WO2017167041A1 - 业务预测数据校正的方法和装置 - Google Patents

业务预测数据校正的方法和装置 Download PDF

Info

Publication number
WO2017167041A1
WO2017167041A1 PCT/CN2017/077243 CN2017077243W WO2017167041A1 WO 2017167041 A1 WO2017167041 A1 WO 2017167041A1 CN 2017077243 W CN2017077243 W CN 2017077243W WO 2017167041 A1 WO2017167041 A1 WO 2017167041A1
Authority
WO
WIPO (PCT)
Prior art keywords
service
time
data
prediction data
service prediction
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/CN2017/077243
Other languages
English (en)
French (fr)
Inventor
郑苏杭
徐萧萧
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.)
Alibaba Group Holding Ltd
Original Assignee
Alibaba Group Holding 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 Alibaba Group Holding Ltd filed Critical Alibaba Group Holding Ltd
Publication of WO2017167041A1 publication Critical patent/WO2017167041A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G06Q30/0202—Market predictions or forecasting for commercial activities
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising

Definitions

  • the present invention relates to the field of electronic technology applications, and in particular, to a method and apparatus for correcting traffic prediction data.
  • the business volume of the individual business on the day is affected by many factors such as activities, marketing means, and promotion time, the business volume of the individual business on the day will be significantly higher than the historical business volume of the business individual.
  • the business system will estimate the traffic volume of the next day, but if the predicted result is less than the actual turnover of the day, it indicates that the forecast result is wrong and the prediction accuracy is too low.
  • an hourly cumulative traffic volume prediction algorithm is generally used, that is, a ratio of accumulated traffic volume per hour to total traffic volume is obtained based on historical data, and then calculated based on the total predicted value of the current day traffic volume. Cumulative traffic forecast per hour.
  • the method can only Only when the business volume is relatively stable can we get better results. Once the real-time business volume changes significantly the next day, not only will the forecast value be inaccurate, but even when the 24 o'clock settlement is performed on the same day, the total traffic forecast will be A serious error that is less than the actual amount of traffic accumulated before 24 o'clock.
  • Embodiments of the present invention provide a method and apparatus for correcting service prediction data, so as to at least solve a technique for correcting a predicted traffic volume in the case where a business entity lacks a significant change in actual traffic volume on the next day due to lack of prior art.
  • the data is lower than the actual business volume, resulting in technical problems with low prediction accuracy.
  • a method for correcting service prediction data includes: extracting first service data at a first moment, where the first service data includes: a traffic volume generated at a first moment; Determining whether the second service prediction data corresponding to the second time is greater than the first service data, wherein the second time is an adjacent time point after the first time; and if the determination result is no, calculating the second service prediction data and The difference of the first service data; correcting the service prediction data corresponding to the second time to the settlement time according to the difference.
  • an apparatus for correcting service prediction data including: an extraction module, configured to extract first service data at a first moment, where the first service data includes: at a first a service volume that is generated at a time; the determining module is configured to determine whether the second service prediction data corresponding to the second time is greater than the first service data, wherein the second time is an adjacent time point after the first time; If the result of the determination is no, the difference between the second service prediction data and the first service data is calculated; and the correction module is configured to correct the service prediction data corresponding to the settlement time according to the difference.
  • the first service data is extracted at the first time, where the first service data includes: a traffic volume generated at the first time; and whether the second service prediction data corresponding to the second time is greater than a service data, wherein the second time is an adjacent time point after the first time; if the determination result is no, calculating a difference between the second service prediction data and the first service data; correcting the second according to the difference
  • the business forecast data corresponding to the time of settlement to the settlement time can achieve the purpose of timely correcting when the actual business volume of the business entity changes in the next day, thereby realizing the technical effect of improving the accuracy of the traffic volume prediction, thereby solving the prior art
  • FIG. 1 is a block diagram showing the hardware structure of a computer terminal of a method for correcting service prediction data according to an embodiment of the present invention
  • FIG. 2 is a flowchart of a method for correcting traffic prediction data according to Embodiment 1 of the present invention
  • FIG. 3a is a schematic flowchart diagram of a method for correcting service prediction data according to Embodiment 1 of the present invention.
  • FIG. 3b is a schematic flowchart diagram of another method for correcting service prediction data according to Embodiment 1 of the present invention.
  • FIG. 4 is a schematic diagram showing the curve of service data and service prediction data in a method for correcting service prediction data according to Embodiment 1 of the present invention
  • FIG. 5 is a schematic structural diagram of an apparatus for correcting service prediction data according to Embodiment 2 of the present invention.
  • FIG. 6 is a schematic structural diagram of an apparatus for correcting service prediction data according to Embodiment 2 of the present invention.
  • FIG. 7 is a schematic structural diagram of another apparatus for correcting service prediction data according to Embodiment 2 of the present invention.
  • FIG. 8 is a schematic structural diagram of another apparatus for correcting service prediction data according to Embodiment 2 of the present invention.
  • FIG. 9 is a schematic structural diagram of another apparatus for correcting service prediction data according to Embodiment 2 of the present invention.
  • a method embodiment of a method of service prediction data correction is also provided, it being noted that the steps illustrated in the flowchart of the accompanying drawings may be in a computer system such as a set of computer executable instructions The steps shown and described may be performed in a different order than the ones described herein, although the logical order is shown in the flowchart.
  • FIG. 1 is a service prediction data school according to an embodiment of the present invention.
  • FIG. 1 is a service prediction data school according to an embodiment of the present invention.
  • FIG. 1 is a service prediction data school according to an embodiment of the present invention.
  • FIG. 1 is a service prediction data school according to an embodiment of the present invention.
  • computer terminal 10 may include one or more (only one shown) processor 102 (processor 102 may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA)
  • a memory 104 for storing data
  • a transmission module 106 for communication functions.
  • FIG. 1 is merely illustrative and does not limit the structure of the above electronic device.
  • computer terminal 10 may also include more or fewer components than those shown in FIG. 1, or have a different configuration than that shown in FIG.
  • the memory 104 can be used to store software programs and modules of application software, such as program instructions/modules corresponding to the method of service prediction data correction in the embodiment of the present invention, and the processor 102 runs the software program and module stored in the memory 104, thereby Perform various functional applications and data processing, that is, implement the vulnerability detection method of the above application.
  • Memory 104 may include high speed random access memory, and may also include non-volatile memory such as one or more magnetic storage devices, flash memory, or other non-volatile solid state memory.
  • memory 104 may further include memory remotely located relative to processor 102, which may be coupled to computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.
  • Transmission device 106 is for receiving or transmitting data via a network.
  • the network specific examples described above may include a wireless network provided by a communication provider of the computer terminal 10.
  • the transmission device 106 includes a Network Interface Controller (NIC) that can be connected to other network devices through a base station to communicate with the Internet.
  • the transmission device 106 can be a Radio Frequency (RF) module for communicating with the Internet wirelessly.
  • NIC Network Interface Controller
  • RF Radio Frequency
  • FIG. 2 is a flow chart of a method of service prediction data correction according to a first embodiment of the present invention.
  • step S202 the first service data is extracted at the first time, where the first service data includes: a transaction volume generated at the first time;
  • the method for correcting the service prediction data provided by the embodiment of the present application can be applied to the correction of the forecast data of the business entity on the same day in the e-commerce platform, and avoiding the total forecast data of the business individual on the same day is less than the current time of the day.
  • the real value of the service data if it is determined whether the total predicted data of the operating entity is smaller than the actual transaction volume, the first service data needs to be first extracted at the first moment.
  • the first service data may be an accumulated value of the transaction volume from 0 o'clock to the first time.
  • the first moment can be accurate to the second, that is, when the first service data is extracted, the structure of the first moment can be expressed as: year-month-day hour: minute: second (yyyy-mm- Dd hh:mm:ss) point.
  • the first service data extracted by the first moment will be the traffic volume accurate to the second, and the transaction volume based on the second to the second is
  • the correction accuracy of the predicted value is improved, and the correction process is as shown in steps S204 to S208.
  • Step S204 determining whether the second service prediction data corresponding to the second time is greater than the first service data, where the second time is an adjacent time point after the first time;
  • the second service prediction data corresponding to the second time is compared with the first service data, and it is determined whether the second service prediction data is greater than the first service.
  • the first service data extracted at the first time is: service data Date1 of 09:15:20
  • the second service prediction data corresponding to the second time may be: service prediction data Date2 corresponding to 10:00, After Date1, Date2 and Date1 are judged to determine whether the Date2 is greater than Date1.
  • Step S206 if the determination result is no, calculating a difference between the second service prediction data and the first service data
  • step S206 of the present application if the result of the determination is that the second prediction data is smaller than the first service data, the difference between the second service prediction data and the first service data may be obtained.
  • step S204 is still taken as an example.
  • Date1 is greater than Date2
  • Step S208 correcting the service prediction data corresponding to the second time to the settlement time according to the difference.
  • the service prediction data corresponding to the second time to the settlement time is corrected according to the difference, and the service data and the service prediction data of the day are determined according to the embodiment of the present application.
  • 24 hours per day is divided into 24 stages, that is, 0 to 24 points, a total of 24 time periods, and when the difference is obtained, the business prediction data corresponding to the second time to 24 points is corrected.
  • step S204 is taken as an example.
  • the difference between Date1 and Date2 is D
  • the second time is 10 points
  • the 15 time points from 10 to 24 are corrected according to the difference D.
  • Business forecast data is
  • the predicted value of the traffic (that is, the second service prediction data provided by the embodiment of the present application) is a pre-business data pre-measure obtained by the prediction algorithm before the predicted service date. It is estimated that the real-time service data is the actual service quantity value obtained in real time on the same day (that is, the first service data provided by the embodiment of the present application).
  • the predicted value of the traffic is a pre-business data pre-measure obtained by the prediction algorithm before the predicted service date.
  • the real-time service data is the actual service quantity value obtained in real time on the same day (that is, the first service data provided by the embodiment of the present application).
  • FIG. 3 is a schematic flowchart of another method for correcting service prediction data according to the first embodiment of the present invention. As shown in FIG. 3a, in the method for correcting service prediction data provided by the embodiment of the present application, the total service is predicted by offline.
  • the ratio of the quantity to the offline predicted hourly sales volume, the offline data ie, the service prediction data in the embodiment of the present application
  • the actual service data obtained by updating at the first time is greater if the service data is greater than the first time.
  • the correction value ie, the difference value in the embodiment of the present application
  • the second time to the settlement time is corrected by the correction value.
  • the solution provided in the first embodiment of the present application is to extract the first service data at the first time, where the first service data includes: the volume of the service generated at the first time; and the second time is determined. Whether the second service prediction data is greater than the first service data; if the determination result is negative, calculating a difference between the second service prediction data and the first service data; correcting the service prediction corresponding to the settlement time according to the difference time.
  • the data achieves the purpose of timely correction when the actual business volume of the business entity changes on the second day, thereby realizing the technical effect of improving the accuracy of the traffic volume prediction, and thus solving the second problem due to the lack of the prior art in the prior art.
  • FIG. 3b is a schematic flowchart of another method for correcting service prediction data according to the first embodiment of the present invention.
  • the method for correcting service prediction data provided by the embodiment of the present application is as follows:
  • the extracting the first service data at the first moment in step S202 includes:
  • Step1 determining whether the first moment is greater than the settlement time
  • Step 1 of the above step S202 before extracting the first service data, it is necessary to determine whether the current extraction time is the settlement time, that is, if the first time is the settlement time, indicating that the day has ended, it is not necessary for the current time.
  • the corresponding service data is compared with the service prediction data, and the verification process ends. Therefore, before extracting the first service data, it is necessary to determine whether the time at which the first service data is extracted is the settlement time.
  • Step 2 When the determination result is no, the first service data is extracted.
  • Step 2 when the result of the determination is that the first time is less than the settlement time, the first service data is extracted.
  • the method for correcting service prediction data is specifically as follows: determining whether the current first time is greater than the settlement time, and determining the result in the process of extracting the first service data. In the case of YES, the first service data extraction process ends; if the determination result is negative, the first service data corresponding to the first time is extracted, wherein the first service data may be accurate to the second time at the first time. In the case of the extracted transaction volume, the volume of the business may be from 0 to the accumulated volume at the first moment.
  • the method for correcting the service prediction data further includes:
  • Step S200 Obtain the proportion of the accumulated traffic of the service prediction data and the preset time to the total amount of the service;
  • the service prediction data in the embodiment of the present application may be exemplified by the daily service prediction data, and the accumulated traffic volume of the preset time may be described by taking the accumulated traffic volume per hour as an example.
  • the business forecast data for each hour of the day will be obtained in advance.
  • the daily business forecast data needs to be acquired, at least 7 days.
  • Historical business data and the ratio of hourly total business data secondly, based on the historical business data of at least 7 days and the ratio of hourly total business data to the ratio of hourly accumulated business volume to total business volume,
  • the specific gravity in the application examples is taken as the average of the proportions of the last 7 days. That is, first calculate the ratio of each hour to the total business data of the day, and then take the average of the ratio of the last 7 days of each hour. This avoids abnormalities that are sudden or low on a certain day.
  • the daily service prediction data in the embodiment of the present application may be calculated by sampling the daily average business data in a time interval, that is, by collecting the daily business transaction data of one week (7 days), by way of summation.
  • the total transaction data of the week is obtained, and the daily average business data is obtained by dividing the total transaction data by dividing by 7 transaction days, and the daily average business data is used as the daily business forecast data.
  • the daily business transaction data can be Date1, Date2, Date3, Date4, Date5, Date6 and Date7.
  • the total transaction data of one week is expressed as follows. :
  • DateZ1 ⁇ 7 Date1+Date2+Date3+Date4+Date5+Date6+Date7;
  • the Date Day1 is used as the daily service prediction data of Day1; or, the daily service prediction data of Day1 is obtained by the prediction algorithm model.
  • Day7 P70 ⁇ P724; among them, the ratio of the daily business data of Day1 to Day7 to the total business data of the day is the proportion of accumulated traffic per hour to the total business data of the day.
  • P 23 ' (P 1 24+P 2 24+...+P 7 24)/7.
  • the method for obtaining the daily service prediction data and the accumulated traffic volume per hour in the total amount of the service provided by the embodiment of the present application is only exemplified by the above, to implement the method for correcting the service prediction data provided by the embodiment of the present application.
  • the standard is not limited.
  • Step S201 Calculate the proportion of the accumulated traffic of the service prediction data and the preset time to the total amount of the service, and obtain the service prediction data corresponding to each time point.
  • the obtained daily service prediction data and the accumulated traffic per hour account for the proportion of the total business volume. Calculate, you will get hourly business forecast data. details as follows:
  • Step1 Calculate the product of the daily business forecast data and the specific gravity, and determine the product as the hourly business forecast data.
  • the obtained product can be determined as the hourly traffic prediction data. details as follows,
  • step S200 assuming that the daily business forecast data is Date Day1 , the cumulative traffic volume per hour accounts for P 0 ', P 1 ', ..., P 24 ': respectively, Date Day1
  • P 0 ', P 1 ', ..., P 24 ' the hourly traffic prediction data D'0 to D'24 of Day1
  • the proportion of the accumulated traffic in the preset time to the total amount of the service may also be the proportion of the accumulated traffic per hour in the second-hour per second in the total amount of the service, which is preset in the embodiment of the present application.
  • the cumulative traffic of time is exemplified by the accumulated traffic volume per hour.
  • the cumulative traffic volume of the preset time accounts for the proportion of the total traffic volume, and the proportion of the accumulated traffic volume per hour to the total traffic volume is taken as an example to implement the present application.
  • Traffic prediction data correction provided by the embodiment The method is correct and the specifics are not limited.
  • the method for correcting the service prediction data provided in the embodiment of the present application obtains the hourly service prediction data by averaging in steps S200 and S201.
  • the service prediction data calculation model can be generated by using the access rate of the store, the addition rate of the shopping cart, the collection rate of the product, and the transaction data as the training feature, and the service prediction data is obtained.
  • the method for predicting the service data provided by the embodiment of the present application is averaged.
  • the value algorithm is taken as an example to implement the method for implementing the service prediction data provided by the embodiment of the present application, which is not limited.
  • the service prediction data corresponding to the second time to the settlement time according to the difference correction in step S208 includes:
  • Step1 generating a correction value according to the difference
  • Step 1 in the above step S208 of the present application in a case where the second service prediction data is smaller than the first service data in step S206, a correction value is generated according to the difference between the first service data and the second service prediction data.
  • Step 2 Adjust the service prediction data corresponding to the second time to the settlement time according to the correction value, and obtain the corrected service prediction data.
  • the service prediction data corresponding to the second time to the settlement time is adjusted based on the correction value, and the corrected service prediction data is obtained.
  • adjusting the service prediction data corresponding to the second time to the settlement time according to the correction value may include two modes, where manner 1 is to determine the difference as the correction value, and add the difference
  • the service prediction data corresponding to the second time to the settlement time is obtained, and the corrected service prediction data is obtained; in the second method, the correction specific gravity is generated according to the difference, and the service prediction data corresponding to the second time to the settlement time is adjusted according to the correction specific gravity,
  • the corrected business forecast data is obtained.
  • mode one performs step A
  • mode two performs step B and step C.
  • step 2 in step S208 adjusts the service prediction data corresponding to the second time to the settlement time according to the correction value, and the corrected service prediction data includes:
  • the service prediction data corresponding to the second time to the settlement time is adjusted according to the correction value, and the corrected service prediction data includes the following two methods:
  • the correction value is the difference value
  • Step A In the case that the difference is determined as the correction value, the difference is added to the service prediction data corresponding to the second time to the settlement time to obtain the corrected service prediction data;
  • step A of the present application the difference between the first service data and the second service prediction data is compared with the hourly service prediction data corresponding to the second time to the settlement time, and the corrected hourly traffic prediction data is obtained.
  • the hourly traffic prediction data to the settlement time may be D'i ⁇ D'24, and D'i ⁇ D'24 are respectively summed with ⁇ d, and the corrected hourly traffic prediction data D'i+ ⁇ d ⁇ D is obtained. '24+ ⁇ d.
  • the correction value is a case of the correction weight generated based on the difference
  • Step B generating a corrected specific gravity according to the difference, and determining the corrected specific gravity as the correction value
  • step B of the present application it is assumed that the difference is ⁇ d, and the corrected specific gravity Bi is generated according to the ⁇ d.
  • step C the service prediction data corresponding to the second time to the settlement time is adjusted according to the corrected specific gravity, and the corrected service prediction data is obtained.
  • step B Based on the corrected specific gravity obtained in step B, in the above step C of the present application, after the corrected specific gravity Bi is obtained, the service prediction data corresponding to the second time to the settlement time is adjusted according to the corrected specific gravity Bi, and the corrected service prediction data is obtained. .
  • the corrected specific gravity Bi, the hourly traffic prediction data from the second time to the settlement time may be D'i to D'24, and D'i to D'24 are respectively multiplied with Bi to obtain the corrected hourly service.
  • the predicted data D'i*Bi ⁇ D'24*Bi.
  • the total traffic volume of the offline prediction day is X (ie, the daily service prediction data in the embodiment of the present application), and the offline traffic volume per hour accounts for the total traffic volume ratio.
  • X the daily service prediction data in the embodiment of the present application
  • the offline traffic volume per hour accounts for the total traffic volume ratio.
  • the real-time data is the actual generated traffic that is tracked in real time on the day of the traffic, and is recorded as Ti (ie, the first service data in the embodiment of the present application), i is accurate to the second, that is, yyyy-mm-dd hh:mm :ss moments.
  • the actually generated service data is re-extracted, that is, the first
  • the service data is extracted until the extracted service data is greater than the service prediction data of the adjacent location time after the extraction time corresponding to the service data; as shown in FIG. 3b, after a calibration process ends, the data is entered.
  • the second correction preparation flow that is, in the case where the service data at the time T1 is greater than the service prediction data at the time T2, the correction flow will be executed.
  • FIG. 4 is a schematic diagram of a service data and service prediction data in a method for correcting service prediction data according to a first embodiment of the present invention.
  • a ratio of accumulated traffic per hour to a total amount of services is obtained according to historical data, and then multiplied
  • the predicted value of the 24 hours of the day is obtained, and a prediction curve with the hour as the horizontal axis and the traffic volume as the vertical axis is drawn.
  • the real-time data is larger than the next-hour data, the difference is calculated, and the predicted value of each hour after the current time is added to the difference to complete the replenishment.
  • the dotted line after 10 o'clock is the predicted value
  • the value of 23 o'clock is the total predicted value of the traffic X
  • the data obtained in real time at 09:15:20 is T9
  • the whole of T9 is followed by one hour. Compare the predicted data at 10 points.
  • the T9 value in the above figure is not larger than the 10 point prediction value, but if the data obtained in real time at 09:15:20 is as thick as the picture, it is obvious that the value is larger than the predicted value of 10 points.
  • the curve appears at this time. Fighting peaks.
  • the method for correcting the service prediction data provided by the embodiment of the present application can be applied to the single-day promotion situation such as “Double Eleven” or “Double Twelve”, and can also be applied to the activity time continuously exceeding A two-day promotion, for example, a three-day multi-day event, a golden week; or an hour or a preset time promotion, based on the above environment, the method for correcting the business prediction data in the embodiment of the present application to the business data Correction principle, that is, regardless of any promotion activity, the business forecast data during the promotion period is generated on the day before or during the previous period of the activity, and when the promotion is performed, if the actual business data is greater than The service prediction data corresponding to the time point at the time is generated by comparing the service prediction data with the service data at the time, and correcting the service prediction data from the adjacent time point to the settlement time according to the correction value.
  • the single-day promotion situation such as “Double Eleven” or “Double Twelve”
  • the business data of the 3-day promotion period activity will be predicted, and then the 3-day business prediction data will be obtained, that is, D day1 and D day2. And D day3 .
  • the service prediction data corresponding to the neighboring time point is further executed to compare the service data with the service prediction data, and compare the generated correction values to correct the service prediction data corresponding to the settlement time of the current time point; similarly, 3 Day promotion
  • the correction of the service prediction data of Day 2 and Day 3 is the same as that in Day 1 and will not be described here.
  • the method of correcting the business forecast data is the same as the 3-day promotion period.
  • the business data of the one-hour promotion day will be predicted. If the promotion is on the day of the promotion, the actual business generated during the hour. The data is larger than the service prediction data corresponding to the adjacent time point of the hour, and is the same as the method for correcting the service prediction data, and the correction value is generated by comparing the service data with the service prediction data, and then the adjacent time point is corrected to the current day according to the correction value. Business forecast data at the time of settlement.
  • the method for correcting the service prediction data provided by the embodiment of the present application can be applied to an Open Data Processing Service (ODPS) deployed by the Facebook Company using a Structured Query Language (SQL). platform.
  • ODPS Open Data Processing Service
  • SQL Structured Query Language
  • the method for correcting the service prediction data according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, can also be through hardware, but In many cases the former is a better implementation.
  • the technical solution of the present invention which is essential or contributes to the prior art, may be embodied in the form of a software product stored in a storage medium (such as ROM/RAM, disk,
  • the optical disc includes a number of instructions for causing a terminal device (which may be a cell phone, a computer, a server, or a network device, etc.) to perform the methods described in various embodiments of the present invention.
  • an apparatus embodiment for implementing the foregoing method embodiments is also provided.
  • the apparatus provided by the foregoing embodiment of the present application may be run on a computer terminal.
  • FIG. 5 is a schematic structural diagram of an apparatus for correcting service prediction data according to Embodiment 2 of the present invention.
  • the apparatus for correcting the service prediction data may include: an extraction module 52, a determination module 54, a calculation module 56, and a correction module 58.
  • the extracting module 52 is configured to extract the first service data at the first time, where the first service data includes: a service volume generated at the first time; the determining module 54 is configured to determine a second time corresponding to the second time The service prediction data is greater than the first service data, wherein the second time is an adjacent time point after the first time; the calculation module 56 is configured to calculate the second service prediction data and the first case if the determination result is no The difference between the service data and the correction module 58 is configured to correct the service prediction data corresponding to the second time to the settlement time according to the difference.
  • the solution provided by the foregoing embodiment 2 of the present application is to extract the first service data at the first time, where the first service data includes: a transaction volume generated at the first time; and a second time corresponding to the second time Whether the second service prediction data is greater than the first service data; if the determination result is negative, calculating a difference between the second service prediction data and the first service data; correcting the service prediction corresponding to the settlement time according to the difference time
  • the data achieves the purpose of timely correction when the actual business volume of the business entity changes on the second day, thereby realizing the technical effect of improving the accuracy of the traffic volume prediction, and thus solving the second problem due to the lack of the prior art in the prior art.
  • the foregoing extraction module 52, the determination module 54, the calculation module 56, and the correction module 58 correspond to the steps S202 to S208 in the first embodiment, and the examples and application scenarios implemented by the four modules and the corresponding steps. The same, but not limited to, the content disclosed in the above embodiment 1. It should be noted that the foregoing module may be implemented in the computer terminal 10 provided in the first embodiment as a part of the device, and may be implemented by software or by hardware.
  • FIG. 6 is a schematic structural diagram of an apparatus for correcting service prediction data according to Embodiment 2 of the present invention.
  • the extraction module 52 includes a determination unit 521 and an extraction unit 522.
  • the determining unit 521 is configured to determine whether the first time is greater than the settlement time; and the extracting unit 522 is configured to: when the determination result is negative, extract the first service data.
  • the foregoing determining unit 521 and extracting unit 522 correspond to Step 1 and Step 2 in step S202 in the first embodiment, and the two modules are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to the above.
  • the foregoing module may be implemented in the computer terminal 10 provided in the first embodiment as a part of the device, and may be implemented by software or by hardware.
  • FIG. 7 is a schematic structural diagram of another apparatus for correcting service prediction data according to Embodiment 2 of the present invention.
  • the apparatus for correcting service prediction data provided by the embodiment of the present application further includes: an obtaining module 50 and a data calculating module 51.
  • the obtaining module 50 is configured to obtain service prediction data and before extracting the first service data at the first moment.
  • the cumulative traffic of the preset time is the proportion of the total traffic;
  • the data calculation module 51 is configured to calculate the proportion of the accumulated traffic of the service prediction data and the preset time to the total amount of the service, and obtain the service prediction corresponding to each time point. data.
  • the foregoing obtaining module 50 and the data calculating module 51 correspond to step S200 and step S201 in the first embodiment.
  • the two modules are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to the above implementation.
  • the foregoing module may be implemented in the computer terminal 10 provided in the first embodiment as a part of the device, and may be implemented by software or by hardware.
  • FIG. 8 is a schematic structural diagram of another apparatus for correcting service prediction data according to Embodiment 2 of the present invention.
  • the correction module 58 includes a numerical value generating unit 581 and a correcting unit 582.
  • the value generating unit 581 is configured to generate a correction value according to the difference value
  • the correcting unit 582 is configured to adjust the service prediction data corresponding to the second time to the settlement time according to the correction value to obtain the corrected service prediction data.
  • the above-mentioned numerical value generating unit 581 and the correcting unit 582 correspond to Step 1 and Step 2 in step S208 in the first embodiment, and the two modules are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited thereto.
  • the foregoing module may be implemented in the computer terminal 10 provided in the first embodiment as a part of the device, and may be implemented by software or by hardware.
  • FIG. 9 is a schematic structural diagram of another apparatus for correcting service prediction data according to Embodiment 2 of the present invention.
  • the correcting unit 582 includes a first correcting subunit 5821, a numerical generating subunit 5822, and a second correcting subunit 5823.
  • the first correction subunit 5821 is configured to: when the difference is determined as the correction value, add the difference to the service prediction data corresponding to the second time to the settlement time to obtain the corrected service prediction data; or
  • the value generating sub-unit 5822 is configured to generate a corrected specific gravity according to the difference, and determine the corrected specific gravity as the correction value.
  • the second correcting sub-unit 5823 is configured to adjust the service prediction data corresponding to the second time to the settlement time according to the corrected specific gravity. , the corrected business forecast data is obtained.
  • first correcting subunit 5821, the numerical generating subunit 5822, and the second correcting subunit 5823 correspond to steps A to C in Step 2 in step S208 in the first embodiment, and the three modules are The examples and application scenarios implemented by the corresponding steps are the same, but are not limited to the contents disclosed in the above embodiment 1. It should be noted that the foregoing module may be implemented in the computer terminal 10 provided in the first embodiment as a part of the device, and may be implemented by software or by hardware.
  • the device for correcting the service prediction data extracts the first time a service data, and comparing the first service data with the second service prediction data at the second moment, and calculating the first service data and the second service prediction data if the second service prediction data is smaller than the first service data
  • the difference between the two, and the business forecast data of the second time to the settlement time is corrected according to the difference, and the effect of the dotted line can be achieved by the complementary technique to smoothly improve the predicted values of all the subsequent points, and the cause is effectively solved in time.
  • the predictions caused by sudden changes in individual effects on the day are inaccurate or even predictive.
  • Embodiments of the present invention also provide a storage medium.
  • the foregoing storage medium may be used to save the program code executed by the method for correcting service prediction data provided in Embodiment 1 above.
  • the foregoing storage medium may be located in any one of the computer terminal groups in the computer network, or in any one of the mobile terminal groups.
  • the storage medium is configured to store program code for performing the following steps: extracting the first service data at the first moment, wherein the first service data comprises: the service generated at the first moment The volume of the second service is determined to be greater than the first service data, wherein the second time is an adjacent time point after the first time; and if the determination result is no, the second service is calculated. The difference between the predicted data and the first service data; the service prediction data corresponding to the settlement time from the second time to the settlement time is corrected according to the difference.
  • the storage medium is configured to store program code for performing the following steps: determining whether the first time is greater than the settlement time; and if the determination result is negative, extracting the first service data.
  • the storage medium is configured to store program code for performing the following steps: acquiring daily business forecast data and accumulative traffic per hour as a percentage of the total business volume; The hourly accumulated business volume accounts for the proportion of the total business volume, and the hourly business forecast data is obtained.
  • the storage medium is arranged to store program code for performing the following steps: calculating a product of daily traffic prediction data and a specific gravity, and determining the product as hourly traffic prediction data.
  • the storage medium is configured to store program code for performing the following steps: adding the difference value to the hourly traffic prediction data corresponding to the current time to the current settlement time, and obtaining the corrected each time Hour business forecast data.
  • the disclosed technical content may be through other Way to achieve.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • multiple units or components may be combined or may be Integrate into another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, unit or module, and may be electrical or otherwise.
  • 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, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, 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 a software functional unit.
  • the integrated unit if implemented in the form of a software functional unit and sold or used as a standalone product, may be stored in a computer readable storage medium.
  • the technical solution of the present invention which is essential or contributes to the prior art, or all or part of the technical solution, may be embodied in the form of a software product stored in a storage medium.
  • a number of instructions are included to cause a computer device (which may be a personal computer, server or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention.
  • the foregoing storage medium includes: a U disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic disk, or an optical disk, and the like. .

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Human Resources & Organizations (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • General Physics & Mathematics (AREA)
  • Marketing (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Game Theory and Decision Science (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Data Mining & Analysis (AREA)

Abstract

本发明公开了一种业务预测数据校正的方法和装置。其中，该方法包括：在第一时刻提取第一业务数据，其中，第一业务数据包括：在第一时刻产生的业务成交量；判断第二时刻对应的第二业务预测数据是否大于第一业务数据，其中，第二时刻为第一时刻之后相邻的时间点；在判断结果为否的情况下，计算第二业务预测数据与第一业务数据的差值；依据差值校正第二时刻至结算时刻对应的业务预测数据。本发明解决了由于现有技术中缺少对经营个体在第二天实际业务量显著变化的情况下校正预测业务量的技术，导致预测数据低于实际业务量，从而带来的预测精度低的技术问题。

Description

业务预测数据校正的方法和装置
本申请要求2016年04月01日递交的申请号为201610204277.5、发明名称为“业务预测数据校正的方法和装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及电子技术应用领域，具体而言，涉及一种业务预测数据校正的方法和装置。
背景技术
随着电商平台的日益发展，对电商平台中未来营业额的预测计算，成为了当前越来越多的电商预测未来营业额的技术手段，而如何精确的预估得到电商的未来营业额成为了亟待解决的问题。
现有技术中由于经营个体当日业务量受活动、营销手段以及促销时间等诸多因素的影响，该经营个体当日的业务量将比该经营个体的历史业务量存在明显提升。而在该当日之前，商务系统将会对第二天的业务量进行预估，但是若预估得到的预测结果小于该当日实际营业额，则说明该预估结果存在错误，预测精度太低。
现有预测经营个体的营业额的技术中，通常使用小时累计业务量预测算法，即，根据历史数据获得各小时累计业务量占总业务量的比例，然后根据当日业务量的总预测值计算出每小时的累计业务量预测值。
但是，根据历史数据预测经营个体第二天的业务量(或，营业额)，以及当天每小时的累计业务量，从而根据24个小时数据点描绘经营个体业务量的变化趋势，该方法仅能只在业务量较平稳时才能取到较好的效果，一旦第二天实时业务量发生显著变化时，不仅会出现预测值不准确的问题，甚至会出现当天24点结算时，业务量预测总值比24点前已累计的真实业务量小的严重错误。
针对上述由于现有技术中缺少对经营个体在第二天实际业务量显著变化的情况下校正预测业务量的技术，导致预测数据低于实际业务量，从而带来的预测精度低的问题，目前尚未提出有效的解决方案。
发明内容
本发明实施例提供了一种业务预测数据校正的方法和装置，以至少解决由于现有技术中缺少对经营个体在第二天实际业务量显著变化的情况下校正预测业务量的技术，导致预测数据低于实际业务量，从而带来的预测精度低的技术问题。
根据本发明实施例的一个方面，提供了一种业务预测数据校正的方法，包括：在第一时刻提取第一业务数据，其中，第一业务数据包括：在第一时刻产生的业务成交量；判断第二时刻对应的第二业务预测数据是否大于第一业务数据，其中，第二时刻为第一时刻之后相邻的时间点；在判断结果为否的情况下，计算第二业务预测数据与第一业务数据的差值；依据差值校正第二时刻至结算时刻对应的业务预测数据。
根据本发明实施例的另一方面，还提供了一种业务预测数据校正的装置，包括：提取模块，用于在第一时刻提取第一业务数据，其中，第一业务数据包括：在第一时刻产生的业务成交量；判断模块，用于判断第二时刻对应的第二业务预测数据是否大于第一业务数据，其中，第二时刻为第一时刻之后相邻的时间点；计算模块，用于在判断结果为否的情况下，计算第二业务预测数据与第一业务数据的差值；校正模块，用于依据差值校正第二时刻至结算时刻对应的业务预测数据。
在本发明实施例中，通过在第一时刻提取第一业务数据，其中，第一业务数据包括：在第一时刻产生的业务成交量；判断第二时刻对应的第二业务预测数据是否大于第一业务数据，其中，第二时刻为第一时刻之后相邻的时间点；在判断结果为否的情况下，计算第二业务预测数据与第一业务数据的差值；依据差值校正第二时刻至结算时刻对应的业务预测数据，达到了对经营个体第二天实际业务量发生变化时能够及时校正的目的，从而实现了提升对业务量预测精度的技术效果，进而解决了由于现有技术中缺少对经营个体在第二天实际业务量显著变化的情况下校正预测业务量的技术，导致预测数据低于实际业务量，从而带来的预测精度低的技术问题。
附图说明
此处所说明的附图用来提供对本发明的进一步理解，构成本申请的一部分，本发明的示意性实施例及其说明用于解释本发明，并不构成对本发明的不当限定。在附图中：
图1是本发明实施例的一种业务预测数据校正的方法的计算机终端的硬件结构框图；
图2是根据本发明实施例一的业务预测数据校正的方法的流程图；
图3a是根据本发明实施例一的一种业务预测数据校正的方法的流程示意图；
图3b是根据本发明实施例一的另一种业务预测数据校正的方法的流程示意图；
图4是根据本发明实施例一的一种业务预测数据校正的方法中业务数据和业务预测数据的曲线示意图；
图5是根据本发明实施例二的业务预测数据校正的装置的结构示意图；
图6是根据本发明实施例二的一种业务预测数据校正的装置的结构示意图；
图7是根据本发明实施例二的另一种业务预测数据校正的装置的结构示意图；
图8是根据本发明实施例二的又一种业务预测数据校正的装置的结构示意图；
图9是根据本发明实施例二的再一种业务预测数据校正的装置的结构示意图。
具体实施方式
为了使本技术领域的人员更好地理解本发明方案，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分的实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都应当属于本发明保护的范围。
需要说明的是，本发明的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换，以便这里描述的本发明的实施例能够以除了在这里图示或描述的那些以外的顺序实施。此外，术语“包括”和“具有”以及他们的任何变形，意图在于覆盖不排他的包含，例如，包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元，而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
实施例1
根据本发明实施例，还提供了一种业务预测数据校正的方法的方法实施例，需要说明的是，在附图的流程图示出的步骤可以在诸如一组计算机可执行指令的计算机系统中执行，并且，虽然在流程图中示出了逻辑顺序，但是在某些情况下，可以以不同于此处的顺序执行所示出或描述的步骤。
本申请实施例一所提供的方法实施例可以在移动终端、计算机终端或者类似的运算装置中执行。以运行在计算机终端上为例，图1是本发明实施例的一种业务预测数据校 正的方法的计算机终端的硬件结构框图。如图1所示，计算机终端10可以包括一个或多个(图中仅示出一个)处理器102(处理器102可以包括但不限于微处理器MCU或可编程逻辑器件FPGA等的处理装置)、用于存储数据的存储器104、以及用于通信功能的传输模块106。本领域普通技术人员可以理解，图1所示的结构仅为示意，其并不对上述电子装置的结构造成限定。例如，计算机终端10还可包括比图1中所示更多或者更少的组件，或者具有与图1所示不同的配置。
存储器104可用于存储应用软件的软件程序以及模块，如本发明实施例中的业务预测数据校正的方法对应的程序指令/模块，处理器102通过运行存储在存储器104内的软件程序以及模块，从而执行各种功能应用以及数据处理，即实现上述的应用程序的漏洞检测方法。存储器104可包括高速随机存储器，还可包括非易失性存储器，如一个或者多个磁性存储装置、闪存、或者其他非易失性固态存储器。在一些实例中，存储器104可进一步包括相对于处理器102远程设置的存储器，这些远程存储器可以通过网络连接至计算机终端10。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
传输装置106用于经由一个网络接收或者发送数据。上述的网络具体实例可包括计算机终端10的通信供应商提供的无线网络。在一个实例中，传输装置106包括一个网络适配器(Network Interface Controller，NIC)，其可通过基站与其他网络设备相连从而可与互联网进行通讯。在一个实例中，传输装置106可以为射频(Radio Frequency，RF)模块，其用于通过无线方式与互联网进行通讯。
在上述运行环境下，本申请提供了如图2所示的业务预测数据校正的方法。图2是根据本发明实施例一的业务预测数据校正的方法的流程图。
步骤S202，在第一时刻提取第一业务数据，其中，第一业务数据包括：在第一时刻产生的业务成交量；
本申请上述步骤S202中，本申请实施例提供的业务预测数据校正的方法可以适用于电商平台中对经营个体当天业务量的预测数据校正，避免经营个体当天业务量总预测数据小于当日某时刻业务数据真实值，在本申请实施例提供的业务预测数据校正的方法中，若要判定经营个体的总预测数据是否小于实际的业务成交量，则需要首先在第一时刻提取第一业务数据，该第一业务数据可以为0点至第一时刻的业务成交量的累计值。
其中，本申请实施例中在第一时刻可以精确至秒，即，提取第一业务数据时，该第一时刻的结构可以表示为：年-月-日小时：分钟：秒(yyyy-mm-dd hh:mm:ss)的时刻 点。在本申请实施例中由于第一时刻可以精确到秒，而由该第一时刻提取的第一业务数据将为精确至秒的业务成交量，综上依据精确至秒的业务成交量在对后续预测数据校正时，将提高预测值的校正精度，其中，校正过程见步骤S204至S208。
步骤S204，判断第二时刻对应的第二业务预测数据是否大于第一业务数据，其中，第二时刻为第一时刻之后相邻的时间点；
基于步骤S202中提取的第一业务数据，本申请上述步骤S204中，将第二时刻对应的第二业务预测数据与该第一业务数据进行比较，判断该第二业务预测数据是否大于第一业务数据，其中，第二时刻为第一时刻之后相邻的时间点。
具体的，假设第一时刻提取的第一业务数据为：09:15:20的业务数据Date1，第二时刻对应的第二业务预测数据可以为：10:00对应的业务预测数据Date2，在得到Date1后将Date2与Date1进行判断，判断该Date2是否大于Date1。
步骤S206，在判断结果为否的情况下，计算第二业务预测数据与第一业务数据的差值；
本申请上述步骤S206中，在判断结果为第二预测数据小于第一业务数据的情况下，可以得到计算第二业务预测数据与第一业务数据的差值。
具体的，仍旧以步骤S204中的示例为例，Date1大于Date2时，计算Date2与Date1之间的差值，即，差值D＝Date1-Date2。
步骤S208，依据差值校正第二时刻至结算时刻对应的业务预测数据。
基于步骤S206得到的差值，本申请上述步骤S208中，依据该差值，对第二时刻至结算时刻对应的业务预测数据进行校正，本申请实施例中将一天的业务数据和业务预测数据依据每天24个小时划分为24个阶段，即，0点至24点，共24个时段，在得到该差值时，将对第二时刻至24点对应的业务预测数据进行校正。
具体的，仍旧以步骤S204中的示例为例，当得到Date1与Date2的差值为D时，若第二时刻为10点，则依据该差值D校正10点至24点这15个时间点的业务预测数据。
结合步骤S202至步骤S208，在本申请实施例中，业务量的预测值(即，本申请实施例提供的第二业务预测数据)是在预测业务日期之前通过预测算法得到的当天业务数据的预估，实时业务数据是当天实时获取的实际业务量值(即，本申请实施例提供的第一业务数据)。为了实现平滑补差，先根据历史数据获取各小时累计业务量占业务总量的比值，然后将其乘以当日业务量预测值，得到当天25个小时点的预测值，据此绘制一条以小时为横轴，业务量为纵轴的预测曲线。当实时数据比下一小时数据大时(即， 第一业务数据大于第二业务预测数据的情况下)，计算该差值(计算第一业务数据与第二业务预测数据的差值)，并将当前时刻后面各小时点的预测值都加上该差值即完成补差。图3a是根据本发明实施例一的另一种业务预测数据校正的方法的流程示意图，如图3a所示，在本申请实施例提供的业务预测数据校正的方法中，通过依据离线预测总业务量与离线预测小时销量的占比，得到离线数据(即，本申请实施例中的业务预测数据)，通过在第一时刻更新得到的实际业务数据，若该业务数据大于与第一时刻相邻的第二时刻对应的业务预测数据，则通过比较业务数据与该业务预测数据，生成校正值(即，本申请实施例中的差值)，进而通过校正值校正第二时刻至结算时刻对应的业务预测数据。
由上可知，本申请上述实施例一所提供的方案，通过在第一时刻提取第一业务数据，其中，第一业务数据包括：在第一时刻产生的业务成交量；判断第二时刻对应的第二业务预测数据是否大于第一业务数据；在判断结果为否的情况下，计算第二业务预测数据与第一业务数据的差值；依据差值校正第二时刻至结算时刻对应的业务预测数据，达到了对经营个体第二天实际业务量发生变化时能够及时校正的目的，从而实现了提升对业务量预测精度的技术效果，进而解决了由于现有技术中缺少对经营个体在第二天实际业务量显著变化的情况下校正预测业务量的技术，导致预测数据低于实际业务量，从而带来的预测精度低的技术问题。
具体的，图3b是根据本发明实施例一的另一种业务预测数据校正的方法的流程示意图，如图3b所示，本申请实施例提供的业务预测数据校正的方法具体如下：
可选的，步骤S202中在第一时刻提取第一业务数据包括：
Step1，判断第一时刻是否大于结算时刻；
本申请上述步骤S202中的Step1中，在提取第一业务数据之前，需要判定当前的提取时间是否为结算时刻，即，若第一时刻为结算时刻，表明当天已经结束，已经没有必要对当前时刻对应的业务数据和业务预测数据进行比较，校验过程结束，所以在提取第一业务数据前，需对提取第一业务数据的时刻是否为结算时刻进行判断。
Step2，在判断结果为否的情况下，提取第一业务数据。
基于Step1的判断，本申请上述步骤Step2中，在判断结果为该第一时刻小于结算时刻的情况下，提取第一业务数据。
具体的，结合Step1和Step2，本申请实施例提供的业务预测数据校正的方法在提取第一业务数据的过程中具体如下：判断当前第一时刻是否大于结算时刻，在判断结果 为是的情况下，第一业务数据提取流程结束；在判断结果为否的情况下，提取第一时刻对应的第一业务数据，其中，第一业务数据可以为在第一时刻精确至秒的情况下，提取的业务成交量，该业务成交量可以为0点至第一时刻的累积成交量。
可选的，在步骤S202中的在第一时刻提取第一业务数据之前，本申请实施例提供的业务预测数据校正的方法还包括：
步骤S200，获取业务预测数据和预设时间的累积业务量占业务总量的比重；
本申请上述步骤S200中，本申请实施例中的业务预测数据可以以日业务预测数据为例，预设时间的累积业务量可以以每小时累积业务量为例进行说明，其中，本申请实施例中在实际业务成交之前的一天将预先得到该日一天每个小时的业务预测数据，其中，在获取每个小时的业务预测数据的过程中，首先，需要获取日业务预测数据、至少7天的历史业务数据和每小时占当天总业务数据的比值；其次，依据至少7天的历史业务数据和每小时占当天总业务数据的比值得到每小时累积业务量占业务总量的比重，其中，本申请实施例中的比重取的是最近7天比重的平均值。即，先算每个小时占当天总业务数据的比值，再取每小时最近7天比值的平均。这样避免某天突高或突低的异常。
具体的，本申请实施例中的日业务预测数据可以通过在一个时间区间内采样计算得到的日均业务数据，即，通过采集一周(7天)的每日业务成交数据，通过求和的方式得到一周的业务成交总数据，进而依据该业务成交总数据通过除以7个成交日得到日均业务数据，并将该日均业务数据作为日业务预测数据。
假设，需要获取Day1(2月1日)的日业务预测数据，则需要获取Day1之前1月26至1月31日7天的成交数据，该经营个体在1月26至1月31日的每日业务成交数据可以为Date1，Date2，Date3，Date4，Date5，Date6和Date7，通过对7天每日业务成交数据的求和，可以得到一周的业务成交总数据，一周的业务成交总数据表示如下：
DateZ1～7＝Date1+Date2+Date3+Date4+Date5+Date6+Date7；
进而为得到预测Day1的日业务预测数据，则一周的业务成交总数据通过除以7个成交日得到日均业务数据，即，DateDay1＝DateZ1～7/7，在得到日均业务数据后可以将该DateDay1作为Day1的日业务预测数据；或者，通过预测算法模型得到Day1的日业务预测数据。
若获取每小时累积业务量占业务总量的比重，则需要先算每个小时占当天总业务数据的比值，再取每小时最近7天比值的平均，最后得到每小时累积业务量占业务总量的比重，例如，先计算7天内每小时占当天总业务数据的比值，可以得到：
Day1:P10～P124；
Day2:P20～P224；
……
Day7:P70～P724；其中，Day1至Day7的每小时占当天总业务数据的比值为每小时累积业务量占当天总业务数据的占比。
进而，再取每小时最近7天比值的平均，可以得到：
P0’＝(P10+P20+……+P70)/7；
P1’＝(P11+P21+……+P71)/7；
……
P23’＝(P124+P224+……+P724)/7。
由上可知，得到每小时累积业务量占业务总量的比重P0’，P1’，……，P24’。
这里需要说明的是本申请实施例提供的获取日业务预测数据和每小时累积业务量占业务总量的比重的方法仅以上述为例，以实现本申请实施例提供的业务预测数据校正的方法为准，具体不做限定。
步骤S201，将业务预测数据和预设时间的累积业务量占业务总量的比重进行计算，得到对应各个时间点的业务预测数据。
基于步骤S200获取的日业务预测数据和每小时累积业务量占业务总量的比重，本申请上述步骤S201中，将获取到的日业务预测数据和每小时累积业务量占业务总量的比重进行计算，将得到每小时业务预测数据。具体如下：
Step1，计算日业务预测数据和比重的乘积，将乘积确定为每小时业务预测数据。
由上述步骤S200和S201可知，通过计算日业务预测数据和比重的乘积，可以将得到乘积确定为每小时业务预测数据。具体如下，
仍旧以步骤S200中的示例为例，假设日业务预测数据为DateDay1，每小时累积业务量占业务总量的比重为P0’，P1’，……，P24’：将DateDay1分别与P0’，P1’，……，P24’进行乘积，可以得到Day1的每小时业务预测数据D’0～D’24。
此外，预设时间的累积业务量占业务总量的比重还可以为以秒为单位的每个小时中秒级的每小时累积业务量占业务总量的比重，本申请实施例中的预设时间的累积业务量以每小时的累积业务量为例，预设时间的累积业务量占业务总量的比重以每小时的累积业务量占业务总量的比重为例进行说明，以实现本申请实施例提供的业务预测数据校正 的方法为准，具体不做限定。
综上，结合步骤S200和步骤S201中计算每小时业务预测数据的方法，本申请实施例提供的业务预测数据校正的方法中除了步骤S200和步骤S201中通过求平均值得到每小时业务预测数据，还可以通过店铺的访问率、购物车的添加率、商品的收藏率以及交易数据作为训练特征生成业务预测数据计算模型，进而得到业务预测数据，本申请实施例提供的业务预测数据的方法以平均值算法为例进行说明，以实现本申请实施例提供的业务预测数据的方法为准，具体不做限定。
可选的，步骤S208中依据差值校正第二时刻至结算时刻对应的业务预测数据包括：
Step1，依据差值生成校正值；
本申请上述步骤S208中的Step1中，在步骤S206中第二业务预测数据小于第一业务数据的情况下，依据第一业务数据和第二业务预测数据的差值生成校正值。
Step2，依据校正值对第二时刻至结算时刻对应的业务预测数据进行调整，得到校正后的业务预测数据。
基于步骤Step1生成的校正值，本申请上述Step2中，依据校正值对第二时刻至结算时刻对应的业务预测数据进行调整，得到校正后的业务预测数据。
具体的，本申请实施例中依据校正值对第二时刻至结算时刻对应的业务预测数据进行调整可以包括两种方式，其中，方式一，为将差值确定为校正值，并将差值加至第二时刻至结算时刻对应的业务预测数据，得到校正后的业务预测数据；方式二，依据差值生成校正比重，并依据校正比重对第二时刻至结算时刻对应的业务预测数据进行调整，得到校正后的业务预测数据。这里方式一执行步骤A，方式二执行步骤B和步骤C。
进一步地，可选的，步骤S208中的Step2中依据校正值对第二时刻至结算时刻对应的业务预测数据进行调整，得到校正后的业务预测数据包括：
本申请实施例中，依据校正值对第二时刻至结算时刻对应的业务预测数据进行调整，得到校正后的业务预测数据包括以下两种方式：
方式一，校正值为差值的情况；
步骤A，在将差值确定为校正值的情况下，将差值加至第二时刻至结算时刻对应的业务预测数据，得到校正后的业务预测数据；
本申请上述步骤A中，将第一业务数据与第二业务预测数据求差得到的差值与第二时刻至结算时刻对应的每小时业务预测数据求和，得到校正后的每小时业务预测数据。
具体的，假设第一业务数据与第二业务预测数据求差得到的差值为△d，第二时刻 至结算时刻的每小时业务预测数据可以为D’i～D’24，将D’i～D’24分别与△d求和，得到校正后的每小时业务预测数据D’i+△d～D’24+△d。
或者，
方式二，校正值为依据差值生成的校正比重的情况；
步骤B，依据差值生成校正比重，并将校正比重确定为校正值；
本申请上述步骤B中，假设该差值为△d，依据该△d生成校正比重Bi。
步骤C，对第二时刻至结算时刻对应的业务预测数据依据校正比重进行调整，得到校正后的业务预测数据。
基于步骤B中得到的校正比重，本申请上述步骤C中，在得到校正比重Bi后，依据该校正比重Bi对第二时刻至结算时刻对应的业务预测数据进行调整，得到校正后的业务预测数据。
具体的，假设校正比重Bi，第二时刻至结算时刻的每小时业务预测数据可以为D’i～D’24，将D’i～D’24分别与Bi乘积，得到校正后的每小时业务预测数据D’i*Bi～D’24*Bi。
结合上述步骤S202至步骤S208，如图3b所示，假设离线预测当天总业务量为X(即，本申请实施例中的日业务预测数据)，离线预测每小时业务量占总业务量占比为pn，n＝0,1,2,…,23(即，本申请实施例中的每小时累积业务量占业务总量的比重)，则24个小时业务量预测值为X*pn(n＝0,1,2,…,23)(即，本申请实施例中的每小时业务预测数据)。实时数据是在业务量当天实时追踪到的实际已产生业务量，记为Ti(即，本申请实施例中的第一业务数据)，i精确到秒，即表示yyyy-mm-dd hh:mm:ss的时刻点。补差技术指，当前时刻i(小时为h)获取的实时数据Ti与(h+1)小时预测值(第二时刻对应的第二业务预测数据)做对比，即Δt＝Ti-X*Ph+1(即，本申请实施例中的第二业务预测数据与第一业务数据的差值)，当Δt>0时将(h+1)小时到23点的每小时预测值加上Δt(即，本申请实施例中的将差值加至第二时刻至结算时刻对应的每小时业务预测数据，得到校正后的每小时业务预测数据)。这里需要说明的是，如图3b所示，当实施数据比小时预测值小时(即，第一业务数据小于第二业务预测数据的情况下)，则重新提取实际产生的业务数据，即，第一时刻之后，结算时刻之前，提取业务数据，直至提取的业务数据大于该业务数据对应的提取时刻之后相邻位置时刻的业务预测数据；如图3b所示，在一次校正流程结束后，将进入第二次校正准备流程，即，在存在T1时刻的业务数据大于T2时刻的业务预测数据的情况下，将执行校正流程。
图4是根据本发明实施例一的一种业务预测数据校正的方法中业务数据和业务预测数据的曲线示意图；先根据历史数据获取各小时累计业务量占业务总量的比值，然后将其乘以当日业务量预测值，得到当天24个小时点的预测值，据此绘制一条以小时为横轴，业务量为纵轴的预测曲线。当实时数据比下一小时数据大时，计算该差值，并将当前时刻后面各小时点的预测值都加上该差值即完成补差。如图4所示，10点之后的虚线部分都是预测值，23点的值即为业务量总预测值X，09:15:20实时获得的数据为T9，将T9与其后一个小时的整点10点的预测数据比较。上图T9值未比10点预测值大不做修正，但假若09:15:20实时拿到的数据如画的粗线条，明显发现其值比10点的预测值大，此时的曲线出现了斗峰。这种预测是完全错误的，因为曲线表现的是小时的累计数据，T10之后的值肯定比T9大，至少是持平，通过补差技术可以达到带点虚线的效果将所有后面点的预测值做个平滑提升，及时有效解决了因当天经营个体效果突变而造成的预测不准确甚至是预测离谱问题。
基于上述步骤S202至步骤S208，本申请实施例提供的业务预测数据校正的方法除了可以适用于“双十一”或“双十二”这类单天促销情况，还可以适用于活动时间持续超过两天的促销活动，例如，连续三天的聚划算活动、黄金周；又或者一个小时或预设时间内的促销活动，基于上述环境，本申请实施例中业务预测数据校正的方法对业务数据的校正原理，即，无论任一促销活动，在该活动开始的前一天或前一时间段内都会生成该促销活动时期内的业务预测数据，当在促销活动进行时，若实际的业务数据大于该时刻相邻时间点对应的业务预测数据，则通过比较该业务预测数据和该时刻的业务数据，生成校正值，并依据该校正值校正该时刻相邻时间点至结算时刻的业务预测数据。
举例来说，以3天的促销期为例，在3天促销期之前，将对该3天促销期活动的业务数据进行预测，进而得到3天的业务预测数据，即，Dday1、Dday2和Dday3。
当在3天促销的活动中时，假设第一天Day1的11:35:30秒的实际产生的业务数据Date1’大于12点至1点该1个小时时间段的业务预测数据Date2，则依据当前的业务数据Date1’与该业务预测数据Date2生成校正值J1，J1可以为Date1’与Date2的差值，在对12点至Day1当天结算时间对应的业务预测数据Date2～Date12进行校正时，可以通过将J1分别添加至该业务预测数据，即，得到Date2+J1～Date12+J1；当本次校正结束后，在Day1当天11:35:30(秒级)之后还存在实际的业务数据大于相邻时间点对应的业务预测数据，则继续执行将业务数据与业务预测数据进行比较，并通过比较生成的校正值，校正该相邻时间点至当天结算时刻对应的业务预测数据；同理，3天促销 的Day2和Day3的业务预测数据的校正与Day1中的方法相同，在此不再赘述。与3天的促销期相同的是，黄金周的促销活动中，对业务预测数据校正的方法与3天的促销期相同。
若是一小时的促销时间，同样的，在该一小时促销活动发起的前一天，将会对发起促销一小时当天的业务数据做一预测，若在促销活动当天，该小时内的实际产生的业务数据大于该小时相邻时间点对应的业务预测数据，则与上述业务预测数据校正的方法相同，通过业务数据与业务预测数据的比较生成校正值，进而依据该校正值校正相邻时间点至当天结算时刻的业务预测数据。
这里本申请实施例提供的业务预测数据校正的方法可以适用于阿里巴巴公司的一种利用结构化查询语言(Structures Query Language，简称SQL)部署的大数据计算服务(Open Data Processing Service，简称ODPS)平台。
需要说明的是，对于前述的各方法实施例，为了简单描述，故将其都表述为一系列的动作组合，但是本领域技术人员应该知悉，本发明并不受所描述的动作顺序的限制，因为依据本发明，某些步骤可以采用其他顺序或者同时进行。其次，本领域技术人员也应该知悉，说明书中所描述的实施例均属于优选实施例，所涉及的动作和模块并不一定是本发明所必须的。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到根据上述实施例的业务预测数据校正的方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端设备(可以是手机，计算机，服务器，或者网络设备等)执行本发明各个实施例所述的方法。
实施例2
根据本发明实施例，还提供了一种用于实施上述方法实施例的装置实施例，本申请上述实施例所提供的装置可以在计算机终端上运行。
图5是根据本发明实施例二的业务预测数据校正的装置的结构示意图。
如图5所示，该业务预测数据校正的装置可以包括：提取模块52、判断模块54、计算模块56和校正模块58。
其中，提取模块52，用于在第一时刻提取第一业务数据，其中，第一业务数据包括：在第一时刻产生的业务成交量；判断模块54，用于判断第二时刻对应的第二业务预测数据是否大于第一业务数据，其中，第二时刻为第一时刻之后相邻的时间点；计算模块56，用于在判断结果为否的情况下，计算第二业务预测数据与第一业务数据的差值；校正模块58，用于依据差值校正第二时刻至结算时刻对应的业务预测数据。
由上可知，本申请上述实施例二所提供的方案，通过在第一时刻提取第一业务数据，其中，第一业务数据包括：在第一时刻产生的业务成交量；判断第二时刻对应的第二业务预测数据是否大于第一业务数据；在判断结果为否的情况下，计算第二业务预测数据与第一业务数据的差值；依据差值校正第二时刻至结算时刻对应的业务预测数据，达到了对经营个体第二天实际业务量发生变化时能够及时校正的目的，从而实现了提升对业务量预测精度的技术效果，进而解决了由于现有技术中缺少对经营个体在第二天实际业务量显著变化的情况下校正预测业务量的技术，导致预测数据低于实际业务量，从而带来的预测精度低的技术问题。
此处需要说明的是，上述提取模块52、判断模块54、计算模块56和校正模块58对应于实施例一中的步骤S202至步骤S208，四个模块与对应的步骤所实现的示例和应用场景相同，但不限于上述实施例一所公开的内容。需要说明的是，上述模块作为装置的一部分可以运行在实施例一提供的计算机终端10中，可以通过软件实现，也可以通过硬件实现。
可选的，图6是根据本发明实施例二的一种业务预测数据校正的装置的结构示意图。如图6所示，提取模块52包括：判断单元521和提取单元522。
其中，判断单元521，用于判断第一时刻是否大于结算时刻；提取单元522，用于在判断结果为否的情况下，提取第一业务数据。
此处需要说明的是，上述判断单元521和提取单元522对应于实施例一中的步骤S202中的Step1和Step2，两个模块与对应的步骤所实现的示例和应用场景相同，但不限于上述实施例一所公开的内容。需要说明的是，上述模块作为装置的一部分可以运行在实施例一提供的计算机终端10中，可以通过软件实现，也可以通过硬件实现。
可选的，图7是根据本发明实施例二的另一种业务预测数据校正的装置的结构示意图。如图7所示，本申请实施例提供的业务预测数据校正的装置还包括：获取模块50和数据计算模块51。
其中，获取模块50，用于在第一时刻提取第一业务数据之前，获取业务预测数据和 预设时间的累积业务量占业务总量的比重；数据计算模块51，用于将业务预测数据和预设时间的累积业务量占业务总量的比重进行计算，得到对应各个时间点的业务预测数据。
此处需要说明的是，上述获取模块50和数据计算模块51对应于实施例一中的步骤S200和步骤S201，两个模块与对应的步骤所实现的示例和应用场景相同，但不限于上述实施例一所公开的内容。需要说明的是，上述模块作为装置的一部分可以运行在实施例一提供的计算机终端10中，可以通过软件实现，也可以通过硬件实现。
可选的，图8是根据本发明实施例二的又一种业务预测数据校正的装置的结构示意图。如图8所示，校正模块58包括：数值生成单元581和校正单元582。
其中，数值生成单元581，用于依据差值生成校正值；校正单元582，用于依据校正值对第二时刻至结算时刻对应的业务预测数据进行调整，得到校正后的业务预测数据。
此处需要说明的是，上述数值生成单元581和校正单元582对应于实施例一中的步骤S208中的Step1和Step2，两个模块与对应的步骤所实现的示例和应用场景相同，但不限于上述实施例一所公开的内容。需要说明的是，上述模块作为装置的一部分可以运行在实施例一提供的计算机终端10中，可以通过软件实现，也可以通过硬件实现。
进一步地，可选的，图9是根据本发明实施例二的再一种业务预测数据校正的装置的结构示意图。如图9所示，校正单元582包括：第一校正子单元5821、数值生成子单元5822和第二校正子单元5823。
其中，第一校正子单元5821，用于在将差值确定为校正值的情况下，将差值加至第二时刻至结算时刻对应的业务预测数据，得到校正后的业务预测数据；或者，数值生成子单元5822，用于依据差值生成校正比重，并将校正比重确定为校正值；第二校正子单元5823，用于对第二时刻至结算时刻对应的业务预测数据依据校正比重进行调整，得到校正后的业务预测数据。
此处需要说明的是，上述第一校正子单元5821、数值生成子单元5822和第二校正子单元5823对应于实施例一中的步骤S208中Step2中的步骤A至步骤C，三个模块与对应的步骤所实现的示例和应用场景相同，但不限于上述实施例一所公开的内容。需要说明的是，上述模块作为装置的一部分可以运行在实施例一提供的计算机终端10中，可以通过软件实现，也可以通过硬件实现。
由上可知，本申请实施例提供的业务预测数据校正的装置，通过在第一时刻提取第 一业务数据，并将该第一业务数据与第二时刻的第二业务预测数据进行比较，在第二业务预测数据小于第一业务数据的情况下，计算第一业务数据与第二业务预测数据之间的差值，并依据该差值校正第二时刻至结算时刻的业务预测数据，通过补差技术可以达到带点虚线的效果将所有后面点的预测值做个平滑提升，及时有效解决了因当天经营个体效果突变而造成的预测不准确甚至是预测离谱问题。
实施例3
本发明的实施例还提供了一种存储介质。可选地，在本实施例中，上述存储介质可以用于保存上述实施例一所提供的业务预测数据校正的方法所执行的程序代码。
可选地，在本实施例中，上述存储介质可以位于计算机网络中计算机终端群中的任意一个计算机终端中，或者位于移动终端群中的任意一个移动终端中。
可选地，在本实施例中，存储介质被设置为存储用于执行以下步骤的程序代码：在第一时刻提取第一业务数据，其中，第一业务数据包括：在第一时刻产生的业务成交量；判断第二时刻对应的第二业务预测数据是否大于第一业务数据，其中，第二时刻为第一时刻之后相邻的时间点；在判断结果为否的情况下，计算第二业务预测数据与第一业务数据的差值；依据差值校正第二时刻至结算时刻对应的业务预测数据。
可选地，在本实施例中，存储介质被设置为存储用于执行以下步骤的程序代码：判断第一时刻是否大于结算时刻；在判断结果为否的情况下，提取第一业务数据。
可选地，在本实施例中，存储介质被设置为存储用于执行以下步骤的程序代码：获取日业务预测数据和每小时累积业务量占业务总量的比重；将日业务预测数据和每小时累积业务量占业务总量的比重进行计算，得到每小时业务预测数据。
可选地，在本实施例中，存储介质被设置为存储用于执行以下步骤的程序代码：计算日业务预测数据和比重的乘积，将乘积确定为每小时业务预测数据。
可选地，在本实施例中，存储介质被设置为存储用于执行以下步骤的程序代码：将差值加至第二时刻至当日结算时刻对应的每小时业务预测数据，得到校正后的每小时业务预测数据。
上述本发明实施例序号仅仅为了描述，不代表实施例的优劣。
在本发明的上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述的部分，可以参见其他实施例的相关描述。
在本申请所提供的几个实施例中，应该理解到，所揭露的技术内容，可通过其它的 方式实现。其中，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，单元或模块的间接耦合或通信连接，可以是电性或其它的形式。
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，在本发明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，也可以采用软件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品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设备(可为个人计算机、服务器或者网络设备等)执行本发明各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、移动硬盘、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述仅是本发明的优选实施方式，应当指出，对于本技术领域的普通技术人员来说，在不脱离本发明原理的前提下，还可以做出若干改进和润饰，这些改进和润饰也应视为本发明的保护范围。

Claims (10)

  1. 一种业务预测数据校正的方法，其特征在于，包括：
    在第一时刻提取第一业务数据，其中，所述第一业务数据包括：在所述第一时刻产生的业务成交量；
    判断第二时刻对应的第二业务预测数据是否大于所述第一业务数据，其中，所述第二时刻为所述第一时刻之后相邻的时间点；
    在判断结果为否的情况下，计算所述第二业务预测数据与所述第一业务数据的差值；
    依据所述差值校正所述第二时刻至结算时刻对应的业务预测数据。
  2. 根据权利要求1所述的方法，其特征在于，所述在第一时刻提取第一业务数据包括：
    判断所述第一时刻是否大于结算时刻；
    在判断结果为否的情况下，提取所述第一业务数据。
  3. 根据权利要求1所述的方法，其特征在于，所述在第一时刻提取第一业务数据之前，所述方法还包括：
    获取业务预测数据和预设时间的累积业务量占业务总量的比重；
    将所述业务预测数据和所述预设时间的累积业务量占业务总量的比重进行计算，得到对应各个时间点的业务预测数据。
  4. 根据权利要求1所述的方法，其特征在于，所述依据所述差值校正所述第二时刻至结算时刻对应的业务预测数据包括：
    依据所述差值生成校正值；
    依据所述校正值对所述第二时刻至所述结算时刻对应的业务预测数据进行调整，得到校正后的所述业务预测数据。
  5. 根据权利要求4所述的方法，其特征在于，所述依据所述校正值对所述第二时刻至所述结算时刻对应的业务预测数据进行调整，得到校正后的所述业务预测数据包括：
    在将所述差值确定为所述校正值的情况下，将所述差值加至所述第二时刻至所述结算时刻对应的业务预测数据，得到校正后的所述业务预测数据；或者，
    依据所述差值生成校正比重，并将所述校正比重确定为所述校正值；
    对所述第二时刻至所述结算时刻对应的业务预测数据依据所述校正比重进行调整， 得到校正后的所述业务预测数据。
  6. 一种业务预测数据校正的装置，其特征在于，包括：
    提取模块，用于在第一时刻提取第一业务数据，其中，所述第一业务数据包括：在所述第一时刻产生的业务成交量；
    判断模块，用于判断第二时刻对应的第二业务预测数据是否大于所述第一业务数据，其中，所述第二时刻为所述第一时刻之后相邻的时间点；
    计算模块，用于在判断结果为否的情况下，计算所述第二业务预测数据与所述第一业务数据的差值；
    校正模块，用于依据所述差值校正所述第二时刻至结算时刻对应的业务预测数据。
  7. 根据权利要求6所述的装置，其特征在于，所述提取模块包括：
    判断单元，用于判断所述第一时刻是否大于结算时刻；
    提取单元，用于在判断结果为否的情况下，提取所述第一业务数据。
  8. 根据权利要求6所述的装置，其特征在于，所述装置还包括：
    获取模块，用于在第一时刻提取第一业务数据之前，获取业务预测数据和预设时间的累积业务量占业务总量的比重；
    数据计算模块，用于将所述业务预测数据和所述预设时间的累积业务量占业务总量的比重进行计算，得到对应各个时间点的业务预测数据。
  9. 根据权利要求6所述的装置，其特征在于，所述校正模块包括：
    数值生成单元，用于依据所述差值生成校正值；
    校正单元，用于依据所述校正值对所述第二时刻至所述结算时刻对应的业务预测数据进行调整，得到校正后的所述业务预测数据。
  10. 根据权利要求9所述的装置，其特征在于，所述校正单元包括：
    第一校正子单元，用于在将所述差值确定为所述校正值的情况下，将所述差值加至所述第二时刻至所述结算时刻对应的业务预测数据，得到校正后的所述业务预测数据；或者，
    数值生成子单元，用于依据所述差值生成校正比重，并将所述校正比重确定为所述校正值；
    第二校正子单元，用于对所述第二时刻至所述结算时刻对应的业务预测数据依据所述校正比重进行调整，得到校正后的所述业务预测数据。
PCT/CN2017/077243 2016-04-01 2017-03-20 业务预测数据校正的方法和装置 Ceased WO2017167041A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610204277.5A CN107292649A (zh) 2016-04-01 2016-04-01 业务预测数据校正的方法和装置
CN201610204277.5 2016-04-01

Publications (1)

Publication Number Publication Date
WO2017167041A1 true WO2017167041A1 (zh) 2017-10-05

Family

ID=59963443

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/077243 Ceased WO2017167041A1 (zh) 2016-04-01 2017-03-20 业务预测数据校正的方法和装置

Country Status (3)

Country Link
CN (1) CN107292649A (zh)
TW (1) TW201737164A (zh)
WO (1) WO2017167041A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108805385A (zh) * 2018-04-02 2018-11-13 阿里巴巴集团控股有限公司 一种评价商户的经营状况的方法、装置及设备
CN111311014B (zh) * 2020-02-27 2024-04-12 广州酷旅旅行社有限公司 业务数据处理方法、装置、计算机设备和存储介质
CN112488377B (zh) * 2020-11-25 2025-01-10 上海中通吉网络技术有限公司 快递日订单量的预测方法、装置、存储介质及电子设备
TWI913083B (zh) * 2024-11-01 2026-01-21 韓商韓領有限公司 確認未來時點之商品之銷售量資訊之方法、電子裝置及記錄媒體

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101729690A (zh) * 2008-10-20 2010-06-09 中兴通讯股份有限公司 一种排班系统和方法
CN102711177A (zh) * 2012-04-26 2012-10-03 北京邮电大学 基于业务预测的负载均衡方法
CN103490956A (zh) * 2013-09-22 2014-01-01 杭州华为数字技术有限公司 基于业务量预测的自适应节能控制方法及设备、系统
CN105205297A (zh) * 2014-06-12 2015-12-30 腾讯科技（深圳）有限公司 一种时间序列预测方法及系统

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103678514B (zh) * 2013-11-26 2017-01-18 科大讯飞股份有限公司 一种业务趋势预测方法及系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101729690A (zh) * 2008-10-20 2010-06-09 中兴通讯股份有限公司 一种排班系统和方法
CN102711177A (zh) * 2012-04-26 2012-10-03 北京邮电大学 基于业务预测的负载均衡方法
CN103490956A (zh) * 2013-09-22 2014-01-01 杭州华为数字技术有限公司 基于业务量预测的自适应节能控制方法及设备、系统
CN105205297A (zh) * 2014-06-12 2015-12-30 腾讯科技（深圳）有限公司 一种时间序列预测方法及系统

Also Published As

Publication number Publication date
CN107292649A (zh) 2017-10-24
TW201737164A (zh) 2017-10-16

Similar Documents

Publication Publication Date Title
CN104463603B (zh) 一种信用评估方法及系统
CN107093141B (zh) 产品对象的处理方法和装置
WO2017167041A1 (zh) 业务预测数据校正的方法和装置
US10504128B2 (en) Systems and methods for improved billing and ordering
CN112580889B (zh) 服务资源预估方法、装置、电子设备及存储介质
CN114255040B (zh) 账户充值提示方法、装置、电子设备以及存储介质
CN108364085A (zh) 一种外卖配送时间预测方法和装置
WO2019184583A1 (zh) 基于电子书的活动内容推送方法及电子设备
CN107404500A (zh) 群组通信方法、装置及移动终端
CN103955821A (zh) 一种充值方法和充值终端以及充值服务器
CN110009429A (zh) 一种预测流量数据的方法、装置和计算机设备
CN104023326A (zh) 流量超额提醒方法和装置
CN108932627A (zh) Cpa广告的消耗速度的预估方法、装置及可读存储介质
CN108965951B (zh) 广告的播放方法及装置
CN111050008A (zh) 一种账户余额提醒方法、装置、电子设备及存储介质
CN116187728A (zh) 客服排班优化方法、装置、设备及存储介质
CN116664289A (zh) 业务信息的预测方法和装置、存储介质及电子装置
CN108564458B (zh) 一种基于蒙特卡洛模拟的消费预测方法及系统
CN103955854A (zh) 一种账户管理设备及方法
HK1245969A1 (zh) 业务预测数据校正的方法和装置
CN106971326B (zh) 订单的用户预期管理系统和方法
CN117649019A (zh) 业务月收入数据预测方法、装置、设备及介质
CN109300037A (zh) 交易进度显示方法及装置
CN110060020B (zh) 项目费用预测方法以及装置
CN110264369B (zh) 费用计算的核对方法、装置、电子设备以及存储介质

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

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

Ref document number: 17773071

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 17773071

Country of ref document: EP

Kind code of ref document: A1