Traffic volume prediction method, device, electronic device and storage medium

Abstract

The present application discloses a business volume prediction method, device, electronic device and storage medium. The business volume prediction method includes: determining the target area and target holiday for the business volume to be predicted; obtaining the historical business volume of the target area on the target holiday; calculating the business volume difference ratio sequence of the target holiday based on the historical business volume; predicting the business volume of the target area on the target holiday based on the business volume difference ratio sequence. The embodiment of the present application realizes the prediction of business volume by calculating the business volume difference ratio sequence of the target area on the target holiday. This process predicts the business volume based on the difference ratio data of a certain holiday in a certain area. Compared with simply observing the number of pieces in each period, this difference ratio sequence corresponding to each holiday can better outline the consumption habits of a region in the face of a certain type of holiday stimulation, making the business volume prediction closer to the actual scene and improving the accuracy of the business volume prediction.

Description

Technical Field

[0001] The present application relates to the field of logistics technology, and in particular to a business volume prediction method, device, electronic device and storage medium. Background Art

[0002] With the continuous deepening of the information management reform of the logistics industry, the information management of the logistics industry is faced with the problem of how to effectively extract useful information from a large amount of complex data. Data mining technology can extract useful information hidden in a large amount of fuzzy data to analyze and predict future development.

[0003] In the field of logistics, holiday volume prediction has always been a major unresolved problem in the express delivery industry. Existing time series models and analysis methods can provide relatively accurate estimates of weekday volume, but cannot estimate the magnitude and rhythm of volume mutations during holidays. The sudden change in holiday volume may be affected by many factors, but the amount of data that can be mastered by the forecaster is generally very limited. After all, the amount of holiday data is much smaller than that of weekdays, which makes it difficult to support the needs of the model, and the correlation between the data and the final mutation results is also difficult to predict. Clustering is a way to increase the available samples of the model, but at present, the industry lacks a clustering method for holiday and city characteristics. A city’s holiday consumption habits cannot be classified by simple volume composition.

[0004] In summary, the existing technology for predicting the quantity of goods has limitations, which leads to inaccurate prediction of the quantity of goods. Summary of the invention

[0005] The embodiments of the present application provide a business volume prediction method, device, electronic device and storage medium, which make the business volume prediction closer to the actual scenario and improve the accuracy of the business volume prediction.

[0006] On the one hand, the present application provides a traffic volume prediction method, the method comprising:

[0007] Determine the target area and target holidays for the business volume to be forecasted;

[0008] Obtain the historical business volume of the target area on the target holiday;

[0009] Calculate the business volume difference ratio sequence of the target holiday according to the historical business volume;

[0010] The business volume of the target area during the target holiday is predicted according to the business volume difference ratio sequence.

[0011] In some embodiments of the present application, the calculating, based on the historical business volume, the business volume difference ratio sequence of the target holiday includes:

[0012] Determine the holiday observation time period and the normal observation time period corresponding to the target holiday;

[0013] From the historical business volume, obtaining the daily historical business volume of the target area during the festival observation period;

[0014] From the historical business volume, obtaining the historical business volume of the target area every day during the normal observation period;

[0015] The business volume difference ratio sequence of the target holiday is calculated based on the historical business volume of each day in the holiday observation period and the historical business volume of each day in the normal observation period.

[0016] In some embodiments of the present application, determining the holiday observation time period and the normal observation time period corresponding to the target holiday includes:

[0017] Fill the target holiday with the first preset number of days in the front and the second preset number of days in the back, respectively, so that the total number of days reaches the target number of days, and obtain the holiday observation time period corresponding to the target holiday;

[0018] The first three preset days of each day in the festival observation time period are taken to obtain a normal observation time period of the same length as the festival observation time period.

[0019] In some embodiments of the present application, the calculation of the target holiday traffic volume difference ratio sequence according to the daily historical traffic volume in the holiday observation period and the daily historical traffic volume in the ordinary observation period includes:

[0020] Generate a first business volume sequence according to the historical business volume of each day in the festival observation period;

[0021] Generate a second business volume sequence according to the historical business volume of each day in the normal observation period;

[0022] Calculating a single-day difference ratio sequence corresponding to the target holiday according to the first business volume sequence and the second business volume sequence;

[0023] The business volume difference ratio sequence of the target holiday is calculated according to the single-day difference ratio sequence.

[0024] In some embodiments of the present application, calculating the target holiday business volume difference ratio sequence according to the single-day difference ratio sequence includes:

[0025] Obtaining the first total historical business volume in the festival observation time period;

[0026] Obtain the second total historical business volume during the normal observation period;

[0027] Calculating a total volume difference ratio corresponding to the target holiday according to the first total historical business volume and the second total historical business volume;

[0028] The business volume difference ratio sequence of the target holiday is calculated according to the single-day difference ratio sequence and the total volume difference ratio.

[0029] In some embodiments of the present application, predicting the business volume of the target area on the target holiday according to the business volume difference ratio sequence includes:

[0030] Determine, according to the target area and the target holiday, a target business volume prediction model for predicting the business volume of the target area on the target holiday;

[0031] The business volume difference ratio sequence is input into the target business volume prediction model to predict the business volume of the target area on the target holiday.

[0032] In some embodiments of the present application, before determining a target business volume prediction model for predicting the business volume of the target area on the target holiday according to the target area and the target holiday, the method further includes:

[0033] Perform holiday clustering on all holidays to divide them into multiple holiday categories;

[0034] Performing regional clustering on the area to be planned, so as to divide each area in the area to be planned into a plurality of regional categories, wherein the area to be planned includes the target area;

[0035] According to the plurality of holiday classifications and the plurality of area classifications, a plurality of holiday and city classifications are formed;

[0036] Collect historical business volume data of the multiple holidays and city classifications;

[0037] Based on the historical business volume data of the multiple holiday city classifications, a preset model is trained to obtain business volume prediction models for the multiple holiday cities.

[0038] On the other hand, the present application provides a traffic volume prediction device, the device comprising:

[0039] A determination unit, used to determine the target area and target holidays for the business volume to be predicted;

[0040] An acquisition unit, configured to acquire the historical business volume of the target area during the target holiday;

[0041] A calculation unit, configured to calculate a business volume difference ratio sequence of the target holiday according to the historical business volume;

[0042] A prediction unit is used to predict the business volume of the target area on the target holiday according to the business volume difference ratio sequence.

[0043] In some embodiments of the present application, the computing unit is specifically used for:

[0044] Determine the holiday observation time period and the normal observation time period corresponding to the target holiday;

[0045] From the historical business volume, obtaining the daily historical business volume of the target area during the festival observation period;

[0046] From the historical business volume, obtaining the historical business volume of the target area every day during the normal observation period;

[0047] The business volume difference ratio sequence of the target holiday is calculated based on the historical business volume of each day in the holiday observation period and the historical business volume of each day in the normal observation period.

[0048] In some embodiments of the present application, the computing unit is specifically used for:

[0049] Fill the target holiday with the first preset number of days in the front and the second preset number of days in the back, respectively, so that the total number of days reaches the target number of days, and obtain the holiday observation time period corresponding to the target holiday;

[0050] The first three preset days of each day in the festival observation time period are taken to obtain a normal observation time period of the same length as the festival observation time period.

[0051] In some embodiments of the present application, the computing unit is specifically used for:

[0052] Generate a first business volume sequence according to the historical business volume of each day in the festival observation period;

[0053] Generate a second business volume sequence according to the historical business volume of each day in the normal observation period;

[0054] Calculating a single-day difference ratio sequence corresponding to the target holiday according to the first business volume sequence and the second business volume sequence;

[0055] The business volume difference ratio sequence of the target holiday is calculated according to the single-day difference ratio sequence.

[0056] In some embodiments of the present application, the computing unit is specifically used for:

[0057] Obtaining the first total historical business volume in the festival observation time period;

[0058] Obtain the second total historical business volume during the normal observation period;

[0059] Calculating a total volume difference ratio corresponding to the target holiday according to the first total historical business volume and the second total historical business volume;

[0060] The business volume difference ratio sequence of the target holiday is calculated according to the single-day difference ratio sequence and the total volume difference ratio.

[0061] In some embodiments of the present application, the prediction unit is specifically used for:

[0062] Determine, according to the target area and the target holiday, a target business volume prediction model for predicting the business volume of the target area on the target holiday;

[0063] The business volume difference ratio sequence is input into the target business volume prediction model to predict the business volume of the target area on the target holiday.

[0064] In some embodiments of the present application, the apparatus further includes a training unit, and the training unit is specifically used to:

[0065] Before determining the target traffic volume prediction model for predicting the traffic volume of the target area on the target holiday according to the target area and the target holiday, performing holiday clustering on all holidays to divide the holidays into a plurality of holiday categories;

[0066] Performing regional clustering on the area to be planned, so as to divide each area in the area to be planned into a plurality of regional categories, wherein the area to be planned includes the target area;

[0067] According to the plurality of holiday classifications and the plurality of area classifications, a plurality of holiday and city classifications are formed;

[0068] Collect historical business volume data of the multiple holidays and city classifications;

[0069] Based on the historical business volume data of the multiple holiday city classifications, a preset model is trained to obtain business volume prediction models for the multiple holiday cities.

[0070] On the other hand, the present application further provides an electronic device, comprising a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the processor performs the following steps:

[0071] Determine the target area and target holidays for the business volume to be forecasted;

[0072] Obtain the historical business volume of the target area on the target holiday;

[0073] Calculate the business volume difference ratio sequence of the target holiday according to the historical business volume;

[0074] The business volume of the target area during the target holiday is predicted according to the business volume difference ratio sequence.

[0075] On the other hand, the present application also provides a computer-readable storage medium on which a computer program is stored, and the computer program is loaded by a processor to execute the steps in the traffic volume prediction method.

[0076] The embodiment of the present application determines the target area and target holiday for the business volume to be predicted; obtains the historical business volume of the target area on the target holiday; calculates the business volume difference ratio sequence of the target holiday based on the historical business volume; and predicts the business volume of the target area on the target holiday based on the business volume difference ratio sequence. Based on the fact that the prior art classifies the business volume by simple piece volume composition, which leads to limitations in the prediction of piece volume and inaccurate prediction of piece volume, the embodiment of the present application predicts the business volume by calculating the business volume difference ratio sequence of the target area on the target holiday. This process predicts the business volume based on the difference ratio data of a certain holiday in a certain area. Compared with simply observing the number of pieces in each period, this difference ratio sequence corresponding to each holiday can better outline the consumption habits of a region in the face of a certain type of holiday stimulation, making the business volume prediction closer to the actual scene and improving the accuracy of the business volume prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0078] Figure 1 is a scenario schematic diagram of a business volume prediction system provided by an embodiment of the present invention;

[0079] Figure 2 is a schematic diagram of an embodiment of a flow chart of a method for predicting business volume provided in an embodiment of the present invention;

[0080] Figure 3is a schematic flow chart of an embodiment of step 203 in an embodiment of the present invention;

[0081] Figure 4 It is a flowchart of an embodiment of predicting the business volume of a target area on the target holiday according to a business volume difference ratio sequence provided in an embodiment of the present invention;

[0082] Figure 5 It is a schematic diagram of the training process of the business volume prediction model for multiple holiday cities in an embodiment of the present invention;

[0083] Figure 6 is a schematic diagram of an embodiment scenario of festival clustering and city clustering in an embodiment of the present invention;

[0084] Figure 7 is a schematic diagram of the structure of an embodiment of a traffic volume prediction device provided in an embodiment of the present invention;

[0085] Figure 8 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0086] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0087] In the following description, the specific embodiments of the present invention will be described with reference to the steps and symbols executed by one or more computers, unless otherwise specified. Therefore, these steps and operations will be mentioned several times as being executed by a computer, and the computer execution referred to herein includes the operation of a computer processing unit by an electronic signal representing data in a structured form. This operation converts the data or maintains it at a location in the memory system of the computer, which can be reconfigured or otherwise change the operation of the computer in a manner familiar to testers in the field. The data structure maintained by the data is a physical location in the memory, which has specific characteristics defined by the data format. However, the principles of the present invention are described in the above text, which does not represent a limitation, and testers in the field will understand that the various steps and operations described below can also be implemented in hardware.

[0088] The term "module" or "unit" used herein can be regarded as a software object executed on the computing system. The different components, modules, engines and services described herein can be regarded as implementation objects on the computing system. The apparatus and method described herein are preferably implemented in software, but can also be implemented in hardware, all within the scope of protection of the present invention.

[0089] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be an intermediate element. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein includes all or any unit and all combinations of one or more associated listed items.

[0090] Embodiments of the present invention provide a traffic volume prediction method, device, server and storage medium.

[0091] See also Figure 1 , Figure 1 Schematic diagram of a scenario of a business volume prediction system provided in an embodiment of the present invention. The business volume prediction system in the embodiment of the present invention can be implemented based on a business platform, such as a logistics service platform. The business volume prediction system may include an electronic device 100, in which a business volume prediction device is integrated.

[0092] In the embodiment of the present invention, the electronic device 100 is mainly used to determine the target area and target holiday for the business volume to be predicted; obtain the historical business volume of the target area on the target holiday; calculate the business volume difference ratio sequence of the target holiday based on the historical business volume; and predict the business volume of the target area on the target holiday based on the business volume difference ratio sequence.

[0093] In an embodiment of the present invention, the server may be an independent server, or a server network or server cluster composed of servers. For example, the server described in the embodiment of the present invention includes but is not limited to a computer, a network host, a single network server, a plurality of network server sets or a cloud server composed of a plurality of servers. Among them, the cloud server is composed of a large number of computers or network servers based on cloud computing. In an embodiment of the present invention, communication between the server and the client can be achieved through any communication method, including but not limited to mobile communications based on the 3rd Generation Partnership Project (3GPP), Long Term Evolution (LTE), Worldwide Interoperability for Microwave Access (WiMAX), or computer network communications based on TCP / IP Protocol Suite (TCP / IP), User Datagram Protocol (UDP), etc.

[0094] Those skilled in the art will understand that Figure 1 The application environment shown in the figure is only one application scenario of the present application solution and does not constitute a limitation on the application scenario of the present application solution. Other application environments may also include Figure 1 More or fewer electronic devices as shown in, e.g. Figure 1 Only one electronic device is shown. It can be understood that the traffic volume prediction system can also include one or more other services, which are not specifically limited here.

[0095] In addition, if Figure 1 As shown, the business volume prediction system may also include a memory 200 for storing data, such as logistics data, for example, various data of the logistics platform, such as logistics transportation information of the transfer yard, specifically, express information, delivery vehicle information and logistics network information.

[0096] It should be noted that Figure 1 The scenario diagram of the business volume prediction system shown is only an example. The business volume prediction system and scenario described in the embodiment of the present invention are intended to more clearly illustrate the technical solution of the embodiment of the present invention, and do not constitute a limitation on the technical solution provided by the embodiment of the present invention. Ordinary technicians in this field can know that with the evolution of the business volume prediction system and the emergence of new business scenarios, the technical solution provided by the embodiment of the present invention is also applicable to similar technical problems.

[0097] The following describes it in detail with reference to specific embodiments.

[0098] In this embodiment, description will be made from the perspective of a traffic volume prediction device, which may be specifically integrated into the electronic device 100 .

[0099] The present invention provides a business volume prediction method, which includes: determining a target area and a target holiday for the business volume to be predicted; obtaining the historical business volume of the target area on the target holiday; calculating a business volume difference ratio sequence for the target holiday based on the historical business volume; and predicting the business volume of the target area on the target holiday based on the business volume difference ratio sequence.

[0100] See also Figure 2 , is a flow chart of an embodiment of a method for predicting traffic volume in an embodiment of the present invention, the method for predicting traffic volume includes:

[0101] 201. Determine the target area and target holidays for the business volume to be predicted.

[0102] Among them, the business volume to be predicted may refer to the volume corresponding to the business to be predicted. In the embodiment of the present invention, the business volume may be the volume of pieces, specifically, the volume of express pieces. It should be noted that in some embodiments of the present invention, it may be further limited to the volume of received pieces or the volume of sent pieces. It is understandable that in other embodiments of the present invention, the business volume may also be the rental and sales volume of other products, such as the sales volume of a target product, specifically, the sales volume of model A air conditioner in city B, which is not limited here.

[0103] In addition, the target area refers to the area where the business volume needs to be predicted. The target area can be based on one or more countries, or only include one country; the target area can also be one or more provinces in China, or only include one province; the target area can also be one or more cities; the target area can also be one or more municipal areas; the target area can also be one or more streets or towns under the municipal area. It can be understood that the above target areas are only examples, and the actual size of the target area can be set according to the actual scenario, and is not limited here.

[0104] The target holiday may be a holiday in a year, and may be any holiday in any country according to actual application needs, such as China's Mid-Autumn Festival or Western Easter, etc., which is not limited here.

[0105] 202. Obtain the historical business volume of the target area during the target holiday.

[0106] Specifically, the historical business volume of the target area on the target holiday may be the historical business volume of the target area for a preset number of days before the target holiday, for example, the historical business volume of the previous year, or one and a half years or two years, or all the saved historical business volumes, which are not specifically limited here. Since the business platform continuously generates business data during the business process, for example, the logistics platform continuously generates logistics data, which is stored in the corresponding storage, for example Figure 1 In the memory 200 shown in , the electronic device 100 can now directly obtain the historical business volume of the target area on the target holiday from the memory 200.

[0107] 203. Calculate a business volume difference ratio sequence of the target holiday based on the historical business volume.

[0108] The traffic volume difference ratio sequence is a sequence generated by calculating the traffic volume difference ratio according to the historical traffic volume of multiple preset observation time periods. Figure 3 As shown, the calculating of the business volume difference ratio sequence of the target holiday according to the historical business volume may include:

[0109] 301. Determine a holiday observation time period and a normal observation time period corresponding to the target holiday.

[0110] The step of determining the holiday observation time period and the normal observation time period corresponding to the target holiday includes:

[0111] (1) The target holiday is filled forward by a first preset number of days and backward by a second preset number of days, so that the total number of days reaches the target number of days, thereby obtaining the holiday observation time period corresponding to the target holiday.

[0112] Specifically, the target number of days can be set as needed, for example, the target number of days is 10 days, 14 days or 21 days, etc. It should be noted that the inventor has determined based on actual tests that it is easier to calculate when the target number of days is an integer multiple of the number of days per week (7 days). Therefore, in an embodiment of the present invention, the target number of days is an integer multiple of the number of days per week (7 days), such as 14 days.

[0113] In the embodiment of the present invention, it is necessary to satisfy: target holiday number of days + first preset number of days + second preset number of days = target number of days, and the time period corresponding to the target number of days is the holiday observation time period corresponding to the target holiday.

[0114] In a specific embodiment, assuming that the first preset number of days is Pf, the second preset number of days is Pt, and the target number of days is 14 days, then:

[0115] Target holiday: H{h1,h2,...,hi,...,h14}

[0116] The holiday observation period corresponding to the target holiday: Pf+H+Pt

[0117] For example, the Mid-Autumn Festival holiday is 3 days (9 / 11-9 / 13), then H is {h1=9-11, h2=9-12, h3=9-13}.

[0118] Front Padding First Preset Days: Pf

[0119] Tail Padding Second preset day: Pt

[0120] Pf and Pt days are filled before and after the normal target holiday period H, respectively, so that the total length of the holiday observation period is 14 days (two weeks).

[0121] Filling needs to meet the following conditions:

[0122] Pf+H+Pt=14

[0123] The time period corresponding to the target holiday should be within the 14-day observation period:

[0124] Pf = Floor ((14-H) / 2) (round down);

[0125] Pt = Cell ((14-H) / 2) (round up);

[0126] For example, the Mid-Autumn Festival holiday is 3 days (9 / 11-9 / 13), then Pf = Floor ((14-3) / 2) = 5 days, that is, 9 / 6-9 / 10;

[0127] Pt = Cell ((14-3) / 2) = 6 days, i.e. 9 / 14-9 / 19.

[0128] In specific applications, when a holiday is filled forward, other holidays may be encountered when filling backward. At this time, other holidays may affect the business volume of the target holiday. Therefore, this situation needs to be excluded during the forecast. Specifically,

[0129] If Pf and Pt include other holidays, Pf will use the forward one week to get the number of days for the dates that overlap with the holidays, and Pt will use the backward one week to get the number of days for the dates that overlap with the holidays. For example, the Mid-Autumn Festival holiday is 9 / 26-9 / 28, and the normal Pt (9 / 29-10 / 4) will include the National Day (10 / 1-10 / 7), then Pt will normally take 9 / 29-9 / 30, and then use 10 / 8-10 / 11 to replace the overlapping dates of 10 / 1-10 / 4. The final value of Pt will be 9 / 29-9 / 30+10 / 8-10 / 11. The purpose of this is to eliminate the interference of other holidays as much as possible and better analyze the impact of the target holiday on the business volume.

[0130] (2) Taking the first three preset days of each day in the festival observation time period, a normal observation time period of the same length as the festival observation time period is obtained.

[0131] In the embodiment of the present invention, the third preset number of days can be directly pushed forward for each day in the holiday observation period to obtain a normal observation period of the same length as the holiday observation period. In this case, the normal observation period is the same length as the holiday observation period, for example, both are 14 days. In addition, the third preset number of days can be taken according to the actual scenario, and holidays are avoided as much as possible. Similarly, the third preset number of days is preferably an integer multiple of the number of days per week (7 days), and the intervals are not too close as much as possible, for example, 42 days (6 weeks).

[0132] In a specific embodiment, it is assumed that the normal observation period (Normal) is also 14 days, specifically: N{n1,n2,...,ni,...,n14}

[0133] After the holiday observation time period corresponding to the target holiday is filled, the corresponding date 6 weeks before each date (42 days ago) is taken to form the normal observation time period, which is a total of 14 days, the same as the holiday observation time period.

[0134] The purpose is to select a set of fluctuating samples of normal business volume for comparison with the target holidays.

[0135] Selection condition: ni = hi-42

[0136] Similarly, if there is a holiday in N, move forward 2 weeks (which can be adjusted according to the actual situation, preferably an integer multiple of a week), that is, ni = hi-56. If there is still a holiday after moving forward two weeks, move forward two weeks again, and so on.

[0137] In other embodiments of the present invention, in addition to taking the number of days in two weeks for the data in the normal observation time period, the data can also be selected as the average of the dates of each week in the first six months, that is, the average of all Mondays in half a year, the average of all Tuesdays, and so on.

[0138] 302. Obtain the daily historical business volume of the target area during the holiday observation period from the historical business volume.

[0139] The historical business volume includes the daily historical business volume of the target area during the festival observation period, so the daily historical business volume of the target area during the festival observation period can be obtained from the historical business volume.

[0140] 303. Obtain the daily historical business volume of the target area during the normal observation period from the historical business volume.

[0141] The historical business volume includes the historical business volume of the target area every day in the normal observation period, so the historical business volume of the target area every day in the normal observation period can be obtained from the historical business volume.

[0142] 304. Calculate a business volume difference ratio sequence of the target holiday according to the historical business volume of each day in the holiday observation period and the historical business volume of each day in the normal observation period.

[0143] Among them, the calculation of the business volume difference ratio sequence of the target holiday based on the historical business volume of each day in the holiday observation time period and the historical business volume of each day in the ordinary observation time period includes: generating a first business volume sequence based on the historical business volume of each day in the holiday observation time period; generating a second business volume sequence based on the historical business volume of each day in the ordinary observation time period; calculating the single-day difference ratio sequence corresponding to the target holiday based on the first business volume sequence and the second business volume sequence; calculating the business volume difference ratio sequence of the target holiday based on the single-day difference ratio sequence.

[0144] In a specific embodiment, the daily difference ratio sequence: D{d1,d2,...,di,...,d14}

[0145] The calculation formula for each value in D is: di = (traffic volume (hi) - traffic volume (ni)) / traffic volume (ni), 1≤i≤14;

[0146] The benefits of using this dimension are as follows:

[0147] (1) Observe the duration of the impact of holidays on business volume (by looking at the day when the absolute value starts to increase, we can know how many days in advance the holiday will begin to affect the area and how long the impact will last)

[0148] (2) Observe the magnitude of the impact (the magnitude of each d value can be used to determine the impact of the holiday on the business volume at each time period, or the sensitivity of the region to the type of holiday)

[0149] (3) Observe the impact type (the fluctuation of the d value can show the impact type of this festival)

[0150] Not affected by the volume of business (because what is observed is the proportion, so large business volume may be grouped together with small business volume, as long as the consumption habits of the two areas are similar).

[0151] Furthermore, the business volume difference ratio sequence of the target holiday is calculated based on the single-day difference ratio sequence, including: obtaining a first total historical business volume in the holiday observation time period; obtaining a second total historical business volume in the normal observation time period; calculating the total amount difference ratio corresponding to the target holiday based on the first total historical business volume and the second total historical business volume; calculating the business volume difference ratio sequence of the target holiday based on the single-day difference ratio sequence and the total amount difference ratio.

[0152] According to the first total historical business volume and the second total historical business volume, the total difference ratio corresponding to the target holiday is calculated as follows in a specific implementation:

[0153] Total difference ratio: C

[0154] C = (first total historical business volume - second total historical business volume) / second total historical business volume

[0155] The purpose of this dimension is to observe whether the holiday increases / decreases the total business volume or just changes the time of business volume distribution.

[0156] If it is observed in subsequent analysis or review that C has a strong positive impact on effective clustering, then dimensionality enhancement of C can also be considered.

[0157] After determining the single-day difference ratio sequence and the total amount difference ratio, the business volume difference ratio sequence of the target holiday can be calculated according to the single-day difference ratio sequence and the total amount difference ratio.

[0158] Taking the daily difference ratio sequence as D={d1,d2,...,di,...,d14} and the total difference ratio as C as an example, the daily difference ratio sequence plus the total difference ratio, i.e. d1,d2,...,d14, plus C, we get a 15-dimensional business volume difference ratio sequence.

[0159] 204. Predict the business volume of the target area on the target holiday according to the business volume difference ratio sequence.

[0160] The embodiment of the present application determines the target area and target holiday for the business volume to be predicted; obtains the historical business volume of the target area on the target holiday; calculates the business volume difference ratio sequence of the target holiday based on the historical business volume; and predicts the business volume of the target area on the target holiday based on the business volume difference ratio sequence. Based on the fact that the prior art classifies the business volume by simple piece volume composition, which leads to limitations in the prediction of piece volume and inaccurate prediction of piece volume, the embodiment of the present application predicts the business volume by calculating the business volume difference ratio sequence of the target area on the target holiday. This process predicts the business volume based on the difference ratio data of a certain holiday in a certain area. Compared with simply observing the number of pieces in each period, this difference ratio sequence corresponding to each holiday can better outline the consumption habits of a region in the face of a certain type of holiday stimulation, making the business volume prediction closer to the actual scene and improving the accuracy of the business volume prediction.

[0161] In other embodiments of the present invention, Figure 4 As shown, predicting the business volume of the target area on the target holiday according to the business volume difference ratio sequence may further include:

[0162] 401. Determine a target business volume prediction model for predicting the business volume of the target area on the target holiday according to the target area and the target holiday.

[0163] Among them, the target business volume prediction model can be a neural network model, such as a convolutional neural network (CNN) model, a recurrent neural network (RNN) model, etc.

[0164] 402. Input the business volume difference ratio sequence into the target business volume prediction model to predict the business volume of the target area on the target holiday.

[0165] It is understandable that before determining the target business volume prediction model for predicting the business volume of the target area on the target holiday according to the target area and the target holiday in step 401, it is necessary to first train the target business volume prediction model.

[0166] Specifically, in the embodiment of the present invention, Figure 5 As shown, before determining a target business volume prediction model for predicting the business volume of the target area on the target holiday according to the target area and the target holiday, the method further includes:

[0167] 501. Perform holiday clustering on all holidays to divide the holidays into multiple holiday categories.

[0168] In the embodiment of the present invention, all holidays may refer to all holidays in a preset area within a natural year, such as all holidays in China within a natural year.

[0169] Specifically, all holidays can be clustered based on holiday business volume data. A holiday classification model can be used to cluster all holidays. The holiday classification model can consider the pre-trained fuzzy c-means clustering (FCM) model or K-means model. The goal is to divide all holidays into several major categories. If two categories are taken, generally holidays with a large increase in business volume will be clustered together, and holidays with a small impact on business volume will be clustered together. If more categories are taken, the holiday classification model may also subdivide holidays according to other characteristics.

[0170] In general, if the sample size of holidays is small, then fewer categories are selected in the embodiment of the present invention, so that sufficient data learning features can be ensured under each holiday category. If the sample size of holidays is large, then more categories can be selected to ensure that the holidays grouped together are highly learnable. Of course, the holiday categories cannot exceed the total number of holidays.

[0171] 502. Perform regional clustering in the area to be planned to divide each area in the area to be planned into a plurality of regional categories, wherein the area to be planned includes the target area.

[0172] After the holiday clustering, the performance of each region in 15 dimensions is clustered using the regional clustering model for a certain type of holiday business volume data. The purpose of this is to screen out regions that respond similarly to similar consumer stimuli. Similarly, this regional clustering can provide more valuable data combinations for subsequent models. At the same time, it also avoids the problems of insufficient data or excessive model size that may be faced by modeling with a single region. Similarly, the regional clustering model can also use a pre-trained FCM model or K-Mean model.

[0173] 503. According to the plurality of holiday classifications and the plurality of area classifications, a plurality of holiday and city classifications are formed.

[0174] After the holiday and region dual clustering is completed, multiple holiday and city classifications can be formed based on the multiple holiday classifications and the multiple region classifications. Figure 6 As shown in Figure 2, assuming that holidays are festivals and regions are cities, after the double clustering of holidays and regions is completed, it can be as follows Figure 6 As shown, it is divided into Class A festival type 1 city, Class A festival type 2 city, etc.

[0175] 504. Collect historical business volume data classified by the plurality of holidays and cities.

[0176] In an embodiment of the present invention, the historical business volume data of the multiple holidays and city classifications are collected directly through the business platform. For example, the business platform stores historical business volume data during operation, and the historical business volume data of the multiple holidays and city classifications can be collected by accessing the corresponding storage of the business platform.

[0177] 505. Based on the historical business volume data of the multiple holiday city classifications, a preset model is trained to obtain business volume prediction models for the multiple holiday cities.

[0178] Specifically, the business volume data divided into festival-city categories are modeled using machine learning methods (obtained by training the initial machine learning model), so that each "festival category-city category" has a prediction model that has learned this type of data, such as Figure 6 The “Class A Festival Class 1 City Model” shown, in which the initial machine learning model can be the CNN model or RNN model described in the above embodiments, etc.

[0179] In some embodiments of the present invention, after the holiday clustering and regional clustering are completed, the distribution ratios thereof are recorded, which are used to determine the weight ratios of the results of each model in subsequent predictions. For example, 60% of the data of a certain holiday is classified into Class A, 30% into Class B, and 10% into Class C. 70% of the data of a certain city is classified into Class 1, 30% into Class 2, and so on.

[0180] In other embodiments of the present invention, since different holiday start dates have different impacts on business volume, different week start date weights may be used. The week start date weights may include: official start times Monday, Tuesday to Sunday, seven types, and a holiday in the previous week and a holiday in the next week, a total of nine types.

[0181] The same type of holidays start at different times of the week, and the impact on the business volume during the holidays is also different. For example, when some holidays start in the middle of the week, people's consumption habits are different from those of holidays that start on the weekend. In addition, if two holidays are close to each other (such as Mid-Autumn Festival and National Day), the same city will also have different consumption curves. Therefore, it is necessary to cluster the holiday data together and then classify them according to the time of the holiday start week. By comparing the growth ratio of these nine situations compared to the holiday average growth baseline, a weight is determined. For example, in the B-type holiday cluster, the official second day of the B-type holiday has a growth ratio of y compared to weekdays. When the official start time of the B-type holiday is Friday, the official second day has a growth ratio of y' compared to weekdays. Then the weight of the B-type holiday starting on Friday is ((y' / y)+c-1) / c. Here c is a coefficient we set to adjust the impact of the weight on the final result. The value is between 1 and positive infinity. The larger c is, the closer the weight will be to 1, and the smaller the impact on the overall prediction model will be.

[0182] To facilitate better implementation of the traffic volume prediction method provided by the embodiment of the present invention, the embodiment of the present invention also provides a device based on the above traffic volume prediction method. The meanings of the terms are the same as those in the above traffic volume prediction method, and the specific implementation details can refer to the description in the method embodiment.

[0183] See also Figure 7 , Figure 7 A schematic diagram of the structure of a traffic volume prediction device provided by an embodiment of the present invention, wherein the traffic volume prediction device 700 may include a determination unit 701, an acquisition unit 702, a calculation unit 703 and a prediction unit 704, wherein:

[0184] A determination unit 701, used to determine a target area and a target holiday for the traffic volume to be predicted;

[0185] An acquisition unit 702 is used to acquire the historical business volume of the target area on the target holiday;

[0186] A calculation unit 703 is used to calculate a business volume difference ratio sequence of the target holiday according to the historical business volume;

[0187] The prediction unit 704 is used to predict the business volume of the target area on the target holiday according to the business volume difference ratio sequence.

[0188] In some embodiments of the present application, the calculation unit 703 is specifically used to:

[0189] Determine the holiday observation time period and the normal observation time period corresponding to the target holiday;

[0190] From the historical business volume, obtaining the daily historical business volume of the target area during the festival observation period;

[0191] From the historical business volume, obtaining the historical business volume of the target area every day during the normal observation period;

[0192] The business volume difference ratio sequence of the target holiday is calculated based on the historical business volume of each day in the holiday observation period and the historical business volume of each day in the normal observation period.

[0193] In some embodiments of the present application, the calculation unit 703 is specifically used to:

[0194] Fill the target holiday with the first preset number of days in the front and the second preset number of days in the back, respectively, so that the total number of days reaches the target number of days, and obtain the holiday observation time period corresponding to the target holiday;

[0195] The first three preset days of each day in the festival observation time period are taken to obtain a normal observation time period of the same length as the festival observation time period.

[0196] In some embodiments of the present application, the calculation unit 703 is specifically used to:

[0197] Generate a first business volume sequence according to the historical business volume of each day in the festival observation period;

[0198] Generate a second business volume sequence according to the historical business volume of each day in the normal observation period;

[0199] Calculating a single-day difference ratio sequence corresponding to the target holiday according to the first business volume sequence and the second business volume sequence;

[0200] The business volume difference ratio sequence of the target holiday is calculated according to the single-day difference ratio sequence.

[0201] In some embodiments of the present application, the calculation unit 703 is specifically used to:

[0202] Obtaining the first total historical business volume in the festival observation time period;

[0203] Obtain the second total historical business volume during the normal observation period;

[0204] Calculating a total volume difference ratio corresponding to the target holiday according to the first total historical business volume and the second total historical business volume;

[0205] The business volume difference ratio sequence of the target holiday is calculated according to the single-day difference ratio sequence and the total volume difference ratio.

[0206] In some embodiments of the present application, the prediction unit 704 is specifically used for:

[0207] Determine, according to the target area and the target holiday, a target business volume prediction model for predicting the business volume of the target area on the target holiday;

[0208] The business volume difference ratio sequence is input into the target business volume prediction model to predict the business volume of the target area on the target holiday.

[0209] In some embodiments of the present application, the apparatus further includes a training unit, and the training unit is specifically used to:

[0210] Before determining the target traffic volume prediction model for predicting the traffic volume of the target area on the target holiday according to the target area and the target holiday, performing holiday clustering on all holidays to divide the holidays into a plurality of holiday categories;

[0211] Performing regional clustering on the area to be planned, so as to divide each area in the area to be planned into a plurality of regional categories, wherein the area to be planned includes the target area;

[0212] According to the plurality of holiday classifications and the plurality of area classifications, a plurality of holiday and city classifications are formed;

[0213] Collect historical business volume data of the multiple holidays and city classifications;

[0214] Based on the historical business volume data of the multiple holiday city classifications, a preset model is trained to obtain business volume prediction models for the multiple holiday cities.

[0215] In the embodiment of the present application, the target area and target holiday of the business volume to be predicted are determined by the determination unit 701; the acquisition unit 702 acquires the historical business volume of the target area on the target holiday; the calculation unit 703 calculates the business volume difference ratio sequence of the target holiday based on the historical business volume; the prediction unit 704 predicts the business volume of the target area on the target holiday based on the business volume difference ratio sequence. In the embodiment of the present application, based on the prior art that classifies by simple piece volume composition, resulting in limitations in piece volume prediction and inaccurate piece volume prediction, the business volume prediction is achieved by calculating the business volume difference ratio sequence of the target area on the target holiday. This process predicts the business volume based on the difference ratio data of a certain holiday in a certain area. Compared with simply observing the number of pieces in each period, this difference ratio sequence corresponding to each holiday can better outline the consumption habits of a region facing a certain type of holiday stimulation, making the business volume prediction closer to the actual scene and improving the accuracy of the business volume prediction.

[0216] The embodiment of the present invention further provides an electronic device, such as Figure 8 As shown, it shows a schematic diagram of the structure of an electronic device involved in an embodiment of the present invention, specifically:

[0217] The electronic device may include components such as a processor 801 with one or more processing cores, a memory 802 with one or more computer-readable storage media, a power supply 803, and an input unit 804. Those skilled in the art will appreciate that Figure 8 The electronic device structure shown in the figure does not constitute a limitation on the electronic device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0218] The processor 801 is the control center of the electronic device, which uses various interfaces and lines to connect various parts of the entire electronic device, and executes various functions of the electronic device and processes data by running or executing software programs and / or modules stored in the memory 802, and calling data stored in the memory 802, so as to monitor the electronic device as a whole. Optionally, the processor 801 may include one or more processing cores; preferably, the processor 801 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operation storage medium, user interface and application program, etc., and the modem processor mainly processes wireless communication. It is understandable that the above-mentioned modem processor may not be integrated into the processor 801.

[0219] The memory 802 can be used to store software programs and modules. The processor 801 executes various functional applications and data processing by running the software programs and modules stored in the memory 802. The memory 802 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating storage medium, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 802 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory 802 may also include a memory controller to provide the processor 801 with access to the memory 802.

[0220] The electronic device also includes a power supply 803 for supplying power to each component. Preferably, the power supply 803 can be logically connected to the processor 801 through a power management storage medium, so as to manage charging, discharging, and power consumption management through the power management storage medium. The power supply 803 can also include any components such as one or more DC or AC power supplies, recharge storage media, power failure detection circuits, power converters or inverters, and power status indicators.

[0221] The electronic device may further include an input unit 804, which may be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

[0222] Although not shown, the electronic device may further include a display unit, etc., which will not be described in detail herein. Specifically in this embodiment, the processor 801 in the electronic device will load the executable files corresponding to the processes of one or more application programs into the memory 802 according to the following instructions, and the processor 801 will run the application programs stored in the memory 802, thereby realizing various functions, as follows:

[0223] Determine a target area and a target holiday for the business volume to be predicted; obtain the historical business volume of the target area on the target holiday; calculate a business volume difference ratio sequence for the target holiday based on the historical business volume; and predict the business volume of the target area on the target holiday based on the business volume difference ratio sequence.

[0224] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0225] To this end, an embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored, and the computer program is loaded by a processor to execute the steps in any one of the traffic volume prediction methods provided by the embodiments of the present invention. For example, the computer program can be loaded by a processor to execute the following steps:

[0226] Determine a target area and a target holiday for the business volume to be predicted; obtain the historical business volume of the target area on the target holiday; calculate a business volume difference ratio sequence for the target holiday based on the historical business volume; and predict the business volume of the target area on the target holiday based on the business volume difference ratio sequence.

[0227] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0228] The computer-readable storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0229] Since the computer program stored in the computer-readable storage medium can execute the steps in any one of the business volume prediction methods provided in the embodiments of the present invention, the beneficial effects that can be achieved by any one of the business volume prediction methods provided in the embodiments of the present invention can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0230] The above is a detailed introduction to a business volume prediction method, device, electronic device and storage medium provided by an embodiment of the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for technical personnel in this field, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A traffic volume prediction method, characterized in that: The method comprises: Determine the target area and target holidays for the business volume to be forecasted; Obtain the historical business volume of the target area on the target holiday; Calculate the business volume difference ratio sequence of the target holiday according to the historical business volume; According to the business volume difference ratio sequence, predict the business volume of the target area on the target holiday; The step of calculating the business volume difference ratio sequence of the target holiday according to the historical business volume includes: Determine the holiday observation time period and the normal observation time period corresponding to the target holiday; From the historical business volume, obtaining the daily historical business volume of the target area during the festival observation period; From the historical business volume, obtaining the historical business volume of the target area every day during the normal observation period; Calculate the business volume difference ratio sequence of the target holiday according to the historical business volume of each day in the holiday observation period and the historical business volume of each day in the normal observation period; The step of determining the holiday observation time period and the normal observation time period corresponding to the target holiday includes: Fill the target holiday with the first preset number of days in the front and the second preset number of days in the back, so that the total number of days reaches the target number of days, and obtain the holiday observation time period corresponding to the target holiday; Each day in the festival observation time period is pushed forward by a third preset number of days to obtain a normal observation time period of the same length as the festival observation time period.

2. The traffic volume prediction method according to claim 1, characterized in that: The calculating of the target holiday business volume difference ratio sequence according to the daily historical business volume in the holiday observation period and the daily historical business volume in the normal observation period includes: Generate a first business volume sequence according to the historical business volume of each day in the festival observation period; Generate a second business volume sequence according to the historical business volume of each day in the normal observation period; Calculating a single-day difference ratio sequence corresponding to the target holiday according to the first business volume sequence and the second business volume sequence; The business volume difference ratio sequence of the target holiday is calculated according to the single-day difference ratio sequence.

3. The traffic volume prediction method according to claim 2, characterized in that: Calculating the target holiday business volume difference ratio sequence according to the single-day difference ratio sequence includes: Obtaining the first total historical business volume in the festival observation time period; Obtain the second total historical business volume during the normal observation period; Calculating a total volume difference ratio corresponding to the target holiday according to the first total historical business volume and the second total historical business volume; The business volume difference ratio sequence of the target holiday is calculated according to the single-day difference ratio sequence and the total volume difference ratio.

4. The traffic volume prediction method according to any one of claims 1 to 3, characterized in that: The predicting the business volume of the target area on the target holiday according to the business volume difference ratio sequence includes: Determine, according to the target area and the target holiday, a target business volume prediction model for predicting the business volume of the target area on the target holiday; The business volume difference ratio sequence is input into the target business volume prediction model to predict the business volume of the target area on the target holiday.

5. The traffic volume prediction method according to claim 1, characterized in that: Before determining a target business volume prediction model for predicting the business volume of the target area on the target holiday according to the target area and the target holiday, the method further includes: Perform holiday clustering on all holidays to divide them into multiple holiday categories; Performing regional clustering on the area to be planned, so as to divide each area in the area to be planned into a plurality of regional categories, wherein the area to be planned includes the target area; According to the plurality of holiday classifications and the plurality of area classifications, a plurality of holiday and city classifications are formed; Collect historical business volume data of the multiple holidays and city classifications; Based on the historical business volume data of the multiple holidays and city classifications, a preset model is trained to obtain a business volume prediction model for the multiple holidays and city classifications.

6. A traffic volume prediction device, characterized in that: The device comprises: A determination unit, used to determine the target area and target holidays for the business volume to be predicted; An acquisition unit, configured to acquire the historical business volume of the target area during the target holiday; A calculation unit, configured to calculate a business volume difference ratio sequence of the target holiday according to the historical business volume; A prediction unit, configured to predict the business volume of the target area on the target holiday according to the business volume difference ratio sequence; The calculation unit is further used to determine the holiday observation time period and the normal observation time period corresponding to the target holiday; From the historical business volume, obtaining the daily historical business volume of the target area during the festival observation period; From the historical business volume, obtaining the historical business volume of the target area every day during the normal observation period; Calculate the business volume difference ratio sequence of the target holiday according to the historical business volume of each day in the holiday observation period and the historical business volume of each day in the normal observation period; The calculation unit is further used to fill the target holiday with a first preset number of days in the front and a second preset number of days in the back, so that the total number of days reaches the target number of days, and obtain the holiday observation time period corresponding to the target holiday; Each day in the festival observation time period is pushed forward by a third preset number of days to obtain a normal observation time period of the same length as the festival observation time period.

7. An electronic device, characterized in that: The invention comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the following steps: Determine the target area and target holidays for the business volume to be forecasted; Obtain the historical business volume of the target area on the target holiday; Calculate the business volume difference ratio sequence of the target holiday according to the historical business volume; According to the business volume difference ratio sequence, predict the business volume of the target area on the target holiday; The step of calculating the business volume difference ratio sequence of the target holiday according to the historical business volume includes: Determine the holiday observation time period and the normal observation time period corresponding to the target holiday; From the historical business volume, obtaining the daily historical business volume of the target area during the festival observation period; From the historical business volume, obtaining the historical business volume of the target area every day during the normal observation period; Calculate the business volume difference ratio sequence of the target holiday according to the historical business volume of each day in the holiday observation period and the historical business volume of each day in the normal observation period; The step of determining the holiday observation time period and the normal observation time period corresponding to the target holiday includes: Fill the target holiday with the first preset number of days in the front and the second preset number of days in the back, so that the total number of days reaches the target number of days, and obtain the holiday observation time period corresponding to the target holiday; Each day in the festival observation time period is pushed forward by a third preset number of days to obtain a normal observation time period of the same length as the festival observation time period.

8. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and the computer program is loaded by a processor to execute the steps in the traffic volume prediction method according to any one of claims 1 to 5.

Images