WO2020140391A1 - 城市网格分配方法、装置及电子终端 - Google Patents

城市网格分配方法、装置及电子终端 Download PDF

Info

Publication number
WO2020140391A1
WO2020140391A1 PCT/CN2019/091101 CN2019091101W WO2020140391A1 WO 2020140391 A1 WO2020140391 A1 WO 2020140391A1 CN 2019091101 W CN2019091101 W CN 2019091101W WO 2020140391 A1 WO2020140391 A1 WO 2020140391A1
Authority
WO
WIPO (PCT)
Prior art keywords
grid
cases
candidate
neighborhood
current
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.)
Ceased
Application number
PCT/CN2019/091101
Other languages
English (en)
French (fr)
Inventor
吴育人
魏萍
王淼石
庄伯金
肖京
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.)
Ping An Technology Shenzhen Co Ltd
Original Assignee
Ping An Technology Shenzhen Co Ltd
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 Ping An Technology Shenzhen Co Ltd filed Critical Ping An Technology Shenzhen Co Ltd
Publication of WO2020140391A1 publication Critical patent/WO2020140391A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06QINFORMATION 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
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06QINFORMATION 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
    • G06Q50/00Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10Services
    • G06Q50/26Government or public services

Definitions

  • This application relates to the field of data processing technology, and in particular, to a method, device and electronic terminal for urban grid distribution.
  • City management has an important impact on the city's economic development and people's lives. It is an objective requirement of city development. Among them, in order to facilitate the management of urban areas, the existing technology proposes grid city management.
  • the grid city management method is to divide the target geographic area of management into multiple city grids, and then perform these city grids. management.
  • the city management is mainly performed by the administrators. Therefore, how to allocate management areas to the administrators to ensure the rationality of the work assigned to the administrators is an urgent problem to be solved.
  • the embodiments of the present application provide a city grid distribution method, device and electronic terminal, which can reasonably distribute the workload of the administrator.
  • an embodiment of the present application provides a method for allocating an urban grid, the method for allocating an urban grid includes:
  • the neighborhood grid with the shortest travel time is selected from the neighborhood grid of the current grid to obtain a candidate grid P 1 ;
  • the current grid and the candidate grid P 1 are allocated to the administrator.
  • an embodiment of the present application provides an urban grid distribution device, the urban grid distribution device includes:
  • An obtaining unit configured to obtain the number of cases corresponding to the current grid where the administrator is located, and the number of cases and travel time corresponding to each neighborhood grid adjacent to the current grid;
  • the selection unit is configured to filter out the neighborhood grid with the shortest travel time from the neighborhood grid of the current grid according to the travel time corresponding to each neighborhood grid to obtain a candidate grid P 1 ;
  • the judging unit is used to judge whether the sum of the first case corresponding to the number of cases in the candidate grid P 1 and the number of cases in the current grid is within a preset range;
  • the processing unit is configured to allocate the current grid and the candidate grid P 1 to the administrator if it is determined that the sum of the first cases is within the preset range.
  • an embodiment of the present application provides an electronic terminal, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, the processor executing the computer When the instructions are readable, the processor implements the urban grid allocation method described in any one of the first aspects.
  • an embodiment of the present application provides a non-volatile computer-readable storage medium having computer-readable instructions stored thereon.
  • the computer When the computer reads and executes the computer-readable instructions, the computer can implement any of the above A method of urban grid allocation.
  • the technical solution provided by the present application when assigning the corresponding city grid to the administrator, selects the neighborhood grid with the shortest travel time from the neighborhood grid of the current grid by using the travel time corresponding to the grid to obtain the candidate grid P 1 , and also use the preset range, when it is determined that the sum of the first case corresponding to the number of cases in the candidate grid P 1 and the number of cases in the current grid is within the preset range, the current grid The candidate grid P 1 is assigned to the administrator.
  • the condition of the shortest travel time is used to filter the neighborhood grid to ensure that the administrator reaches the destination in the shortest time, saving the administrator's travel time as much as possible, so as to make full use of the administrator time.
  • the current grid and the candidate grid P 1 are assigned to the administrator, to a certain extent, the number of cases assigned by the administrator is limited, and the administrator is also guaranteed Handle the number of assigned cases within a certain time to achieve a reasonable distribution of the workload of the administrator.
  • FIG. 1 is a schematic flowchart of an urban grid allocation method provided by an embodiment of the present application
  • FIG. 2 is a schematic flowchart of another urban grid allocation method provided by an embodiment of the present application.
  • FIG. 3 is a functional block diagram of an urban grid distribution device provided by an embodiment of the present application.
  • FIG. 4 is a functional block diagram of an electronic terminal provided by an embodiment of the present application.
  • the word “if” as used herein may be interpreted as “when” or “when” or “in response to determination” or “in response to detection”.
  • the phrases “if determined” or “if detected (statement or event stated)” can be interpreted as “when determined” or “in response to determination” or “when detected (statement or event stated) )” or “in response to detection (statement or event stated)”.
  • An embodiment of the present application provides an urban grid allocation method.
  • FIG. 1 is a schematic flowchart of a method for allocating an urban grid provided by an embodiment of the present application. As shown in FIG. 1, the method includes the following steps:
  • the above steps of acquiring the number of cases and travel time corresponding to the grid may be performed periodically.
  • the execution cycle can be set to days, weeks, months, etc. according to demand.
  • the number of cases corresponding to the above grid may be the average value of the number of historical cases in the grid, or it may be predicted based on the number of historical cases in the grid, or the number of cases may also be the elapse of a predetermined period of time
  • the number of cases obtained by the statistical processing is not limited in this embodiment with regard to the specific form of the number of cases corresponding to the grid.
  • the number of cases corresponding to the grid may be: obtaining the total number of historical cases in the grid within a specified period before the current time, The number of cases corresponding to the unit time is calculated.
  • the specific values of the specified time period and unit time are not limited.
  • the number of cases corresponding to the acquisition grid may be: sending an acquisition request to the case statistics device, the acquisition request carrying a predetermined period of time, thereby The case statistics device counts the total number of cases in the predetermined time period based on the predetermined time, and feeds back the total number of cases obtained by the statistics.
  • the number of cases corresponding to the grid may be obtained by processing the historical data of the grid by a case prediction model constructed using a machine learning algorithm. The number of predicted cases.
  • the case prediction model needs to be established before the case prediction model is used to predict the number of cases in each grid.
  • the case prediction model is obtained by using machine learning algorithms to process labeled sample data through machine learning algorithms.
  • the machine learning algorithm used to build the case prediction model can be one of the regression algorithm, random forest decision tree algorithm, convolutional neural network learning algorithm, support vector machine, etc.
  • the machine used in this application for the case prediction model The learning algorithm is not limited.
  • the specific establishment process is: first, obtain the number of cases in each unit time included in the specified historical time of each grid to obtain training samples; select a specified number of unit time from the specified historical time as the prediction time period; Obtain the number of cases in each grid during the prediction period from the training sample, take the number of cases in each grid during the prediction period as the output sample, and use the training sample as the input sample; use the input sample and the output sample Input to the training model, and train the training model through machine learning to get the case prediction model.
  • the travel time of each neighbor grid includes all the time consumed by the administrator when passing through all areas in the grid In addition to the time, it also includes the travel time of the administrator from the current grid to the neighborhood grid. The way of obtaining the travel time for each neighborhood grid will be explained in detail below.
  • the preset range of the number of cases is configured to balance the number of cases that each administrator ultimately needs to process as much as possible, so as to effectively avoid the occurrence of too many or too few cases assigned by some administrators, resulting in After the administrator with too few assigned cases completes the work, the administrator with too many assigned cases also has a backlog of cases where more cases have not been processed, and thus provides human resource utilization to a certain extent to improve the efficiency of case processing.
  • the technical solution provided by the present application when assigning the corresponding city grid to the administrator, selects the neighborhood grid with the shortest travel time from the neighborhood grid of the current grid by using the travel time corresponding to the grid to obtain the candidate grid P 1 , and also use the preset range, when it is determined that the sum of the first case corresponding to the number of cases in the candidate grid P 1 and the number of cases in the current grid is within the preset range, the current grid The candidate grid P 1 is assigned to the administrator.
  • the condition of the shortest travel time is used to filter the neighborhood grid to ensure that the administrator reaches the destination in the shortest time, saving the administrator's travel time as much as possible, so as to make full use of the administrator time.
  • the current grid and the candidate grid P 1 are assigned to the administrator, to a certain extent, the number of cases assigned by the administrator is limited, and the administrator is also guaranteed Handle the number of assigned cases within a certain time to achieve a reasonable distribution of the workload of the administrator.
  • each city grid has at least two domain grids, that is, in the implementation step 104, the neighborhood network with the shortest travel time is selected from the neighborhood grids of the current grid In the process of the grid, you may get at least two domain grids that meet the shortest driving time. In this case, in order to ensure that the number of cases handled by the administrator in a certain period of time is as high as possible, determine which is the candidate domain network.
  • the grid time can be further determined by performing the following steps: First, first determine the number of cases in the domain grid with the shortest travel time; thus, based on the number of cases, the neighborhood grid with the shortest travel time is sorted in the specified order to obtain The case number sequence; further, based on the case number sequence, the neighborhood grid with the largest number of cases is acquired, and the neighborhood grid with the largest number of cases is determined as the candidate grid P 1 .
  • the designated order may be the order of the number of cases from small to large, or the designated order may also be the order of the number of cases from small to large.
  • the grid corresponding to the last bit in the case number sequence is determined as the candidate grid P 1 ; if the case number sequence is based on the largest to the smallest Is obtained, the grid corresponding to the first place in the case number sequence is determined as the candidate grid P 1 .
  • the preset range of the set number of cases includes an upper limit value and a lower limit value
  • the first case sum is not in the preset range includes two cases, that is, the first case sum exceeds the upper limit or the The sum of one case does not reach the lower limit corresponding to the preset range.
  • this application provides another implementation method, as shown in FIG. 2, After 106, the following steps need to be performed:
  • the number of cases and travel time corresponding to the surrounding grid adjacent to the candidate grid P i is similar to the method of obtaining the number of cases and travel time corresponding to the neighboring grid of the current grid in step 102 above
  • this embodiment will not repeat them here.
  • the central grid is the candidate grid P i obtained before, so the number of cases corresponding to the central grid before the previous realization determines whether the sum of the first cases is within the preset range, It has been learned that when it is realized this time whether the sum of the second cases is within the preset range, there is no need to perform the operation of acquiring the number of cases corresponding to the center grid again.
  • step 201 is performed to determine the candidate grid P i as the center grid.
  • the obtained candidate grid P i+1 is used as the central grid, and the above steps 202-204 are cyclically executed until the total number of cases of all the candidate grids obtained And the sum of the second cases corresponding to the current grid is within the preset range, the loop is stopped, and step 205 is executed to assign the current grid and all the obtained candidate grids to the administrator.
  • re-selecting the corresponding candidate grid may be: performing the domain grids on the neighborhood grid in the order of the travel time from small to large Sorting to obtain a time series, and determining the domain grid in the time series after the current candidate grid and corresponding to the next position adjacent to the current candidate grid as the newly selected candidate grid P 1 .
  • the domain grid with the shortest travel time is sorted in the order specified by the number of cases, and after obtaining the case number sequence, the The domain grid corresponding to the next position adjacent to the domain grid with the largest number of cases in the number sequence is used as the candidate grid, and the candidate grid is determined as the reselected candidate grid P 1 , and so on, specifying the selection Until the sum of the candidate grid P 1 and the number of cases in the current grid is less than or equal to the upper limit of the preset range, or it is determined that all the field grids in the case number sequence cannot be satisfied so that the sum of the first cases Less than or equal to the upper limit value of the preset range, based on the time series obtained in the above embodiment
  • the above-described re-selection candidate when it is judged the upper limit value is greater than the sum of the second case of a preset range, the above-described re-selection candidate can mesh P according to Embodiment 1, the user selects another candidate grid. Omitted herein for reselection in a manner other candidate grid can be found in the above-described specific re-selection candidate grid mode P 1.
  • the administrator does not need to perform related management on the grid.
  • the grid with 0 cases can be assigned to any An administrator.
  • the number of cases and the travel time corresponding to each neighborhood grid of the neighbor may include: acquiring the number of cases and the travel time corresponding to the neighborhood grid indicated as an unassigned administrator in the judgment result;
  • the neighborhood grid assigned to the administrator is not processed.
  • the implementation for obtaining the travel time corresponding to each neighborhood grid adjacent to the current grid may include: first, obtaining road information of the current grid and its neighbor grid, and then, according to The road information, obtaining the first time required for the current grid to reach any neighboring grid; and obtaining the second time required for the administrator to finish driving all roads in any neighboring grid; Therefore, the total time of the first time and the second time is obtained, and the total time is used as the travel time corresponding to the any neighboring grid.
  • the road information acquisition of the grid may be: acquiring the map information corresponding to each grid, and then extracting the road information from the obtained map information.
  • the road information may include, but is not limited to: the number of roads, the number of forks, and so on.
  • a driving time estimation model can be established in advance based on the machine learning algorithm, so that based on the preset driving time estimation model, the predicted information can be obtained based on the road information of the grid Travel time.
  • the way to obtain the first time is: first, according to the road information of the current grid and the road information of its domain grid, obtain the feasible route information from the current grid to the domain grid, and then the feasible route The information is input into the first driving time estimation model, and the first time is obtained after the first driving time estimation model is processed; for example, the way to obtain the second time is to input the road information of the neighborhood grid into the first
  • the second travel time estimation model is processed by the first travel time estimation model to obtain the second time.
  • the above first driving time estimation model and second driving time estimation model may be based on support vector machine, convolutional neural network, linear regression algorithm, Markov decision algorithm, or gradient boosting decision tree (Gradient Boosting Decision Tree, GBDT) algorithm Wait for the establishment of machine learning algorithms.
  • GDT gradient boosting decision tree
  • the process of establishing the time prediction model specifically includes: Step 1, according to the map information, obtain the connected route information of two adjacent grids as training The feasible route of the sample; Step 2: Obtain the road condition information of each road segment in the feasible route of the training sample and the actual travel time of each road segment, where the road segment is the road between two forks, or, the fork and The road between the starting point (end point), or the road between the starting point and the end; Step 3, using the deep learning algorithm model, train the road condition information of each road segment and the actual driving time of each road segment to obtain Estimated travel time model.
  • a connected road ie, road
  • a connected road ie, road
  • the historical road condition information is vectorized to obtain the corresponding feature vector
  • the obtained feature vector is input into the initial driving time estimation model, and the training and training are performed to obtain the training output value.
  • the training output value and the actual driving time value are obtained
  • the training process is ended and the first travel time estimation model is obtained.
  • the time-consuming acquisition method to reach the specified grid is similar to the above-mentioned method to acquire the travel time corresponding to each neighborhood grid adjacent to the current grid, and this embodiment will not repeat them here. .
  • the target city area is divided into multiple preset grids of a specified size, and there are multiple auto insurance case managers distributed in the target city area.
  • the allocation method of assigning these preset grids to each auto insurance case manager can be quickly and accurately calculated, and the minimum travel time can be
  • construct an objective function that minimizes the case arrival time of the car insurance case manager, and divides it with each preset grid; and, the preset grid is not divided repeatedly to multiple people; and, each car insurance case manager
  • the solution of the objective function can characterize the specific allocation method of the above-mentioned divided preset grid to the auto insurance case managers
  • the objective function and its corresponding constraint conditions can be used to construct a correlation matrix X of the auto insurance case manager and the preset grid, where the correlation matrix
  • the behavior car insurance case manager of X is listed as the preset grid. If the specified car insurance case manager is associated with the specified preset grid based on the constraint condition, the association matrix X is associated with the specified car insurance case manager and The element value corresponding to the specified preset grid is 1, otherwise, the element value is 0.
  • the constructed correlation matrix X is a 5*50 dimension matrix, and the elements in the 5*50 dimension matrix The specific value is obtained based on the set objective function and the corresponding constraints. After obtaining the 5*50 dimension matrix, the 5*50 dimension matrix can be obtained to ensure that the work efficiency of each car insurance case manager is fully The preset grid allocation method used.
  • the target candidate grid includes a plurality of grids
  • it may further be performed: counting each candidate management The total amount of cases currently to be processed by the administrator; the candidate administrator with the smallest total amount of cases to be processed is determined as the target administrator; and the designated grid is assigned to the target administrator.
  • the execution subject of the above embodiment may be an urban grid distribution device, which may be an application located in a local terminal, or may also be a plug-in or a software development kit (Software Development Kit) in an application located in a local terminal. SDK) and other functional units, which are not particularly limited in the embodiments of the present application.
  • the application may be an application program (nativeApp) installed on the terminal, or may also be a webpage program (webApp) of a browser on the terminal, which is not limited in this embodiment of the present application.
  • embodiments of the present application further provide device embodiments that implement the steps and methods in the foregoing method embodiments.
  • FIG. 3 is a functional block diagram of an urban grid distribution device provided by an embodiment of the present application.
  • the device includes an acquisition unit 31, a selection unit 32, a judgment unit 33, and a processing unit 34.
  • the obtaining unit 31 is used to obtain the number of cases corresponding to the current grid where the administrator is located, and to obtain the number of cases and travel time corresponding to each neighborhood grid adjacent to the current grid;
  • the selection unit 32 Used to filter out the neighborhood grid with the shortest travel time from the neighborhood grid of the current grid according to the travel time corresponding to each neighborhood grid to obtain a candidate grid P 1 ;
  • the judgment unit 33 It is used to determine whether the sum of the first case corresponding to the number of cases in the candidate grid P 1 and the number of cases in the current grid is within a preset range;
  • the processing unit 34 is used to determine if the sum of the first case is in all Within the preset range, the current grid and the candidate grid P 1 are allocated to the administrator.
  • the number of cases corresponding to the above grid may be the average value of the number of historical cases in the grid, or it may be predicted based on the number of historical cases in the grid, or the number of cases may also be the elapse of a predetermined period of time
  • the number of cases obtained by the statistical processing is not limited in this embodiment with regard to the specific form of the number of cases corresponding to the grid.
  • the number of cases corresponding to the grid may be: obtaining the total number of historical cases in the grid within a specified period before the current time, The number of cases corresponding to the unit time is calculated.
  • the specific values of the specified time period and unit time are not limited.
  • the number of cases corresponding to the acquisition grid may be: sending an acquisition request to the case statistics device, the acquisition request carrying a predetermined period of time, thereby The case statistics device counts the total number of cases in the predetermined time period based on the predetermined time, and feeds back the total number of cases obtained by the statistics.
  • the number of cases corresponding to the grid may be obtained by processing the historical data of the grid by a case prediction model constructed using a machine learning algorithm. The number of predicted cases.
  • the case prediction model needs to be established before the case prediction model is used to predict the number of cases in each grid.
  • the case prediction model is obtained by using machine learning algorithms to process labeled sample data through machine learning algorithms.
  • the machine learning algorithm used to build the case prediction model can be one of the regression algorithm, random forest decision tree algorithm, convolutional neural network learning algorithm, support vector machine, etc.
  • the machine used in this application for the case prediction model The learning algorithm is not limited.
  • the specific establishment process is: first, obtain the number of cases in each unit time included in the specified historical time of each grid to obtain training samples; select a specified number of unit time from the specified historical time as the prediction time period; Obtain the number of cases in each grid during the prediction period from the training sample, take the number of cases in each grid during the prediction period as the output sample, and use the training sample as the input sample; use the input sample and the output sample Input to the training model, and train the training model through machine learning to get the case prediction model.
  • the selection unit 32 performs The neighborhood grid with the shortest travel time is selected from the neighborhood grid to obtain the candidate grid P 1 , which further includes: sorting the neighborhood grid with the shortest travel time according to the number of cases in a specified order to obtain the number of cases Sequence; and, based on the case number sequence, the neighborhood grid with the largest number of cases is acquired, and the neighborhood grid with the largest number of cases is determined as the candidate grid P 1 .
  • the selection unit 32 is also used to select from each surrounding grid according to the number of cases and travel time corresponding to each surrounding grid The neighborhood grid with the shortest travel time is selected to obtain the candidate grid P i+1 ;
  • the judgment unit 33 is also used to judge whether the total number of cases of all candidate grids and the sum of the second cases of the number of cases in the current grid
  • the selection unit 32 may specifically perform the following operation: obtain the maximum number of cases in the case number sequence The candidate grid adjacent to the domain grid of is determined as the candidate grid P 1 to be reselected.
  • the selection unit 32 is further used to filter out the first designated candidate grid with the number of cases being 0 from the neighborhood grid; and/or from the surrounding grid of the central grid A second designated candidate grid with a number of cases of 0 is selected from the grid; thus, the processing unit 34 is also used to allocate the first designated candidate grid and/or the second designated candidate grid to the administrator.
  • the acquiring unit 31 performs acquiring the travel time corresponding to each neighborhood grid adjacent to the current grid, including: acquiring road information of the current grid and its neighbor grid ; And, based on the road information, obtain the first time required for the administrator to travel from the current grid to any neighborhood grid; and, obtain the administrator to complete the travel of any neighborhood grid The second time required by all roads in the road; and, obtaining the total time of the first time and the second time, and using the total time as the travel time corresponding to any neighboring grid.
  • the processing unit 34 is further used to determine the Whether multiple grids in the target geographic area have been assigned to corresponding administrators; and, if it is determined that there are multiple grids in the target geographical area that have no designated grids assigned by administrators, according to each Road information of the grid, detecting a target candidate grid that takes a predetermined time threshold to reach the specified grid; and, determining a candidate administrator corresponding to the target candidate grid; and, using the specified network The grid is assigned to the candidate administrator.
  • the processing unit 34 is further used to count each Candidate administrators, the total number of cases currently to be processed; and, the candidate administrator with the smallest total amount of current pending cases is determined as the target administrator; and, the designated grid is assigned to the target administrator .
  • FIG. 4 is a functional block diagram of an electronic terminal provided by an embodiment of the present application.
  • the electronic terminal includes a memory 41, a processor 42, and computer-readable instructions stored in the memory 41 and executable on the processor 42, and the processor 42 executes the computer When reading an instruction, the processor 42 implements any of the above urban grid allocation methods.
  • Embodiments of the present application also provide a non-volatile computer-readable storage medium on which computer-readable instructions are stored.
  • the computer reads and executes the computer-readable instructions, the computer is enabled to implement any of the above The urban grid allocation method described above.
  • the disclosed system, device, and method may be implemented in other ways.
  • the device embodiments described above are only schematic.
  • the division of the unit is only a logical function division, and there may be other divisions in actual implementation, for example, multiple units or components may be combined Or it can be integrated into another system, or some features can be ignored or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical, or other forms.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
  • the above integrated unit can be implemented in the form of hardware, or in the form of hardware plus software functional units.
  • the above integrated unit implemented in the form of a software functional unit may be stored in a computer-readable storage medium.
  • the above software functional unit is stored in a storage medium, and includes several instructions to enable a computer device (which may be a personal computer, server, or network device, etc.) or processor (Processor) to perform the methods described in the embodiments of the present application Partial steps.
  • the foregoing storage media include: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other media that can store program code .

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Tourism & Hospitality (AREA)
  • Strategic Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Development Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Educational Administration (AREA)
  • Physics & Mathematics (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Navigation (AREA)

Abstract

一种城市网格分配方法、装置及电子终端,涉及数据处理技术领域。通过获取管理员所在当前网格对应的案件数,以及与所述当前网格相邻的每个邻域网格对应的案件数和行驶时间(102);从而,根据所述每个邻域网格对应的行驶时间,从所述当前网格的邻域网格中筛选出行驶时间最短的邻域网格,得到候选网格P1(104);进而,判断所述候选网格P1的案件数与当前网格的案件数对应的第一案件总和是否在预设范围内(106);若判断出所述第一案件总和在所述预设范围内,将所述当前网格与所述候选网格P1分配给所述管理员(108)。上述城市网格分配方法能够合理分配管理员的工作量。

Description

城市网格分配方法、装置及电子终端
本申请以2018年12月31日提交的申请号为201811650804.0,名称为“城市网格分配方法、装置及电子终端”的中国发明专利申请为基础,并要求其优先权。
【技术领域】
本申请涉及数据处理技术领域,尤其涉及城市网格分配方法、装置及电子终端。
【背景技术】
城市管理对于城市的经济发展以及人们的生活都有着重要的影响,其是城市发展的客观要求。其中,为了便于对城市区域进行管理,现有技术提出了网格化城市管理,该网格化城市管理方式是将管理的目标地理区域划分成多个城市网格,然后对这些城市网格进行管理。
其中,对于城市管理主要是由管理员执行管理工作的,因此,如何为管理员分配管理区域,以保证为管理员分配工作的合理性是目前亟待解决的问题。
【发明内容】
有鉴于此,本申请实施例提供了一种城市网格分配方法、装置及电子终端,能够合理分配管理员的工作量。
第一方面,本申请实施例提供了一种城市网格分配方法,所述城市网格分配方法包括:
获取管理员所在当前网格对应的案件数,以及,获取与所述当前网格相邻的每个邻域网格对应的案件数和行驶时间;
根据所述每个邻域网格对应的行驶时间,从所述当前网格的邻域网格中筛选出行驶时间最短的邻域网格,得到候选网格P 1
判断所述候选网格P 1的案件数与当前网格的案件数对应的第一案件总和是否在预设范围内;
若判断出所述第一案件总和在所述预设范围内,将所述当前网格与所述候选网格P 1分配 给所述管理员。
第二方面,本申请实施例提供了一种城市网格分配装置,所述城市网格分配装置包括:
获取单元,用于获取管理员所在当前网格对应的案件数,以及,与所述当前网格相邻的每个邻域网格对应的案件数和行驶时间;
选择单元,用于根据所述每个邻域网格对应的行驶时间,从所述当前网格的邻域网格中筛选出行驶时间最短的邻域网格,得到候选网格P 1
判断单元,用于判断所述候选网格P 1的案件数与当前网格的案件数对应的第一案件总和是否在预设范围内;
处理单元,用于若判断出所述第一案件总和在所述预设范围内,将所述当前网格与所述候选网格P 1分配给所述管理员。
第三方面,本申请实施例提供了一种电子终端,包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机可读指令,所述处理器执行所述计算机可读指令时,所述处理器实现第一方面中任一项所述的城市网格分配方法。
第四方面,本申请实施例提供了一种非易失性计算机可读存储介质,其上存储有计算机可读指令,当计算机读取并执行所述计算机可读指令时,使得计算机实现上述任一项所述的城市网格分配方法。
上述技术方案具有如下有益效果:
本申请提供的技术方案,在为管理员分配对应的城市网格时,通过利用网格对应的行驶时间,从所述当前网格的邻域网格中筛选出行驶时间最短的邻域网格,得到候选网格P 1,并且还利用预设范围,在确定候选网格P 1的案件数与当前网格的案件数对应的第一案件总和在预设范围的情况下,将当前网格与候选网格P 1分配给管理员。其中,利用行驶时间最短的这一条件对邻域网格进行筛选,保证管理员在最短时间内到达目的地,尽可能节省管理员的行驶时间,从而充分利用管理员的时间。另外,通过在判断出第一案件总和在预设范围内,将当前网格与候选网格P 1分配给管理员,一定程度上限定了管理员分配得到的案件数,也在保证管理员在一定时间处理分配的案件数,实现合理分配管理员的工作量。
【附图说明】
为了更清楚地说明本申请实施例的技术方案,下面将对实施例中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本申请的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其它的附图。
图1是本申请实施例所提供的一种城市网格分配方法的流程示意图;
图2是本申请实施例所提供的另一种城市网格分配方法的流程示意图;
图3是本申请实施例所提供的一种城市网格分配装置的功能框图;
图4是本申请实施例所提供的一种电子终端的功能框图。
【具体实施方式】
为了更好的理解本申请的技术方案,下面结合附图对本申请实施例进行详细描述。
应当明确,所描述的实施例仅仅是本申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其它实施例,都属于本申请保护的范围。
在本申请实施例中使用的术语是仅仅出于描述特定实施例的目的,而非旨在限制本申请。在本申请实施例和所附权利要求书中所使用的单数形式的“一种”、“所述”和“该”也旨在包括多数形式,除非上下文清楚地表示其他含义。
应当理解,本文中使用的术语“和/或”仅仅是一种描述关联对象的关联关系,表示可以存在三种关系,例如,A和/或B,可以表示:单独存在A,同时存在A和B,单独存在B这三种情况。另外,本文中字符“/”,一般表示前后关联对象是一种“或”的关系。
取决于语境,如在此所使用的词语“如果”可以被解释成为“在……时”或“当……时”或“响应于确定”或“响应于检测”。类似地,取决于语境,短语“如果确定”或“如果检测(陈述的条件或事件)”可以被解释成为“当确定时”或“响应于确定”或“当检测(陈述的条件或事件)时”或“响应于检测(陈述的条件或事件)”。
本申请实施例给出一种城市网格分配方法。
具体的,请参考图1,其为本申请实施例所提供的城市网格分配方法的流程示意图,如图1所示,该方法包括以下步骤:
102、获取管理员所在当前网格对应的案件数,以及,获取与所述当前网格相邻的每个邻域网格对应的案件数和行驶时间。
这里需要说明的是,上述对于网格对应的案件数和行驶时间的获取步骤可以是周期性执行。其中,该执行的周期可以根据需求设置几天,几周,几个月等。
其中,上述网格对应的案件数可以是该网格历史案件数的平均值,或者,为基于该网格历史案件数预测得到的,又或者,该案件数还可以是预定时间段内的经过统计处理得 到的案件数,本实施例对于上述网格对应的案件数的具体形式不做限定。
例如,若上述网格对应的案件数为该网格历史案件数的平均值时,则获取网格对应的案件数可以为:获取网格在当前时间之前的指定时段内的历史总案件数,计算单位时间内对应的案件数,其中,本实施例对于指定时段和单位时间的具体值不作限定。又例如,若案件数是预定时间段内的经过统计处理得到的案件数时,则获取网格对应的案件数可以为:向案件统计设备发送获取请求,该获取请求中携带预定时间段,从而该案件统计设备基于该预定时间,统计该预定时间段内案件总量,并反馈该统计得到的案件总量。又例如,若案件数是基于网格历史案件数预测得到的时,则获取网格对应的案件数可以为:利用机器学习算法构建的案件预测模型对该网格的历史数据进行处理,从而得到预测的案件数。
这里需要说明的是,在利用案件预测模型预测每个网格的案件数之前需要先建立该案件预测模型。该案件预测模型是利用机器学习算法,经过对标记样本数据进行机器学习算法处理得到的。其中,建立该案件预测模型所使用的机器学习算法可以为回归算法、随机森林决策树算法、卷积神经网络学习算法、支持向量机等算法中的一种,本申请对于案件预测模型使用的机器学习算法不做限定。
具体建立过程为:首先,获取每个网格在指定历史时间内包括的每个单位时间内的案件数,得到训练样本;从指定历史时间中选择指定数量的单位时间,以作为预测时间段;从训练样本中获取每个网格在预测时间段内发生的案件数,以每个网格在预测时间段内发生的案件数作为输出样本,以训练样本作为输入样本;将输入样本以及输出样本输入到训练模型,通过机器学习对训练模型进行训练,得到案件预测模型。
其中,考虑到管理员由当前所在的当前网格到达任一邻域网格也需要一定时间,因此,每个邻域网格的行驶时间除了包括管理员经过网格内所有区域时所消耗的时间之外,还包括管理员由当前网格到达该邻域网格的行驶时间。针对于每个邻域网格的行驶时间的获取方式将在下文中具体阐述。
104、根据所述每个邻域网格对应的行驶时间,从所述当前网格的邻域网格中筛选出行驶时间最短的邻域网格,得到候选网格P 1
106、判断所述候选网格P 1的案件数与当前网格的案件数对应的第一案件总和是否在预设范围内。
108、若判断出所述第一案件总和在所述预设范围内,将所述当前网格与所述候选网格P 1分配给所述管理员。
这里需要说明的是,配置案件数目的预设范围是为了尽可能均衡每个管理员最终分 配到的需要处理案件数量,从而有效避免出现某些管理员分配的案件数量过多或过少,造成分配案件数量过少的管理员完成工作后,分配案件数量过多的管理员还积压较多的案件未处理的情况的出现,进而在一定程度上提供人力资源利用率,提高案件处理效率。
本申请提供的技术方案,在为管理员分配对应的城市网格时,通过利用网格对应的行驶时间,从所述当前网格的邻域网格中筛选出行驶时间最短的邻域网格,得到候选网格P 1,并且还利用预设范围,在确定候选网格P 1的案件数与当前网格的案件数对应的第一案件总和在预设范围的情况下,将当前网格与候选网格P 1分配给管理员。其中,利用行驶时间最短的这一条件对邻域网格进行筛选,保证管理员在最短时间内到达目的地,尽可能节省管理员的行驶时间,从而充分利用管理员的时间。另外,通过在判断出第一案件总和在预设范围内,将当前网格与候选网格P 1分配给管理员,一定程度上限定了管理员分配得到的案件数,也在保证管理员在一定时间处理分配的案件数,实现合理分配管理员的工作量。
在本申请实施例中,每一个城市网格都有至少两个领域网格,也就是说,在实现步骤104从所述当前网格的邻域网格中筛选出行驶时间最短的邻域网格的过程中,可能得到至少两个满足行驶时间最短的领域网格的情况,针对于这种情况,为了尽可能保证管理员在一定时间内处理的案件数高,在确定哪个为候选领域网格时,可以进一步通过执行以下步骤来确定:首先,先确定这些行驶时间最短的领域网格的案件数;从而,基于案件数量,将行驶时间最短的邻域网格按照指定顺序进行排序,得到案件数序列;进而,根据所述案件数序列,获取案件数最多的邻域网格,并将该案件数最多的邻域网格确定为候选网格P 1。具体的,该指定顺序可以是案件数量从小到大的顺序,或者,该指定顺序还可以为案件数量从小到大的顺序。若该案件数序列是基于从小到大的排序得到的,则将该案件数序列中排在最后一位对应的网格确定为候选网格P 1;若该案件数序列是基于从大到小的排序得到的,则将该案件数序列中排在第一位对应的网格确定为候选网格P 1
进一步的,设定的案件数的预设范围包括上限值和下限值,第一案件总和不在预设范围包括两种情况,即第一案件总和超过预设范围对应的上限值或者第一案件总和未达到预设范围对应的下限值。当第一案件总和未达到预设范围的下限值时(即第一案件总和小于预设范围的下限值)时,本申请提供了另一种实现方式,如图2所示,在步骤106之后,还需要执行以下步骤:
201、确定候选网格P i为中心网格,其中,i为大于或等于1的正整数,当i=1时,该中心网格为所述候选网格P 1
202、获取与所述中心网格相邻的每个周围网格对应的案件数和行驶时间,其中,所述中心网格的领域网格不包括候选网格P i-1,其中,当i=1时,候选网格P i-1为所述当前网格。
这里需要说明的是,候选网格P i相邻的周围网格对应的案件数和行驶时间获取方式与前述步骤102中当前网格的邻域网格对应的案件数和行驶时间的获取方式相似,对于候选网格P i相邻的周围网格对应的案件数和行驶时间的具体获取过程,本实施例在此不再赘述。
另外,还需要说明一下的是,中心网格为之前得到的候选网格P i,因此该中心网格对应的案件数在经过上一次实现判断第一案件总和是否在预设范围内之前,就已经获知,在实现本次判断第二案件总和是否在所述预设范围内时,不需要再次执行获取该中心网格对应的案件数的操作。
203、根据所述每个周围网格对应的案件数和行驶时间,从所述每个周围网格中筛选出行驶时间最短的邻域网格,得到候选网格P i+1
204、判断所有候选网格的案件总量以及当前网格的案件数的第二案件总和是否在所述预设范围内。
205、若判断出所述第二案件总和在所述预设范围内,将所述当前网格以及得到的所有候选网格分配给所述管理员。
若判断出所述第二案件总和小于在所述预设范围的下限值,令i=i+1,执行步骤201确定候选网格P i为中心网格的步骤。
如果得到的第二案件总和仍然小于预设范围的下限值,表明目前确定出的所有候选网格以及当前网格的案件数量相对较少,不能相对充分利用人力资源,因此,当得到的第二案件总和仍然小于预设范围的下限值时,则以得到的候选网格P i+1为中心网格,循环执行上述步骤202-步骤204,直到得到的所有候选网格的案件总量以及当前网格对应的第二案件总和在该预设范围内,停止循环,并执行步骤205将当前网格以及得到的所有候选网格分配给该管理员。
另外,当判断出案件总和大于所述预设范围的上限值时,为了保证管理员在一定时间内能够处理所分配的网格内的案件,需要重新选择对应的候选网格,以调整得到案件总和,使得得到在案件总和在预设范围。具体的,当上述步骤中得到的行驶时间最短的领域网格只有一个时,重新选择对应的候选网格可以为:将这些领域网格,按照行驶时间从小到大的顺序对邻域网格进行排序,得到时间序列,将该时间序列中在当前候选网格之后且 与该当前候选网格相邻下一位置对应的领域网格,确定为重新选择的候选网格P 1
可选的是,结合上述步骤104从所述当前网格的邻域网格中筛选出行驶时间最短的邻域网格的过程中,可能得到至少两个满足行驶时间最短的领域网格这一具体场景,当判断出第一案件总和大于所述预设范围的上限值时,在对这些行驶时间最短的领域网格按照案件数量指定顺序进行排序,得到案件数序列之后,获取在该案件数序列中与案件数最多的领域网格相邻下一位置对应的领域网格作为备选网格,并将该备选网格确定为重新选择的候选网格P 1,如此类推,指定选择出的候选网格P 1与当前网格的案件数之和小于或等于该预设范围上限值为止,或者,确定出案件数序列中所有的领域网格的都无法满足使得第一案件总和小于或等于预设范围的上限值,基于上述实施例中得到的时间序列,将该时间序列中在当前候选网格之后且与该当前候选网格相邻下一位置对应的领域网格,确定为重新选择的候选网格P 1
这里需要说明的是,在上述实施例中,当判断出第二案件总和大于预设范围的上限值时,可以根据上述重新选择候选网格P 1的方式,重新选择其他候选网格。这里针对于其他候选网格的重新选择方式不再赘述,具体可参看上述重新选择候选网格P 1的方式。
在一具体实施例中,如果某个网格内的案件数为0时,管理员不需要对该网格执行相关的管理,这种情况下,可以该案件数为0的网格分配给任何一个管理员。为了在一定程度上提高网格分配的效率,在得到对应邻域网格的案件数之后,还可以执行:从当前网格的邻域网格中筛选出案件数目为0的第一指定候选网格;和/或,从每个中心网格的周围网格中筛选出案件数目为0的第二指定候选网格;然后,得到的第一指定候选网格和第二指定候选网格也分配给该管理员。
可选的是,为了便于管理,同时保证每个城市网格中的案件能够及时处理,通过会为目标地理区域内的多个网格分配给多个管理员,而为了避免一个网格重复分配给多个管理员,在获取与每个邻域网格对应的案件数和行驶时间之前,需要先对这些邻域网格进行筛选,选择出未分配管理员的邻域网格,然后,再从将这些未分配管理员的邻域网格,分配给对应的管理员。基于此,在执行上述提供的城市网格分配方法前,还需要:判断所述当前网格的邻域网格是否已分配管理员,得到判断结果;其中,执行获取与所述当前网格相邻的每个邻域网格对应的案件数和行驶时间可以包括:获取所述判断结果中指示为未分配管理员的邻域网格对应的案件数和行驶时间;而对于判断结果中指示已分配管理员的邻域网格不做处理。
本申请中,针对于获取与所述当前网格相邻的每个邻域网格对应的行驶时间的实现可以包括:首先,获取当前网格及其邻域网格的道路信息,然后,根据所述道路信息,获 取所述当前网格到任一邻域网格所需要的第一时间;以及,获取管理员行驶完所述任一邻域网格内所有道路所需要的第二时间;从而,获取所述第一时间和所述第二时间的总时间,并将所述总时间作为所述任一邻域网格对应的行驶时间。其中,网格的道路信息获取可以为:获取每个网格对应的地图信息,然后,从得到的地图信息中提取出道路信息。其中,道路信息可以包括但不限于:道路数量、岔路口数量等。
具体的,在实现获取第一时间和第二时间时,可以基于机器学习算法,预先建立一个行驶时间预估模型,从而基于预设的行驶时间预估模型,根据网格的道路信息,获取预计行驶时间。例如,获取第一时间的方式为:首先,根据当前网格的道路信息以及其领域网格的道路信息,获取由该当前网格行驶到该领域网格的可行路线信息,然后将该可行路线信息输入到第一行驶时间预估模型中,经过该第一行驶时间预估模型处理后得到第一时间;例如,获取第二时间的方式为,将该邻域网格的道路信息输入到第二行驶时间预估模型,经过该第一行驶时间预估模型处理后得到第二时间。
上述第一行驶时间预估模型和第二行驶时间预估模型可以基于支持向量机、卷积神经网络、线性回归算法、马尔可夫决策算法或者梯度提升决策树(Gradient Boosting Decision Tree,GBDT)算法等机器学习算法建立。具体的,以第一行驶时间预估模型的建立为例,具体说明该时间预估模型建立的过程包括:步骤1,根据地图信息,获取两个相邻网格的连通路线信息,以作为训练样本的可行路线;步骤2,获取训练样本的可行路线中的每个路段的道路路况信息以及每个路段的实际行驶时间,其中,路段为两个岔路口之间的道路,或者,岔路口与起始点(终点)之间的道路,或者,起始点与终之间的道路;步骤3,利用深度学习算法模型,对每个路段的道路路况信息以及每个路段的实际行驶时间进行训练,得到行驶时间预估模型。
举例说明,假设两个相邻网格之间的一条连通道路(即可行道路)包括3个路段,分别获取每个路段15天的历史道路路况信息和实际行驶时间,将每个路段15天的历史道路路况信息作向量化处理,得到对应的特征向量,将得到特征向量输入到初始行驶时间预估模型中,进行学习训练,得到训练输出值,当得到训练输出值与实际行驶时间值之间的误差小于一定值时,结束该训练过程,得到该第一行驶时间预估模型。
进一步的,为了保证目标地理区域内的所有网格都有对应的管理员,为实现此目的,在按照上述实施例提供的方法为每一个管理员分配对应的网格之后,还需要执行:判断所述目标地理区域内的多个网格是否都已分配给对应的管理员;若判断出所述目标地理区域内的多个网格存在未分配管理员的指定网格,根据每个网格的道路信息,检测到达所述指定网格耗时满足预设时间阈值的目标备选网格;确定所述目标备选网格对应的候选管理员; 将所述指定网格分配给所述候选管理员。
这里需要说明的是,到达所述指定网格耗时的获取方式与上述获取与所述当前网格相邻的每个邻域网格对应的行驶时间的方式相似,本实施例在此不在赘述。
下面结合一个具体场景描述上述技术方案。在该具体场景中目标城市区域被划分成多个指定大小的预设网格,且该目标城市区域分布有多个车险案件管理员。在保证让每个车险案件管理员的工作效率得到充分的利用的前提下,能够快速地准确地计算出将这些预设网格分配每个车险案件管理员的分配方式,可以以行驶时间最小为原则,构建最小化车险案件管理员的案件到达时间的目标函数,并且以每个预设网格都划分;以及,预设网格不重复划分给多个人;以及,每个车险案件管理员均有最小覆盖案件数和最大覆盖案件数;这3点作为约束条件,构建该目标函数的约束函数,从而,基于该约束函数,利用启发式算法或遗传算法对该目标函数进行求解,其得到的该目标函数的解能够表征上述划分的预设网格分配给对个车险案件管理员的具体分配方式。
具体的,在启发式算法或遗传算法对该目标函数进行求解时,可以将该目标函数和其对应的约束条件,构建一个车险案件管理员和预设网格的关联矩阵X,其中该关联矩阵X的行为车险案件管理员,列为预设网格,如果基于该约束条件,得出指定车险案件管理员与指定预设网格关联,则该关联矩阵X中与该指定车险案件管理员和指定预设网格对应的元素值为1,否则,该元素值为0。例如,目标城市区域中分布的车险案件管理员为5个,预设网格有50个,则构建的关联矩阵X是一个5*50维度的矩阵,该5*50维度的矩阵中的元素的具体数值基于设定目标函数和对应的约束条件得到,得到该5*50维度的矩阵后,通过该5*50维度的矩阵即可得到能够保证让每个车险案件管理员的工作效率得到充分的利用的预设网格分配方式。
结合上述实现方式,可选的是,当目标备选网格包括多个网格时,在所述确定所述目标备选网格对应的候选管理员之后,还可以执行:统计每个候选管理员当前待处理的案件总量;将所述当前待处理的案件总量最小的候选管理员确定为目标管理员;将所述指定网格分配给所述目标管理员。
需要说明的是,上述实施例的执行主体可以为城市网格分配装置,该装置可以位于本地终端的应用,或者还可以为位于本地终端的应用中的插件或软件开发工具包(Software Development Kit,SDK)等功能单元,本申请实施例对此不进行特别限定。
可以理解的是,所述应用可以是安装在终端上的应用程序(nativeApp),或者还可以是终端上的浏览器的一个网页程序(webApp),本申请实施例对此不进行限定。
基于上述实施例所提供的城市网格分配方法,本申请实施例进一步给出实现上述方法实施例中各步骤及方法的装置实施例。
请参考图3,其为本申请实施例所提供的城市网格分配装置的功能方块图。如图3所示,该装置包括:获取单元31,选择单元32,判断单元33,处理单元34。其中,获取单元31,用于获取管理员所在当前网格对应的案件数,以及,获取与所述当前网格相邻的每个邻域网格对应的案件数和行驶时间;选择单元32,用于根据所述每个邻域网格对应的行驶时间,从所述当前网格的邻域网格中筛选出行驶时间最短的邻域网格,得到候选网格P 1;判断单元33,用于判断所述候选网格P 1的案件数与当前网格的案件数对应的第一案件总和是否在预设范围内;处理单元34,用于若判断出所述第一案件总和在所述预设范围内,将所述当前网格与所述候选网格P 1分配给所述管理员。
其中,上述网格对应的案件数可以是该网格历史案件数的平均值,或者,为基于该网格历史案件数预测得到的,又或者,该案件数还可以是预定时间段内的经过统计处理得到的案件数,本实施例对于上述网格对应的案件数的具体形式不做限定。
例如,若上述网格对应的案件数为该网格历史案件数的平均值时,则获取网格对应的案件数可以为:获取网格在当前时间之前的指定时段内的历史总案件数,计算单位时间内对应的案件数,其中,本实施例对于指定时段和单位时间的具体值不作限定。又例如,若案件数是预定时间段内的经过统计处理得到的案件数时,则获取网格对应的案件数可以为:向案件统计设备发送获取请求,该获取请求中携带预定时间段,从而该案件统计设备基于该预定时间,统计该预定时间段内案件总量,并反馈该统计得到的案件总量。又例如,若案件数是基于网格历史案件数预测得到的时,则获取网格对应的案件数可以为:利用机器学习算法构建的案件预测模型对该网格的历史数据进行处理,从而得到预测的案件数。
这里需要说明的是,在利用案件预测模型预测每个网格的案件数之前需要先建立该案件预测模型。该案件预测模型是利用机器学习算法,经过对标记样本数据进行机器学习算法处理得到的。其中,建立该案件预测模型所使用的机器学习算法可以为回归算法、随机森林决策树算法、卷积神经网络学习算法、支持向量机等算法中的一种,本申请对于案件预测模型使用的机器学习算法不做限定。
具体建立过程为:首先,获取每个网格在指定历史时间内包括的每个单位时间内的案件数,得到训练样本;从指定历史时间中选择指定数量的单位时间,以作为预测时间段;从训练样本中获取每个网格在预测时间段内发生的案件数,以每个网格在预测时间段内发 生的案件数作为输出样本,以训练样本作为输入样本;将输入样本以及输出样本输入到训练模型,通过机器学习对训练模型进行训练,得到案件预测模型。
本申请中,可选的是,如果当前网格的邻域网格中筛选出行驶时间最短的邻域网格包括至少两个邻域网格,该选择单元32执行从所述当前网格的邻域网格中筛选出行驶时间最短的邻域网格,得到候选网格P 1,还包括:基于案件数量,将所述行驶时间最短的邻域网格按照指定顺序进行排序,得到案件数序列;以及,根据所述案件数序列,获取案件数最多的邻域网格,并将该案件数最多的邻域网格确定为候选网格P 1
本申请中,可选的是,该城市网格分类装置还包括确定单元(图3中未示出)当判断出所述第一案件总和小于所述预设范围的下限值时,该确定单元用于确定候选网格P i为中心网格,其中,i为大于或等于1的正整数,当i=1时,所述中心网格为所述候选网格P 1;该获取单元31还用于获取与所述中心网格相邻的每个周围网格对应的案件数和行驶时间,其中,所述中心网格的领域网格不包括候选网格P i-1,其中,当i=1时,候选网格P i-1为所述当前网格;该选择单元32还用于根据所述每个周围网格对应的案件数和行驶时间,从所述每个周围网格中筛选出行驶时间最短的邻域网格,得到候选网格P i+1;该判断单元33还用于判断所有候选网格的案件总量以及当前网格的案件数的第二案件总和是否在所述预设范围内;若判断出所述第二案件总和在所述预设范围内,处理单元34用于将所述当前网格、所述所有候选网格分配给所述管理员;若判断出所述第二案件总和小于在所述预设范围的下限值,处理单元34用于令i=i+1,执行所述确定候选网格P i为中心网格的步骤。
本申请中,可选的是,当判断出第一案件总和大于所述预设范围的上限值时,选择单元32具体还可以执行以下操作:获取在所述案件数序列中与案件数最多的领域网格相邻的备选网格;将所述备选网格确定为重新选择的候选网格P 1
本申请中,可选的是,选择单元32还用于从所述邻域网格中筛选出案件数目为0的第一指定候选网格;和/或,从所述中心网格的周围网格中筛选出案件数目为0的第二指定候选网格;从而,处理单元34还用于将将所述第一指定候选网格和/或第二指定候选网格分配给所述管理员。
本申请中,可选的是,获取单元31在执行获取与所述当前网格相邻的每个邻域网格对应的行驶时间,包括:获取当前网格及其邻域网格的道路信息;以及,根据所述道路信息,获取所述管理员从所在的当前网格行驶到任一邻域网格所需要的第一时间;以及,获取管理员行驶完所述任一邻域网格内所有道路所需要的第二时间;以及,获取所述第一时 间和所述第二时间的总时间,并将所述总时间作为所述任一邻域网格对应的行驶时间。
本申请中,可选的是,当将目标地理区域内的多个网格分配给多个管理员时,在为每个管理员分配对应的网格之后,处理单元34还用于判断所述目标地理区域内的多个网格是否都已分配给对应的管理员;以及,若判断出所述目标地理区域内的多个网格存在未分配管理员的指定网格,根据所述每个网格的道路信息,检测到达所述指定网格耗时满足预设时间阈值的目标备选网格;以及,确定所述目标备选网格对应的候选管理员;以及,将所述指定网格分配给所述候选管理员。
本申请中,可选的是,当所述目标备选网格包括多个网格时,在所述确定所述目标备选网格对应的候选管理员之后,处理单元34还用于统计每个候选管理员当前待处理的案件总量;以及,将所述当前待处理的案件总量最小的候选管理员确定为目标管理员;以及,将所述指定网格分配给所述目标管理员。
由于本实施例中的各单元能够执行上述所述的方法,本实施例未详细描述的部分,可参考对上述城市网格分配方法的相关说明。
基于上述实施例所提供的城市网格分配方法,本申请实施例进一步给出实现上述方法实施例中各步骤及方法的电子终端。请参考图4,其为本申请实施例所提供的电子终端的功能方块图。如图4所示,该电子终端包括存储器41、处理器42以及存储在所述存储器41中并可在所述处理器42上运行的计算机可读指令,所述处理器42执行所述计算机可读指令时,所述处理器42实现上述任一项所述的城市网格分配方法。
本申请实施例还提供了一种非易失性计算机可读存储介质,其上存储有计算机可读指令,当计算机读取并执行所述计算机可读指令时,使得计算机实现上述任一项所述的城市网格分配方法。
所属领域的技术人员可以清楚地了解到,为描述的方便和简洁,上述描述的系统,装置和单元的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。
在本申请所提供的几个实施例中,应该理解到,所揭露的系统,装置和方法,可以通过其它的方式实现。例如,以上所描述的装置实施例仅仅是示意性的,例如,所述单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如,多个单元或组件可以结合或者可以集成到另一个系统,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口,装置或单元的间接耦合或通信连接,可以是电性,机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示 的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外,在本申请各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现,也可以采用硬件加软件功能单元的形式实现。
上述以软件功能单元的形式实现的集成的单元,可以存储在一个计算机可读取存储介质中。上述软件功能单元存储在一个存储介质中,包括若干指令用以使得一台计算机装置(可以是个人计算机,服务器,或者网络装置等)或处理器(Processor)执行本申请各个实施例所述方法的部分步骤。而前述的存储介质包括:U盘、移动硬盘、只读存储器(Read-Only Memory,ROM)、随机存取存储器(Random Access Memory,RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述仅为本申请的较佳实施例而已,并不用以限制本申请,凡在本申请的精神和原则之内,所做的任何修改、等同替换、改进等,均应包含在本申请保护的范围之内。

Claims (20)

  1. 一种城市网格分配方法,其特征在于,所述城市网格分配方法包括:
    获取管理员所在当前网格对应的案件数,以及,获取与所述当前网格相邻的每个邻域网格对应的案件数和行驶时间;
    根据所述每个邻域网格对应的行驶时间,从所述当前网格的邻域网格中筛选出行驶时间最短的邻域网格,得到候选网格P 1
    判断所述候选网格P 1的案件数与当前网格的案件数的第一案件总和是否在预设范围内;
    若判断出所述第一案件总和在所述预设范围内,将所述当前网格与所述候选网格P 1分配给所述管理员。
  2. 根据权利要求1所述的方法,其特征在于,如果当前网格的邻域网格中筛选出行驶时间最短的邻域网格包括至少两个邻域网格,所述从所述当前网格的邻域网格中筛选出行驶时间最短的邻域网格,得到候选网格P 1,包括:
    基于案件数量,将所述行驶时间最短的邻域网格按照指定顺序进行排序,得到案件数序列;
    根据所述案件数序列,获取案件数最多的邻域网格,并将该案件数最多的邻域网格确定为候选网格P 1
  3. 根据权利要求1所述的方法,其特征在于,当判断出所述第一案件总和小于所述预设范围的下限值时,所述方法还包括:
    确定候选网格P i为中心网格,其中,i为大于或等于1的正整数,当i=1时,所述中心网格为所述候选网格P 1
    获取与所述中心网格相邻的每个周围网格对应的案件数和行驶时间,其中,所述中心网格的领域网格不包括候选网格P i-1,其中,当i=1时,候选网格P i-1为所述当前网格;
    根据所述每个周围网格对应的案件数和行驶时间,从所述每个周围网格中筛选出行驶时间最短的邻域网格,得到候选网格P i+1
    判断所有候选网格的案件总量以及当前网格的案件数的第二案件总和是否在所述预设范围内;
    若判断出所述第二案件总和在所述预设范围内,将所述当前网格、所述所有候选网格分配给所述管理员;
    若判断出所述第二案件总和小于所述预设范围的下限值,令i=i+1,执行所述确定候选网格P i为中心网格的步骤。
  4. 根据权利要求2所述的方法,其特征在于,当判断出第一案件总和大于所述预设范围的上限值时,所述方法还包括:
    获取在所述案件数序列中与案件数最多的领域网格相邻的备选网格;
    将所述备选网格确定为重新选择的候选网格P 1
  5. 根据权利要求3所述的方法,其特征在于,所述方法还包括:
    从所述邻域网格中筛选出案件数目为0的第一指定候选网格;和/或,从所述中心网格的周围网格中筛选出案件数目为0的第二指定候选网格;
    将所述第一指定候选网格和/或第二指定候选网格分配给所述管理员。
  6. 根据权利要求1所述的方法,其特征在于,获取与所述当前网格相邻的每个邻域网格对应的行驶时间包括:
    获取所述当前网格及其邻域网格的道路信息;
    根据所述道路信息,获取所述管理员从所在的所述当前网格行使到任一邻域网格所需要的第一时间;以及,
    获取所述管理员行驶完所述任一邻域网格内所有道路所需要的第二时间;
    获取所述第一时间和所述第二时间的总时间,并将所述总时间作为所述任一邻域网格对应的行驶时间。
  7. 根据权利要求1所述的方法,其特征在于,获取每个网格对应的案件数,包括:
    获取每个网格在预测时间段之前的指定时段内案件数的历史数据;
    根据所述指定时段内发生的历史案件数,通过预先建立的案件预测模型,对所述每个网格在所述预测时间段内的发生案件数进行预测。
  8. 根据权利要求7所述的方法,其特征在于,在根据所述指定时段内发生的历史案件数,通过预先建立的案件预测模型,对所述每个网格在所述预测时间段内的发生案件数进行预测之前,所述方法还包括:
    获取每个网格在指定历史时间内包括的每个单位时间内的案件数,得到训练样本;
    从所述指定历史时间中选择指定数量的单位时间,以作为预测时间段;
    从所述训练样本中获取每个网格在所述预测时间段内发生的案件数,以每个网格在所述预测时间段内发生的案件数作为输出样本,以所述训练样本作为输入样本;
    将所述输入样本以及所述输出样本输入到训练模型,通过机器学习对所述训练模型进行训练,得到所述案件预测模型。
  9. 一种城市网格分配装置,其特征在于,所述城市网格分配装置包括:
    获取单元,用于获取管理员所在当前网格对应的案件数,以及,获取与所述当前网格相邻的每个邻域网格对应的案件数和行驶时间;
    选择单元,用于根据所述每个邻域网格对应的行驶时间,从所述当前网格的邻域网格中筛选出行驶时间最短的邻域网格,得到候选网格P 1
    判断单元,用于判断所述候选网格P 1的案件数与当前网格的案件数对应的第一案件总和是否在预设范围内;
    处理单元,用于若判断出所述第一案件总和在所述预设范围内,将所述当前网格与所述候选网格P 1分配给所述管理员。
  10. 根据权利要求9所述的装置,其特征在于,如果当前网格的邻域网格中筛选出行驶时间最短的邻域网格包括至少两个邻域网格,所述处理单元还具体用于:
    基于案件数量,将所述行驶时间最短的邻域网格按照指定顺序进行排序,得到案件数序列;
    根据所述案件数序列,获取案件数最多的邻域网格,并将该案件数最多的邻域网格确定为候选网格P 1
  11. 根据权利要求9所述的装置,其特征在于,所述处理单元还具体用于:
    确定候选网格P i为中心网格,其中,i为大于或等于1的正整数,当i=1时,所述中心网格为所述候选网格P 1
    获取与所述中心网格相邻的每个周围网格对应的案件数和行驶时间,其中,所述中心网格的领域网格不包括候选网格P i-1,其中,当i=1时,候选网格P i-1为所述当前网格;
    根据所述每个周围网格对应的案件数和行驶时间,从所述每个周围网格中筛选出行驶时间最短的邻域网格,得到候选网格P i+1
    判断所有候选网格的案件总量以及当前网格的案件数的第二案件总和是否在所述预设范围内;
    若判断出所述第二案件总和在所述预设范围内,将所述当前网格、所述所有候选网格分配给所述管理员;
    若判断出所述第二案件总和小于所述预设范围的下限值,令i=i+1,执行所述确定候选网格P i为中心网格的步骤。
  12. 根据权利要求10所述的装置,其特征在于,所述判断单元还具体用于:
    获取在所述案件数序列中与案件数最多的领域网格相邻的备选网格;
    将所述备选网格确定为重新选择的候选网格P 1
  13. 一种电子终端,包括存储器、处理器以及存储在所述存储器中并可在所述处理器 上运行的计算机可读指令,其特征在于,所述处理器执行所述计算机可读指令时,所述处理器实现如下步骤:
    获取管理员所在当前网格对应的案件数,以及,获取与所述当前网格相邻的每个邻域网格对应的案件数和行驶时间;
    根据所述每个邻域网格对应的行驶时间,从所述当前网格的邻域网格中筛选出行驶时间最短的邻域网格,得到候选网格P 1
    判断所述候选网格P 1的案件数与当前网格的案件数的第一案件总和是否在预设范围内;
    若判断出所述第一案件总和在所述预设范围内,将所述当前网格与所述候选网格P 1分配给所述管理员。
  14. 根据权利要求13所述的电子终端,其特征在于,所述处理器执行所述计算机可读指令时,所述处理器还实现如下步骤:
    基于案件数量,将所述行驶时间最短的邻域网格按照指定顺序进行排序,得到案件数序列;
    根据所述案件数序列,获取案件数最多的邻域网格,并将该案件数最多的邻域网格确定为候选网格P 1
  15. 根据权利要求13所述的电子终端,其特征在于,所述处理器执行所述计算机可读指令时,所述处理器还实现如下步骤:
    确定候选网格P i为中心网格,其中,i为大于或等于1的正整数,当i=1时,所述中心网格为所述候选网格P 1
    获取与所述中心网格相邻的每个周围网格对应的案件数和行驶时间,其中,所述中心网格的领域网格不包括候选网格P i-1,其中,当i=1时,候选网格P i-1为所述当前网格;
    根据所述每个周围网格对应的案件数和行驶时间,从所述每个周围网格中筛选出行驶时间最短的邻域网格,得到候选网格P i+1
    判断所有候选网格的案件总量以及当前网格的案件数的第二案件总和是否在所述预设范围内;
    若判断出所述第二案件总和在所述预设范围内,将所述当前网格、所述所有候选网格分配给所述管理员;
    若判断出所述第二案件总和小于所述预设范围的下限值,令i=i+1,执行所述确定候选网格P i为中心网格的步骤。
  16. 根据权利要求14所述的电子终端,其特征在于,所述处理器执行所述计算机可读指令时,所述处理器还实现如下步骤:
    获取在所述案件数序列中与案件数最多的领域网格相邻的备选网格;
    将所述备选网格确定为重新选择的候选网格P 1
  17. 一种计算机非易失性可读存储介质,所述计算机非易失性可读存储介质存储有计算机可读指令,其特征在于,所述计算机可读指令被处理器执行时实现如下步骤:
    获取管理员所在当前网格对应的案件数,以及,获取与所述当前网格相邻的每个邻域网格对应的案件数和行驶时间;
    根据所述每个邻域网格对应的行驶时间,从所述当前网格的邻域网格中筛选出行驶时间最短的邻域网格,得到候选网格P 1
    判断所述候选网格P 1的案件数与当前网格的案件数的第一案件总和是否在预设范围内;
    若判断出所述第一案件总和在所述预设范围内,将所述当前网格与所述候选网格P 1分配给所述管理员。
  18. 根据权利要求17所述的计算机非易失性可读存储介质,其特征在于,所述计算机可读指令被一个或多个处理器执行时,使得所述一个或多个处理器还实现如下步骤:
    基于案件数量,将所述行驶时间最短的邻域网格按照指定顺序进行排序,得到案件数序列;
    根据所述案件数序列,获取案件数最多的邻域网格,并将该案件数最多的邻域网格确定为候选网格P1。
  19. 根据权利要求17所述的计算机非易失性可读存储介质,其特征在于,所述计算机可读指令被一个或多个处理器执行时,使得所述一个或多个处理器还实现如下步骤:
    确定候选网格P i为中心网格,其中,i为大于或等于1的正整数,当i=1时,所述中心网格为所述候选网格P 1
    获取与所述中心网格相邻的每个周围网格对应的案件数和行驶时间,其中,所述中心网格的领域网格不包括候选网格P i-1,其中,当i=1时,候选网格P i-1为所述当前网格;
    根据所述每个周围网格对应的案件数和行驶时间,从所述每个周围网格中筛选出行驶时间最短的邻域网格,得到候选网格P i+1
    判断所有候选网格的案件总量以及当前网格的案件数的第二案件总和是否在所述预设范围内;
    若判断出所述第二案件总和在所述预设范围内,将所述当前网格、所述所有候选网格分配给所述管理员;
    若判断出所述第二案件总和小于所述预设范围的下限值,令i=i+1,执行所述确定候选网格P i为中心网格的步骤。
  20. 根据权利要求18所述的计算机非易失性可读存储介质,其特征在于,所述计算机可读指令被一个或多个处理器执行时,使得所述一个或多个处理器还实现如下步骤:
    获取在所述案件数序列中与案件数最多的领域网格相邻的备选网格;
    将所述备选网格确定为重新选择的候选网格P 1
PCT/CN2019/091101 2018-12-31 2019-06-13 城市网格分配方法、装置及电子终端 Ceased WO2020140391A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811650804.0 2018-12-31
CN201811650804.0A CN109800974A (zh) 2018-12-31 2018-12-31 城市网格分配方法、装置及电子终端

Publications (1)

Publication Number Publication Date
WO2020140391A1 true WO2020140391A1 (zh) 2020-07-09

Family

ID=66556971

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/091101 Ceased WO2020140391A1 (zh) 2018-12-31 2019-06-13 城市网格分配方法、装置及电子终端

Country Status (2)

Country Link
CN (1) CN109800974A (zh)
WO (1) WO2020140391A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109800974A (zh) * 2018-12-31 2019-05-24 平安科技(深圳)有限公司 城市网格分配方法、装置及电子终端
CN112070247B (zh) * 2020-09-01 2024-09-13 上海钧正网络科技有限公司 一种运维网格的管理方法、装置、介质及电子设备
CN117151560B (zh) * 2023-11-01 2024-01-30 南京万物互联科技发展有限公司 一种基于云平台的智慧养老数字化管理系统及方法

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106355325A (zh) * 2016-08-30 2017-01-25 长威信息科技发展股份有限公司 一种警力资源与案件快速关联的方法及系统
CN107154954A (zh) * 2016-03-04 2017-09-12 苏州俊峰伟业信息技术有限公司 上门服务排单方法和系统
CN107330612A (zh) * 2017-06-28 2017-11-07 北京惠赢天下网络技术有限公司 一种配送区域调整方法、系统及服务器
CN108257025A (zh) * 2017-05-04 2018-07-06 平安科技(深圳)有限公司 孤儿保单分配方法和装置
CN109003028A (zh) * 2018-07-17 2018-12-14 北京百度网讯科技有限公司 用于划分物流区域的方法和装置
CN109754185A (zh) * 2018-12-31 2019-05-14 平安科技(深圳)有限公司 城市网格分配方法及装置、电子终端
CN109800974A (zh) * 2018-12-31 2019-05-24 平安科技(深圳)有限公司 城市网格分配方法、装置及电子终端

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10534776B2 (en) * 2011-04-29 2020-01-14 International Business Machines Corporation Proximity grids for an in-memory data grid
CN102591955B (zh) * 2011-12-29 2016-10-05 北京城市系统工程研究中心 网格化城市管理数据分析方法
CN108932553A (zh) * 2017-05-25 2018-12-04 北京嘀嘀无限科技发展有限公司 确定交通工具需求数据的方法及装置

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107154954A (zh) * 2016-03-04 2017-09-12 苏州俊峰伟业信息技术有限公司 上门服务排单方法和系统
CN106355325A (zh) * 2016-08-30 2017-01-25 长威信息科技发展股份有限公司 一种警力资源与案件快速关联的方法及系统
CN108257025A (zh) * 2017-05-04 2018-07-06 平安科技(深圳)有限公司 孤儿保单分配方法和装置
CN107330612A (zh) * 2017-06-28 2017-11-07 北京惠赢天下网络技术有限公司 一种配送区域调整方法、系统及服务器
CN109003028A (zh) * 2018-07-17 2018-12-14 北京百度网讯科技有限公司 用于划分物流区域的方法和装置
CN109754185A (zh) * 2018-12-31 2019-05-14 平安科技(深圳)有限公司 城市网格分配方法及装置、电子终端
CN109800974A (zh) * 2018-12-31 2019-05-24 平安科技(深圳)有限公司 城市网格分配方法、装置及电子终端

Also Published As

Publication number Publication date
CN109800974A (zh) 2019-05-24

Similar Documents

Publication Publication Date Title
CN109615122B (zh) 配送范围的生成方法、装置、电子设备和存储介质
CN108415975B (zh) 基于bdch-dbscan的出租车载客热点识别方法
CN108364467B (zh) 一种基于改进型决策树算法的路况信息预测方法
CN101299861B (zh) 一种基于最短回路的基站系统巡检路径自动化确定方法
CN111294819B (zh) 一种网络优化方法及装置
CN108616553B (zh) 云计算资源池进行资源调度的方法及装置
CN102711177A (zh) 基于业务预测的负载均衡方法
WO2020140391A1 (zh) 城市网格分配方法、装置及电子终端
CN106912015A (zh) 一种基于移动网络数据的人员出行链识别方法
CN109360421B (zh) 一种基于机器学习的交通信息预测方法及装置、电子终端
CN106535204B (zh) 一种业务覆盖质量评估方法及装置
CN105471670B (zh) 流量数据分类方法及装置
CN110213780A (zh) 网络切片的管理方法、管理和编排实体及存储介质
CN108092706B (zh) 一种映射方法
CN115460700A (zh) 基于联邦学习的网络资源配置方法、装置、电子设备及介质
CN106713396A (zh) 服务器调度方法和系统
CN113052217A (zh) 预测结果标识及其模型训练方法、装置及计算机存储介质
CN108647910B (zh) 城市公交站点的设置方法、装置、终端及计算机存储介质
CN113570867A (zh) 一种城市交通状态预测方法、装置、设备及可读存储介质
CN111367632A (zh) 一种基于周期特征的容器云调度方法
CN109661671A (zh) 使用边界位图对图像分类的改善
CN104270789B (zh) 基于数据共享的无线传感器网络的采样任务调度方法
CN101159976B (zh) 一种预测话务量及信道配置的方法及装置
CN109951856A (zh) 网元状态的检测方法、装置以及计算机可读存储介质
CN104915430B (zh) 一种基于MapReduce的约束关系粗糙集规则获取方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19907858

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19907858

Country of ref document: EP

Kind code of ref document: A1