing the cases according to the case types, and calculating the expected workload of the inspectors for handling each type of cases according to the expected workload of the cases;
expected workload T of the casesAs shown in the following formula:
Figure BDA0003077902410000011
wherein dtsRepresenting cases sDuration of case handling, picRepresenting the weight of the c iteration, and rho representing the total number of cases in a fixed time interval;
calculating the average expected workload D of each type of caseSAs shown in the following formula:
Figure BDA0003077902410000012
wherein S represents the total number of cases of each type;
the average expected workload F for the inspector to handle each case is obtained as shown in the following formula:
Figure BDA0003077902410000021
where Q represents the total number of cases that the inspector transacted each type of case.
And step 3: according to the case currently handled by the inspection officer, the expected workload of each case currently handled by the inspection officer is predicted by using a regression decision tree model, and the current workload of the inspection officer is calculated based on the expected workload of the historical cases handled by the inspection officer, and the method specifically comprises the following steps:
step 3.1: in an input space where a training data set is located, recursively dividing each region into two sub-regions and determining an output value on each sub-region, and constructing a binary decision tree:
step 3.2: integrating the data set, namely integrating the expected workload of the cases, the average expected workload of each type of case, the average expected workload of the inspectors handling each type of case and the historical case set to generate a data set D;
step 3.3: establishing a regression prediction tree model, wherein the input is a data set D containing case expected workload, and the output is the current workload f of an inspectors(X)；
Step 3.4: selecting an optimal segmentation variable j and a segmentation point s of a data set D, solving the following formula, traversing the variable j, scanning the segmentation point s for the segmentation variable j, and selecting a pair (j, s) which enables the formula to reach the minimum value;
Figure BDA0003077902410000022
wherein the parameter c1,c2Sample values, R, representing leaf nodes1And R2The two regions represent two regions divided by a binary tree, i.e. a sample set, xiRepresenting case data, y, in a data set DiRepresenting the expected workload of each case in the data set D; the binary tree takes the characteristic value s of the variable j as a logic judgment condition;
step 3.5: the pair (j, s) obtained in step 3.4 is used to partition the region of the data set D and determine the corresponding output value, as shown in the following equation:
Figure BDA0003077902410000023
wherein N ismThe number of samples representing the two regions is,
Figure BDA0003077902410000024
represents a parameter c1,c2X represents the case data set.
Wherein R is1(j，s)＝{x|x(j)≤s}，R2(j,s)＝{x|x(j)>s}
Wherein x(j)Representing case data set under the segmentation variable j;
step 3.6: continue to two R1，R2The sub-region executes the step 3.4 and the step 3.5 until a fixed iteration number is met and stops;
step 3.7: dividing an input data set D into M regions R1，R2，...，RMGenerating a regression decision tree:
Figure BDA0003077902410000025
wherein I is a function of the indicator,
Figure BDA0003077902410000031
step 3.8: based on regression decisionsThe tree predicts the expected workload of handling cases of the current inspector and calculates the current workload f of the inspectors(X)：
Figure BDA0003077902410000032
Wherein N represents the total number of current case handling pieces of the inspector, f (x)i) Representing the expected workload of case i;
and 4, step 4: according to the expected workload of the historical case of the inspector and the current workload of the inspector, the improved KNN algorithm is utilized to carry out personalized recommendation of the inspector, and the inspector most suitable for handling the newly arrived case is obtained;
the personalized recommendation of the inspection officers is to select the inspection officer which is most matched with a newly arrived case from an inspection officer set to be recommended, and each case in the test data set is assigned with an optimal inspection officer;
step 4.1: predicting an expected workload of a new case as W using a regression prediction tree modelnewCalculating the average expected workload N of the newly arrived case and the casesavgDeviation Δ W ofnew；
ΔWnew＝Wnew-Navg
Step 4.2: calculating the expected workload M of the inspectors for handling the cases according to K inspectors in the set of inspectors to be recommended and by combining with the moving case of the newly-added casesnowAverage workload with this type of case NavgStandard deviation Δ M ofnowThe greater the deviation, the greater the ability of the reviewing officer to handle the type of case.
ΔMnow＝Navg-Mnow
Step 4.3: finding out Δ W according to nearest neighbor algorithm (KNN)newNearest neighbor Δ MnowIn combination with the current load f of the inspectors(X), recommending an optimal inspector;
Figure BDA0003077902410000033
wherein f iss(X)avgRepresents the average workload of the K inspectors to be recommended, fs(X)maxRepresents the maximum workload of the K inspectors to be recommended, fs(X)minRepresenting the minimum workload of the inspector to be recommended.
The invention has the following beneficial effects:
the technical scheme provides an inspection officer personalized recommendation method based on improved KNN, and the inspection officer personalized recommendation method can realize real-time personalized recommendation according to the complexity of cases and the current load of the inspection officer, improve the distribution rationality of the cases, and improve the case handling efficiency.
Drawings
FIG. 1 is a flow chart of a personalized recommendation method according to an embodiment of the invention;
FIG. 2 is a flow chart of a method for calculating an expected workload of a inspector according to an embodiment of the present invention;
FIG. 3 is a diagram of the number of cases handled by the inspector and the expected workload for an embodiment of the present invention;
wherein, the figure (a) is a schematic diagram of the number of cases handled by the inspector, and the figure (b) is a schematic diagram of the expected workload of the inspector in handling the cases of each type.
Detailed Description
The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
A personalized recommendation method for scout organs based on improved KNN, as shown in fig. 1, comprising the following steps:
step 1: preprocessing the case handling data of the inspection hall, extracting the data of all inspectors to obtain an inspection operator list, and obtaining a handling history case set of each inspector and a current handling case set of the inspectors;
in the embodiment, data are acquired at regular time and stored in the historical case database through an API (application program interface) data interface provided by a hospital for inspection. The collected case handling data comprises: case basic information table, personnel code table, case category code table, department business allocation table, personnel role allocation table, business code table, unit code table, file basic information table, file catalogue table, related property information table, related property identification information table, acceptance log table, case delivery log table and case delivery suspect record table.
Step 2: calculating expected workload of each case born by each inspector according to the cases historically handled by each inspector; then dividing the cases according to the case types, and calculating the expected workload of the inspectors for handling each type of cases according to the expected workload of the cases, as shown in FIG. 2;
the expected workload is calculated by adopting an EM iterative average calculation method, then according to the handling days of the historical cases, combining the historical handling case set and the current handling case set of each inspector, calculating the expected workload of the historical cases by adopting the EM iterative average calculation method, dividing the cases according to the case types, calculating the average expected workload of each type of case, and calculating the average expected workload of each type of case for the inspector, as shown in FIG. 3.
Expected workload T of the casesAs shown in the following formula:
Figure BDA0003077902410000041
wherein dtsIndicates the case-handling duration of the case s, picRepresenting the weight of the c iteration, and rho representing the total number of cases in a fixed time interval;
calculating the average expected workload D of each type of caseSAs shown in the following formula:
Figure BDA0003077902410000042
wherein S represents the total number of cases of each type;
the average expected workload F for the inspector to handle each case is obtained as shown in the following formula:
Figure BDA0003077902410000043
where Q represents the total number of cases that the inspector transacted each type of case.
And step 3: according to the case currently handled by the inspection officer, the expected workload of each case currently handled by the inspection officer is predicted by using a regression decision tree model, and the current workload of the inspection officer is calculated based on the expected workload of the historical cases handled by the inspection officer, and the method specifically comprises the following steps:
step 3.1: in an input space where a training data set is located, recursively dividing each region into two sub-regions and determining an output value on each sub-region, and constructing a binary decision tree:
step 3.2: integrating the data set, namely integrating the expected workload of the cases, the average expected workload of each type of case, the average expected workload of the inspectors handling each type of case and the historical case set to generate a data set D;
step 3.3: establishing a regression prediction tree model, wherein the input is a data set D containing case expected workload, and the output is the current workload f of an inspectors(X)；
Step 3.4: selecting an optimal segmentation variable j and a segmentation point s of a data set D, solving the following formula, traversing the variable j, scanning the segmentation point s for the segmentation variable j, and selecting a pair (j, s) which enables the formula to reach the minimum value;
Figure BDA0003077902410000051
wherein the parameter c1，c2Sample values, R, representing leaf nodes1And R2The two regions represent two regions divided by a binary tree, i.e. a sample set, xiRepresenting case data, y, in a data set DiRepresenting the expected workload of each case in the data set D; the binary tree takes the characteristic value s of the variable j as a logic judgment condition;
step 3.5: the pair (j, s) obtained in step 3.4 is used to partition the region of the data set D and determine the corresponding output value, as shown in the following equation:
Figure BDA0003077902410000052
wherein N ismThe number of samples representing the two regions is,
Figure BDA0003077902410000053
represents a parameter c1，c2X represents the case data set.
Wherein R is1(j，s)＝{x|x(j)≤s}，R2(j，s)＝{x|x(j)>s}
Wherein x(j)Representing case data set under the segmentation variable j;
step 3.6: continue to two R1，R2The sub-region executes the step 3.4 and the step 3.5 until a fixed iteration number is met and stops;
step 3.7: dividing an input data set D into M regions R1，R2，...，RMGenerating a regression decision tree:
Figure BDA0003077902410000054
wherein I is a function of the indicator,
Figure BDA0003077902410000055
step 3.8: predicting the expected workload of the current inspection officer handling case according to the regression decision tree, and calculating the current workload f of the inspection officers(X)：
Figure BDA0003077902410000056
Wherein N represents the total number of current case handling pieces of the inspector, f (x)i) Representing the expected workload of case i;
and 4, step 4: according to the expected workload of the historical case of the inspector and the current workload of the inspector, the improved KNN algorithm is utilized to carry out personalized recommendation of the inspector, and the inspector most suitable for handling the newly arrived case is obtained;
the personalized recommendation of the inspection officers is to select the inspection officer which is most matched with the newly arrived case from the set of inspection officers to be recommended, and each case in the test data set is assigned with the optimal inspection officer. In order to avoid the backlog of cases and inspectors frequently classified into a certain number of cases,
step 4.1: predicting an expected workload of a new case as W using a regression prediction tree modelnewCalculating the average expected workload N of the newly arrived case and the casesavgDeviation Δ W ofnew(ii) a The larger the deviation, the more complicated it is to reflect the newly arrived case.
ΔWnew＝Wnew-Navg
Step 4.2: calculating the expected workload M of the inspectors for handling the cases according to K inspectors in the set of inspectors to be recommended and by combining with the moving case of the newly-added casesnowAverage workload with this type of case NavgStandard deviation Δ M ofnowThe greater the deviation, the greater the ability of the reviewing officer to handle the type of case.
ΔMnow＝Navg-Mnow
Step 4.3: finding out Δ W according to nearest neighbor algorithm (KNN)newNearest neighbor Δ MnowIn combination with the current load f of the inspectors(X), recommending an optimal inspector;
Figure BDA0003077902410000061
wherein f iss(X)avgRepresents the average workload of the K inspectors to be recommended, fs(X)maxRepresents the maximum workload of the K inspectors to be recommended, fs(X)minRepresenting the minimum workload of the inspector to be recommended.
The personalized recommendation method for the inspection officer provided in the embodiment tests total 1364 case data of a certain-level inspection yard and 2 subordinate regional inspection yards, calculates the personalized recommendation accuracy rate of the inspection officer, and test results are as follows:
TABLE 1 survey officer personalized recommendation List
Inspection yard Number of examiners Total number of case types Assign exact number Total number of cases Rate of accuracy of case classification
Inspection institute of city 27 56 265 305 88.2％
Regional inspection yard of city 15 50 518 588 90.1％
Regional inspection yard of city 18 46 414 471 89.3％
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present 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 solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; such modifications and substitutions do not depart from the spirit of the corresponding technical solutions and scope of the present invention as defined in the appended claims.

Claims (4)

1. A personalized recommendation method for a scout organ based on improved KNN is characterized by comprising the following steps:
step 1: preprocessing the case handling data of the inspection hall, extracting the data of all inspectors to obtain an inspection operator list, and obtaining a handling history case set of each inspector and a current handling case set of the inspectors;
step 2: calculating expected workload of each case born by each inspector according to the cases historically handled by each inspector; then dividing the cases according to the case types, and calculating the expected workload of the inspectors for handling each type of cases according to the expected workload of the cases;
and step 3: according to the case currently handled by the inspection officer, predicting the expected workload of each case currently handled by the inspection officer by using a regression decision tree model, and calculating the current workload of the inspection officer based on the expected workload of the historical cases handled by the inspection officer;
and 4, step 4: and selecting the inspector which is most matched with the newly arrived case from the set of inspectors to be recommended by utilizing an improved KNN algorithm according to the expected workload of the historical cases of the inspector and the current workload of the inspector, distributing an optimal inspector to each case in the test data set, and performing personalized recommendation of the inspectors to obtain the inspector which is most suitable for handling the newly arrived case.
2. The personalized recommendation method for scout officer based on improved KNN as claimed in claim 1, wherein the case expected workload T in step 2sAs shown in the following formula:
Figure FDA0003077902400000011
wherein dtsIndicates the case-handling duration of the case s, picRepresenting the weight of the c iteration, and rho representing the total number of cases in a fixed time interval;
calculating the average expected workload D of each type of caseSAs shown in the following formula:
Figure FDA0003077902400000012
wherein S represents the total number of cases of each type;
the average expected workload F for the inspector to handle each case is obtained as shown in the following formula:
Figure FDA0003077902400000013
where Q represents the total number of cases that the inspector transacted each type of case.
3. The personalized recommendation method for scout officers based on improved KNN as claimed in claim 1, wherein the step 3 specifically comprises the following steps:
step 3.1: in an input space where a training data set is located, recursively dividing each region into two sub-regions and determining an output value on each sub-region, and constructing a binary decision tree:
step 3.2: integrating the data set, namely integrating the expected workload of the cases, the average expected workload of each type of case, the average expected workload of the inspectors handling each type of case and the historical case set to generate a data set D;
step 3.3: establishing a regression prediction tree model, wherein the input is a data set D containing case expected workload, and the output is the current workload f of an inspectors(X)；
Step 3.4: selecting an optimal segmentation variable j and a segmentation point s of a data set D, solving the following formula, traversing the variable j, scanning the segmentation point s for the segmentation variable j, and selecting a pair (j, s) which enables the formula to reach the minimum value;
Figure FDA0003077902400000014
wherein the parameter c1,c2Sample values, R, representing leaf nodes1And R2The two regions represent two regions divided by a binary tree, i.e. a sample set, xiRepresenting case data, y, in a data set DiRepresenting the expected workload of each case in the data set D; the binary tree takes the characteristic value s of the variable j as a logic judgment condition;
step 3.5: the pair (j, s) obtained in step 3.4 is used to partition the region of the data set D and determine the corresponding output value, as shown in the following equation:
Figure FDA0003077902400000021
wherein N ismThe number of samples representing the two regions is,
Figure FDA0003077902400000022
represents a parameter c1,c2X represents a case data set;
wherein R is1(j,s)＝{x|x(j)≤s},R2(j,s)＝{x|x(j)>s}
Wherein x(j)Representing case data set under the segmentation variable j;
step 3.6: continue to two R1，R2The sub-region executes the step 3.4 and the step 3.5 until a fixed iteration number is met and stops;
step 3.7: dividing an input data set D into M regions R1，R2，...，RMGenerating a regression decision tree:
Figure FDA0003077902400000023
wherein I is a function of the indicator,
Figure FDA0003077902400000024
step 3.8: predicting the expected workload of the current inspection officer handling case according to the regression decision tree, and calculating the current workload f of the inspection officers(X)：
Figure FDA0003077902400000025
Wherein N represents the total number of current case handling pieces of the inspector, f (x)i) Representing the expected workload of case i.
4. The personalized recommendation method for scout officers based on improved KNN as claimed in claim 1, wherein the step 4 comprises the following steps:
step 4.1: predicting an expected workload of a new case as W using a regression prediction tree modelnewCalculating the average expected workload N of the newly arrived case and the casesavgDeviation Δ W ofnew；
ΔWnew＝Wnew-Navg
Step 4.2: calculating the expected workload M of the inspectors for handling the cases according to K inspectors in the set of inspectors to be recommended and by combining with the moving case of the newly-added casesnowAverage workload with this type of case NavgStandard deviation Δ M ofnowThe larger the deviation is, the stronger the ability of the inspection officer to handle the type of case is reflected;
ΔMnow＝Navg-Mnow
step 4.3: finding out Δ W according to nearest neighbor algorithm (KNN)newNearest neighbor Δ MnowIn combination with the current load f of the inspectors(X), optimal inspection is recommendedAn officer;
Figure FDA0003077902400000031
wherein f iss(X)avgRepresents the average workload of the K inspectors to be recommended, fs(X)maxRepresents the maximum workload of the K inspectors to be recommended, fs(X)minRepresenting the minimum workload of the inspector to be recommended.
CN202110558100.6A 2021-05-21 2021-05-21 Inspection officer personalized recommendation method based on improved KNN Active CN113177165B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110558100.6A CN113177165B (en) 2021-05-21 2021-05-21 Inspection officer personalized recommendation method based on improved KNN

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110558100.6A CN113177165B (en) 2021-05-21 2021-05-21 Inspection officer personalized recommendation method based on improved KNN

Publications (2)

Publication Number Publication Date
CN113177165A true CN113177165A (en) 2021-07-27
CN113177165B CN113177165B (en) 2023-07-21

Family

ID=76929613

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110558100.6A Active CN113177165B (en) 2021-05-21 2021-05-21 Inspection officer personalized recommendation method based on improved KNN

Country Status (1)

Country Link
CN (1) CN113177165B (en)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2715433A1 (en) * 2009-09-23 2011-03-23 Courtview Justice Solutions Judicial case management computer system and method for providing the same
CA2820657A1 (en) * 2012-07-13 2014-01-13 Teleride/Sage Ltd. System and method for reducing trial requests and backlog of court cases
CN106845777A (en) * 2016-12-21 2017-06-13 北京华宇信息技术有限公司 Case workload evaluation method, judge's performance evaluation techniques, modeling method, computing device and computer-readable recording medium
CN110009182A (en) * 2019-01-28 2019-07-12 广州大学 A kind of law court's case automation division method and system
CN111369208A (en) * 2019-12-30 2020-07-03 云南省昆明市人民检察院 Intelligent auxiliary case handling system for criminal penalty change execution inspection
CN111899133A (en) * 2020-08-28 2020-11-06 四川省广安市人民检察院 Informatization platform construction method and application based on civil inspection service
CN112053074A (en) * 2020-09-11 2020-12-08 南京通达海科技股份有限公司 Automatic case dividing system and method for court cases
CN112163423A (en) * 2020-09-29 2021-01-01 东北大学 Method and system for calculating handling workload of inspection case
CN112200484A (en) * 2020-10-26 2021-01-08 东北大学 Intelligent case division system and method based on strategy definition

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2715433A1 (en) * 2009-09-23 2011-03-23 Courtview Justice Solutions Judicial case management computer system and method for providing the same
CA2820657A1 (en) * 2012-07-13 2014-01-13 Teleride/Sage Ltd. System and method for reducing trial requests and backlog of court cases
CN106845777A (en) * 2016-12-21 2017-06-13 北京华宇信息技术有限公司 Case workload evaluation method, judge's performance evaluation techniques, modeling method, computing device and computer-readable recording medium
CN110009182A (en) * 2019-01-28 2019-07-12 广州大学 A kind of law court's case automation division method and system
CN111369208A (en) * 2019-12-30 2020-07-03 云南省昆明市人民检察院 Intelligent auxiliary case handling system for criminal penalty change execution inspection
CN111899133A (en) * 2020-08-28 2020-11-06 四川省广安市人民检察院 Informatization platform construction method and application based on civil inspection service
CN112053074A (en) * 2020-09-11 2020-12-08 南京通达海科技股份有限公司 Automatic case dividing system and method for court cases
CN112163423A (en) * 2020-09-29 2021-01-01 东北大学 Method and system for calculating handling workload of inspection case
CN112200484A (en) * 2020-10-26 2021-01-08 东北大学 Intelligent case division system and method based on strategy definition

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
杜豫苏; 高伟; 何育凯: "专业法官会议运行机制构建再思考——基于审委会、审联会、专业法官会"三会"比较的视角", 山东法官培训学院学报, vol. 34, no. 6, pages 139 - 148 *
林文水;王建东;: "工作流技术在法院管理系统中的应用", 计算机与现代化, no. 04, pages 124 - 127 *
郑曦;: "人工智能技术在司法裁判中的运用及规制", 中外法学, vol. 32, no. 03, pages 674 - 696 *

Also Published As

Publication number Publication date
CN113177165B (en) 2023-07-21

Similar Documents

Publication Publication Date Title
WO2019165673A1 (en) Reimbursement form risk prediction method, apparatus, terminal device, and storage medium
CN116523262B (en) Intelligent planning method, system and medium for production plan based on big data
CN107037980A (en) Many expressions storage of time series data
US20230269265A1 (en) Systems and methods for cybersecurity risk mitigation and management
CN114742477B (en) Enterprise order data processing method, device, equipment and storage medium
CN115757660A (en) Method, system, electronic equipment and program product for automatically generating incidence relation between database tables
CN115358481A (en) Early warning and identification method, system and device for enterprise ex-situ migration
CN113177165A (en) Improved KNN-based personalized inspection officer recommendation method
Hu Overdue invoice forecasting and data mining
Deng et al. A novel method for elimination of inconsistencies in ordinal classification with monotonicity constraints
CN115660730A (en) Loss user analysis method and system based on classification algorithm
Hwang et al. Building an Analytical Platform of Big Data for Quality Inspection in the Dairy Industry: A Machine Learning Approach
CN115034520A (en) Risk prediction method, device, equipment and storage medium
CN114493234A (en) Method for identifying key pressure control points of water supply pipe network
Akinci et al. Comparison of iron and steel production defects using classification algorithms
CN117150097B (en) Automatic matching method for law enforcement checklist
CN115423148B (en) Agricultural machinery operation performance prediction method and device based on Ke Li jin method and decision tree
AlDarmaki et al. Prediction of the closing price in the Dubai financial market: a data mining approach
CN117807377B (en) Multidimensional logistics data mining and predicting method and system
CN117541200B (en) Project management method and system based on LTC flow
US20230334496A1 (en) Automated transaction clustering based on rich, non-human filterable risk elements
Pakhomova et al. The income prediction module of the retail store's network
CN111626586B (en) Data quality detection method, device, computer equipment and storage medium
Kheirallah et al. Anomaly Detection in a Production Line: an Online/Offline Statistical Learning Approach
TR2022010385A2 (en) FACTORING CUSTOMER ANALYSIS SYSTEM AND METHOD

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