WO2023246954A1 - 物品推荐方法、装置、电子设备及存储介质 - Google Patents

物品推荐方法、装置、电子设备及存储介质 Download PDF

Info

Publication number
WO2023246954A1
WO2023246954A1 PCT/CN2023/114022 CN2023114022W WO2023246954A1 WO 2023246954 A1 WO2023246954 A1 WO 2023246954A1 CN 2023114022 W CN2023114022 W CN 2023114022W WO 2023246954 A1 WO2023246954 A1 WO 2023246954A1
Authority
WO
WIPO (PCT)
Prior art keywords
item
shipping
candidate
volume
candidate item
Prior art date
Application number
PCT/CN2023/114022
Other languages
English (en)
French (fr)
Inventor
李博涵
曾庆维
陈桓
Original Assignee
顺丰科技有限公司
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 顺丰科技有限公司 filed Critical 顺丰科技有限公司
Publication of WO2023246954A1 publication Critical patent/WO2023246954A1/zh

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2457—Query processing with adaptation to user needs
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2462—Approximate or statistical queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/083—Shipping
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Definitions

  • This application relates to the technical field of item recommendation, specifically to an item recommendation method, device, electronic equipment and storage medium.
  • This application provides an item recommendation method, device, electronic equipment and storage medium, aiming to solve the problem of low recognition accuracy of current item recommendation methods.
  • this application provides an item recommendation method, including:
  • a target item to be recommended is selected from the candidate items.
  • determining the shipping fluctuation value of each candidate item based on the shipping time and shipping volume of each candidate item in the logistics information includes:
  • the shipping volume of each candidate item is divided to obtain the target shipping volume of each candidate item and the history of each candidate item. Shipping volume;
  • the shipping fluctuation value of each candidate item is calculated based on the target shipping volume of each candidate item and the historical shipping volume of each candidate item.
  • the global fluctuation value of each candidate item is determined based on the shipping fluctuation value of each candidate item and the average shipping volume of candidate items in the logistics information.
  • the shipping fluctuation value of each candidate item is weighted to obtain the global fluctuation value of each candidate item.
  • selecting a target item to be recommended from each candidate item based on the global fluctuation value of each candidate item includes:
  • Candidate items with global fluctuation values greater than the preset score threshold are set as target items to be recommended.
  • the method before determining the shipping fluctuation value of each candidate item based on the shipping time and shipping volume of each candidate item in the logistics information, the method further include:
  • the first item whose average change rate of shipping volume is less than the preset change rate threshold is set as a candidate item.
  • setting the first item whose average change rate of mailing volume is less than a preset change rate threshold as a candidate item includes:
  • the second items and the third items are screened out from the first items, and the first items whose average change rate of mailing volume after screening is less than a preset change rate threshold are used as candidate items.
  • the method further includes:
  • the items corresponding to the selection instructions are input into the preset area of the preset waybill.
  • this application provides an item recommendation device, including:
  • Acquisition unit used to obtain logistics information
  • a first determination unit configured to determine the shipping fluctuation value of each candidate item based on the shipping time and shipping volume of each candidate item in the logistics information
  • a second determination unit configured to determine the global fluctuation value of each candidate item based on the shipping fluctuation value of each candidate item and the average shipping volume of candidate items in the logistics information
  • a selection unit configured to select a target item to be recommended from the candidate items according to the global fluctuation value of each candidate item.
  • the first determination unit is also used to:
  • the shipping volume of each candidate item is divided to obtain the target shipping volume of each candidate item and the history of each candidate item. Shipping volume;
  • the shipping fluctuation value of each candidate item is calculated based on the target shipping volume of each candidate item and the historical shipping volume of each candidate item.
  • the second determination unit is also used to:
  • the shipping fluctuation value of each candidate item is weighted to obtain the global fluctuation value of each candidate item.
  • the selection unit is also used to:
  • Candidate items with global fluctuation values greater than the preset score threshold are set as target items to be recommended.
  • the first determination unit is also used to:
  • the first item whose average change rate of shipping volume is less than the preset change rate threshold is set as a candidate item.
  • the first determination unit is also used to:
  • the second items and the third items are screened out from the first items, and the first items whose average change rate of mailing volume after screening is less than a preset change rate threshold are used as candidate items.
  • the selection unit is also used to:
  • the items corresponding to the selection instructions are input into the preset area of the preset waybill.
  • the present application also provides an electronic device.
  • the electronic device includes a processor, a memory, and a computer program stored in the memory and executable on the processor.
  • the processor calls the computer program in the memory, it executes the method provided by the present application. The steps in any of the item recommendation methods.
  • this application also provides a storage medium.
  • a computer program is stored on the storage medium.
  • the computer program is executed by a processor, the steps in any of the item recommendation methods provided by this application are implemented.
  • the item recommendation method includes: obtaining logistics information; and determining the shipping fluctuation of each candidate item according to the shipping time and shipping volume of each candidate item in the logistics information. value; determine the global fluctuation value of each candidate item based on the shipping fluctuation value of each candidate item and the average shipping volume of candidate items in the logistics information; determine the global fluctuation value of each candidate item based on the global fluctuation value of each candidate item value, select the target item to be recommended from the candidate items.
  • the item recommendation method provided by the embodiment of the present application can recommend consignment items for the user when the user fills in the consignment, reducing the time for the user to fill in the waybill, and when determining the recommended target item, through the candidate items
  • the average shipping volume corrects the shipping fluctuation value of each candidate item to obtain a global fluctuation value that can more accurately represent the change in shipping volume over time, solving the problem that the shipping fluctuation value of a small sample of candidate items cannot accurately represent the shipping volume.
  • the problem changes over time, and the target items obtained are more accurate.
  • the embodiment of the present application can determine the target items to be recommended without the need for specific user information, so a cold start of filling in consigned items can be achieved.
  • Figure 1 is a schematic diagram of an application scenario of the item recommendation method provided in an embodiment of the present application
  • Figure 2 is a schematic flowchart of an item recommendation method provided in an embodiment of the present application.
  • Figure 3 is a schematic flowchart of an item recommendation method provided in another embodiment of the present application.
  • Figure 4 is a schematic flowchart of an item recommendation method provided in yet another embodiment of the present application.
  • Figure 5 is a schematic flowchart of an item recommendation method provided in yet another embodiment of the present application.
  • Figure 6 is a schematic flowchart of obtaining candidate items provided in an embodiment of the present application.
  • Figure 7 is a schematic flowchart of obtaining candidate items provided in another embodiment of the present application.
  • Figure 8 is a schematic flowchart of an item recommendation method provided in another embodiment of the present application.
  • Figure 9 is a schematic structural diagram of an item recommendation device provided in an embodiment of the present application.
  • Figure 10 is a schematic structural diagram of an electronic device provided in an embodiment of the present application.
  • first and second are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. quantity. Thus, features defined as “first” and “second” may explicitly or implicitly include one or more of the described features. In the description of the embodiments of this application, “plurality” means two or more, unless otherwise explicitly and specifically limited.
  • Embodiments of the present application provide an item recommendation method, device, electronic device, and storage medium.
  • the item recommendation device can be integrated in an electronic device, and the electronic device can be a server, a terminal or other equipment.
  • the execution subject of the item recommendation method in the embodiment of the present application may be the item recommendation device provided in the embodiment of the present application, or different types of electronic equipment such as server equipment, physical hosts, or user equipment (UE) that integrate the item recommendation device.
  • the item recommendation device can be implemented in the form of hardware or software, and the UE can specifically be a terminal device such as a smartphone, tablet computer, notebook computer, handheld computer, desktop computer or personal digital assistant (Personal Digital Assistant, PDA).
  • PDA Personal Digital Assistant
  • the electronic device can operate individually or in a cluster of devices.
  • Figure 1 is a schematic diagram of an application scenario of the item recommendation method provided in an embodiment of the present application.
  • the item recommendation system may include an electronic device 101, in which an item recommendation device is integrated.
  • the item recommendation system may also include a memory 102 for storing data, such as text data.
  • the scene schematic diagram of the item recommendation system shown in Figure 1 is only an example.
  • the item recommendation system and the scene described in the embodiment of the present application are for the purpose of more clearly explaining the technical solution of the embodiment of the present application, and do not constitute a provision for the embodiment of the present application.
  • the limitations of the technical solutions Persons of ordinary skill in the art know that with the evolution of item recommendation systems and the emergence of new business scenarios, the technical solutions provided by the embodiments of the present invention are also applicable to similar technical problems.
  • the item recommendation method includes: obtaining Logistics information; determine the shipping fluctuation value of each candidate item based on the shipping time and shipping volume of each candidate item in the logistics information; determine the shipping fluctuation value of each candidate item based on the shipping fluctuation value of each candidate item and the average shipping value of the candidate items in the logistics information
  • the quantity of items is determined to determine the global fluctuation value of each candidate item; based on the global fluctuation value of each candidate item, the target item to be recommended is selected from each candidate item.
  • Figure 2 is a schematic flowchart of an item recommendation method provided by an embodiment of the present application. Although a logical sequence is shown in the flowchart diagrams, in some cases, the steps shown or described may be performed in a sequence different from that described herein. step.
  • the item recommendation method may specifically include the following steps S201 to S204, wherein:
  • Step S201 Obtain logistics information.
  • the item recommendation method can be applied to express logistics software.
  • the consignment items are automatically recommended to the user, reducing the time for the user to fill in the waybill and improving the user experience.
  • the electronic device can use the item recommendation method provided by the embodiment of the present application. , determine the consignments to be recommended, and display the recommended consignments on the consignment item filling page for users to choose. Users do not need to manually fill in the waybill, but can complete the electronic order filling through selection methods such as touch screens.
  • Logistics information can refer to information generated when items are shipped.
  • the electronic device can obtain logistics information based on historical waybill data, that is, the electronic device uses the information contained in the historical waybill data as logistics information.
  • the logistics information can include time information corresponding to each historical waybill.
  • historical waybills may refer to all waybills generated within a historical period. For example, one month before the current time can be used as the above historical period, and all waybills generated within one month before the current time can be used as historical waybills.
  • the data of the historical waybill can be stored in the backend database of the express logistics software.
  • the electronic device can read the backend database of the express logistics software, obtain the historical waybill, and use the information contained in the historical waybill data as the logistics information.
  • the logistics information may include consigned item information of each historical waybill, and shipping time information of each historical waybill. Referring to Table 1, Table 1 shows a situation of logistics information. Assuming that there are 6 historical waybills, the logistics information can include the following information:
  • the logistics information can also include the express product information corresponding to each historical waybill.
  • the express product information can be understood as express type information, which is set by the express company.
  • express delivery product information may include express delivery type information such as "express delivery”, "delivery today", etc.
  • Step S202 Determine the shipping fluctuation value of each candidate item based on the shipping time and shipping volume of each candidate item in the logistics information.
  • Candidate items can include all consignment items in the logistics information.
  • the candidate items may include all consigned items corresponding to the historical waybill.
  • the logistics information is the information included in Table 1, the candidate items include apples, fish, and oranges, a total of three types of consignment items.
  • Shipping volume can refer to the number of times a shipment is sent.
  • the shipping quantity of each candidate item may refer to the number of times each candidate item is shipped.
  • the shipping volume of each candidate item in the logistics information may include the number of shipments of each candidate item per day during the historical period corresponding to the logistics information, which can be obtained from the number of historical waybills corresponding to each candidate item within each day. .
  • the logistics information is the information contained in Table 1, and assuming that the historical time corresponding to the logistics information is January 1 and January 2, for the candidate item "Apple", the corresponding historical waybill quantity is 3, and 3
  • the shipping time of historical waybills is January 1st, so the number of shipments of the candidate item "Apple” is: 3 pieces in total on January 1st, and 3 pieces in total on January 2nd. 0 in total.
  • the corresponding historical waybill quantity is 1 on January 1 and 1 on January 2. Therefore, the shipping quantity of the candidate item "fish” is: a total of 1 on January 1 , a total of 1 on January 2.
  • the shipping volume of each candidate item is considered to include the number of shipments of each candidate item per day during the historical period corresponding to the logistics information.
  • the shipping time of each candidate item may refer to the time each candidate item is sent.
  • the shipping time of each candidate item in the logistics information may include the shipping time information contained in the historical waybill corresponding to each candidate item.
  • the logistics information is the information included in Table 1
  • the corresponding historical waybills are historical waybills 1, 4, and 5, so the shipping time of the candidate item "Apple” can be January 1.
  • the corresponding historical waybills are historical waybills 3 and 6, so the shipping time of the candidate item "oranges" includes January 1 and January 2.
  • the shipping fluctuation value of each candidate item is the calculated value of the shipping volume changing with time based on the information related to each candidate item. It is used to express the total shipping volume as the measurement standard. How the shipment volume of candidate items changes over time.
  • the shipping fluctuation value of each candidate item can be used to represent the change in the recent shipping volume of each candidate item compared to the historical shipping volume, using its own total shipping volume as a measure. For example, the larger the shipping fluctuation value is, it means that when only considering the own shipping volume, the recent shipping volume of the corresponding candidate item has changed more compared to the historical shipping volume. In the near future, the shipping volume of the corresponding candidate item has appeared. surge.
  • the electronic device can divide the shipping volume according to the shipping time to obtain the recent shipping volume and the historical shipping volume. For example, if the historical time corresponding to the logistics information is one month before the current time, the shipping volume within half a month before the current time can be regarded as the recent shipping volume, and the remaining shipping volume can be regarded as the historical shipping volume. Assume The current time is February 1st, then the electronic device can regard the daily mailing volume from January 1st to January 15th as the historical mailing volume, and the daily mailing volume from January 16th to January 31st as the recent mailing volume. Shipping volume.
  • step S202 Determine the shipping fluctuation value of each candidate item based on the shipping time and shipping volume of each candidate item in the logistics information" can be performed in the following manner:
  • Step S2021 Divide the shipping volume of each candidate item according to the shipping time of each candidate item in the logistics information, and obtain the target shipping volume of each candidate item and the historical shipping volume of each candidate item.
  • the target shipping volume may refer to the recent shipping volume.
  • the description of the recent shipping volume can be referred to above, and the details will not be repeated.
  • the description of historical shipping volume can also be referred to above.
  • Step S2022 Calculate the shipping fluctuation value of each candidate item based on the target shipping volume of each candidate item and the historical shipping volume of each candidate item.
  • the method of calculating the shipping fluctuation value can be referred to the above, and the details will not be repeated.
  • the shipping fluctuation value calculated by this method will be greater than the shipping fluctuation value calculated when the shipment volume of the corresponding candidate items has surged in the near future, or the shipment fluctuation value calculated in the near future will be greater than the shipment fluctuation value calculated by this method.
  • the shipping fluctuation value calculated when the shipment volume of the corresponding candidate item does not change significantly can be used to characterize the change in the shipment volume of each candidate item over time, using its own total shipment volume as a measure.
  • formula (1) can be used to calculate the shipping fluctuation value of each candidate item:
  • S i refers to the shipping fluctuation value of the i-th candidate item, refers to the first total value of the i-th candidate item, refers to the second total value of the i-th candidate item.
  • the shipping fluctuation value is calculated by formula (1), but this cannot be understood as a limitation on the embodiments of the present application.
  • the reason why the difference between the first total value and the second total value, that is, the recent growth/decrease in shipping volume, is not directly used as the shipping fluctuation value is that for candidate items with a large total value difference, if their corresponding The total shipping volume, then the shipping fluctuation value cannot accurately represent the degree of change in the recent shipping volume of the candidate items.
  • candidate items A 1 and A 2 as examples. Assume that the first total value of A 1 is 278 and the second total value is 221. The first total value of A 2 is 50 and the second total value is 5. It can be seen that the difference corresponding to A 1 is greater than the difference corresponding to A 2. However, it is obvious from the first and second total values of A 1 and A 2 that A 2’s recent mailing volume has surged. If the difference between the first total value and the second total value is used as the shipping fluctuation value, it will be judged that A 1 's recent shipping volume has surged compared to A 2 .
  • Step S203 Determine the global fluctuation value of each candidate item based on the shipping fluctuation value of each candidate item and the average shipping volume of the candidate items in the logistics information.
  • the average shipping volume of candidate items refers to the average shipping value of all candidate items.
  • the electronic device can sum up the corresponding shipping amounts to obtain the total shipping amount of each candidate item (equivalent to the first total value and the second total value of each candidate item). (the total value of the value), and sum the total shipment quantity of each candidate item to obtain the total shipment quantity of all candidate items, and then calculate the difference between the total shipment quantity of all candidate items and the number of candidate items.
  • the ratio is used as the average shipping volume of candidate items. Take Table 1 as an example for illustrative explanation, in which the total shipment quantity of the candidate item "apple" is 3, the total shipment quantity of the candidate item "fish” is 1, and the total shipment quantity of the candidate item "orange” is 2. After summing, the total shipping quantity of all candidate items is 6, and the number of candidate items is 3. Therefore, for Table 1, the corresponding average shipping quantity of candidate items is 2.
  • the average shipment volume of candidate items is used to solve the problem of inaccurate shipment fluctuation values of small sample candidate items.
  • Small sample candidate items refer to the smaller total shipment volume, that is, the sum of the first total value and the second total value is larger.
  • Small candidate items Since the total number of shipments corresponding to a small sample of candidate items is small, when the fluctuation value of shipments is calculated according to equation (1), even if smaller, but It is also small, so the calculated S i may be too large.
  • candidate items B 1 and B 2 as an example. Assume that the first total value of B 1 is 1 and the second total value is 0.
  • the first total value of B 2 is 50 and the second total value is 5, then for the candidate item B 1 , the calculated S 1 is e, and for the candidate item B 2 , the calculated S 2 is S 1 is greater than S 2 , and the electronic device will determine that compared to B 2 , B 1 's recent mail volume has surged. However, it can be seen from the total values of B 1 and B 2 that B 2's recent mail volume has increased. There has been a surge, and B 1 's recent shipping volume has not changed significantly. If the items to be recommended are determined directly based on the shipping fluctuation value calculated by equation (1), candidate items that have not experienced a surge will be recommended. to users.
  • the global fluctuation value is a more accurate fluctuation value obtained by correcting the shipment fluctuation value by the average shipment volume of candidate items. It can be understood as expanding the total shipment volume of each candidate item through the average shipment volume of candidate items, so as to A value calculated for each candidate item that represents changes in shipping volume over time after avoiding small samples of candidate items.
  • the coefficient for correcting the shipping fluctuation value can be calculated based on the average shipping volume of candidate items, the target shipping volume of each candidate item, and the historical shipping volume of each candidate item, and based on the obtained The coefficient and the shipping fluctuation value are calculated to obtain the global fluctuation value.
  • step S203 Determine the global fluctuation value of each candidate item based on the shipping fluctuation value of each candidate item and the average shipping volume of the candidate items in the logistics information.
  • Step S2031 Count the number of items of each candidate item in the logistics information.
  • the number of items for each candidate item can refer to the explanation above. Taking Table 1 as an example, the number of items in the logistics information corresponding to Table 1 is 3.
  • Step S2032 Calculate the average shipping volume of candidate items in the logistics information based on the total shipping volume of each candidate item and the quantity of the items.
  • the total shipping volume of each candidate item refers to the total shipping volume obtained by summing the total shipping volume of each candidate item after calculating the first total value and the second total value for each candidate item.
  • Piece quantity Take Table 1 as an example. In the logistics information corresponding to Table 1, the total shipment quantity of each candidate item is 6.
  • the average shipping volume of the candidate items can be obtained.
  • Step S2033 Calculate the fluctuation correction coefficient of each candidate item based on the average shipping volume of the candidate items, the target shipping volume of each candidate item, and the historical shipping volume of each candidate item.
  • the electronic device can calculate the first total value and the second total value corresponding to each candidate item according to the target shipping volume and the historical shipping volume through the above method, and then Calculate the total value sum and total value difference corresponding to each candidate item, and calculate the fluctuation correction coefficient through formula (2):
  • T i refers to the fluctuation correction coefficient of the i-th candidate item
  • first total value of the i-th candidate item refers to the second total value of the i-th candidate item
  • avg refers to the average shipping volume of candidate items.
  • the total shipment volume of candidate items can be expanded by the average shipment volume of candidate items, and the proportion of the growth/decrease value of the recent shipment volume in the total shipment volume after expansion can be calculated. , and then determine the degree of growth/decrease in recent shipping volume.
  • the smaller the increase/decrease in volume the smaller the size of the increase/decrease value, thereby avoiding the problem of inaccurate recommendations for small sample candidate items due to the small total shipment volume.
  • formula (3) can also be used, that is, a preset expansion value can be added to the denominator in formula (2) to obtain the global fluctuation value, so that each candidate item includes multiple small When sampling candidate items, further expand the total shipping volume of each candidate item to avoid the problem that the average shipping volume of candidate items is still small, resulting in inaccurate recommendations:
  • T i refers to the fluctuation correction coefficient of the i-th candidate item
  • avg refers to the average shipping volume of candidate items
  • W is the preset expansion value, which can be set according to the needs of the actual scenario.
  • Step S2034 Weight the shipping fluctuation value of each candidate item according to the fluctuation correction coefficient to obtain the global fluctuation value of each candidate item.
  • G i refers to the global fluctuation value of the i-th candidate item
  • Si refers to the shipping fluctuation value of the i-th candidate item
  • Ti refers to the fluctuation correction coefficient of the i-th candidate item.
  • Step S204 Select a target item to be recommended from each candidate item based on the global fluctuation value of each candidate item.
  • the target items to be recommended can be understood as currently popular consignment items.
  • the electronic device may select the candidate item with the largest global fluctuation value and use the candidate item as the target item.
  • the electronic device may compare the global fluctuation value of each candidate item with a preset fluctuation threshold, and use the candidate item with a global fluctuation value greater than the fluctuation threshold as the target item.
  • the preset fluctuation threshold is used to evaluate the size of the global fluctuation value, and the specific value can be set according to the needs of the actual scenario.
  • step S204 selecting target items to be recommended from each candidate item according to the global fluctuation value of each candidate item.
  • Step S2041 Compare the global fluctuation value of each candidate item with a preset score threshold to obtain candidate items whose global fluctuation value is greater than the preset score threshold.
  • the preset score threshold is the fluctuation threshold mentioned above, and the details will not be described again.
  • Step S2042 Set the candidate items whose global fluctuation value is greater than the preset score threshold as target items to be recommended.
  • the method of steps S201 to S204 can also achieve the cold filling in of consigned items when sending, since it is not necessary to obtain the historical data of a specific user to select the target items to be recommended.
  • start By clicking on the button, you can recommend candidate items that the user has not filled in before. For example, for some candidate items with seasonal attributes, such as rice dumplings, moon cakes, etc., they can be identified as target items to be recommended before a specific season to solve the user cold start problem.
  • the item recommendation method includes: obtaining logistics information; determining the shipping fluctuation value of each candidate item based on the shipping time and shipping volume of each candidate item in the logistics information; The shipping fluctuation value of each candidate item and the average shipping volume of candidate items in the logistics information are used to determine the global fluctuation value of each candidate item; based on the global fluctuation value of each candidate item, the target to be recommended is selected from each candidate item. thing.
  • the item recommendation method provided by the embodiment of the present application can recommend consignment items for the user when the user fills in the consignment, reducing the time for the user to fill in the waybill, and when determining the recommended target item, through the candidate items
  • the average shipping volume corrects the shipping fluctuation value of each candidate item to obtain a global fluctuation value that can more accurately represent the change in shipping volume over time, solving the problem that the shipping fluctuation value of a small sample of candidate items cannot accurately represent the shipping volume.
  • the problem changes over time, and the target items obtained are more accurate.
  • the embodiment of the present application can determine the target items to be recommended without the need for specific user information, so a cold start of filling in consigned items can be achieved.
  • the electronic device may first screen the consignment items included in the logistics information, and select consignment items that are less prone to sudden changes in shipment volume as candidate items.
  • the method also includes :
  • Step S301 Based on the shipping time of the first item in the logistics information, count the shipping volume of the first item in each preset historical time period.
  • the time range corresponding to the logistics information can be divided according to the preset time interval to obtain each preset historical time period. For example, if the electronic device uses historical waybill information within one month before the current time as logistics information, that is, the time range corresponding to the logistics information refers to one month, and the preset time interval is 5 days, then one month can be divided into multiple There are three preset historical time periods, and each preset historical time period corresponds to 5 days in the month.
  • the first items refer to all consigned items included in the logistics information. Taking Table 1 as an example, for the logistics information in Table 1, the first items include "apples”, “fish” and “oranges”.
  • the shipping volume of the first item in the preset historical time period refers to the number of times the first item is sent in the preset historical time period.
  • Table 1 assume that the time range corresponding to the logistics information in Table 1 is divided according to the preset time interval of 1 day, and two preset historical time periods are obtained, "January 1st" and "January 2nd". Then for the first item "Apple”, the shipping volume in the preset time period "January 1st” includes the number of times "Apple” was sent on January 1st.
  • the shipping volume in the time period "January 1st” includes the number of shipments of "fish” in January 1st.
  • the first item "orange” in the preset time period "January 1st” The shipping volume includes the number of shipments sent by "Orange” on January 1.
  • Step S302 Calculate the average change rate of the shipping volume of the first item based on the shipping volume in each preset historical time period and the time interval corresponding to each preset historical time period.
  • the time interval corresponding to each preset historical time period refers to the time interval preset above, that is, the time range included in each preset historical time period.
  • the electronic device can calculate the change rate of the mailing volume corresponding to each two adjacent preset historical event segments based on the mailing volume in each two adjacent preset historical event segments and the time interval, Then, all the change rates of the shipping volume corresponding to each preset historical time period are averaged to obtain the average change rate of the shipping volume of the first item.
  • an example is given below for illustrative explanation, but it should not be understood as a limitation to the embodiment of the present application: Assume that there are three preset historical time periods a, b, and c, and the first item is in the preset historical time period a.
  • the shipping volumes in b and c are 10, 20, and 30 respectively, and the time interval corresponding to the preset historical time period is 5 days.
  • the change rates of the shipping volume are respectively the change rate of the shipping volume between a and b 2/day, and the change rate of the shipping volume between b and c 2/day.
  • the average change rate of shipping volume is 2/day.
  • Step S303 Set the first item whose average change rate of mailing quantity is less than the preset change rate threshold as a candidate item.
  • the preset change rate threshold is used to evaluate the average change rate of mailing volume, and the specific value can be set according to actual scene requirements.
  • the item recommendation method provided by the embodiment of the present application determines the target item based on the changes in the shipping volume, , therefore, for the first item whose shipment volume often changes suddenly, it cannot be judged by the item recommendation method provided by the embodiment of the present application, and needs to be screened out to avoid recommendation errors and reduce the amount of calculation.
  • step S303 “Set the first item with an average change rate of shipment volume less than the preset change rate threshold as a candidate item” may include:
  • Step S3031 Obtain second items whose total historical shipping volume is less than the first preset frequency threshold from the shipping records of the target users to be recommended.
  • the target users to be recommended may refer to users who open express logistics software. For details, please refer to the description in step S201.
  • Shipping records can include the user's historical shipping behavior data, for example, they can include waybill data created by the user. Therefore, the target user's shipping record contains historical waybill information created by the target user.
  • the historical waybill information here can refer to all waybills created by the target user, or it can refer to the waybills created by the target user within a period of time before the current time.
  • the shipping record can be stored in the backend database of the express logistics software.
  • the electronic device can query the backend database according to the user identity of the target user to obtain the shipping record of the target user.
  • the user identity identifier is used to distinguish the identities of different users, and may refer to the login name used by the user when logging into the express logistics software, etc.
  • the total number of historical shipments refers to the total number of shipments in the shipping record.
  • the electronic device can query each historical waybill in the shipping record and obtain each historical waybill in the shipping record. Corresponding consignment items, and then count the total number of shipments of different consignment items in the target user's shipping records to obtain the total historical shipment volume of different consignment items.
  • the first preset number threshold is used to evaluate the total size of historical mailings.
  • the specific value can be set according to actual scenario requirements, for example, it can be set to 1.
  • Taking the first preset frequency threshold as 1 as an example if the total number of historical shipments is less than the first preset frequency threshold, it means that the target user has never sent the corresponding consignment item. Even if the global fluctuation value is large, the current shipment If there is a surge in volume, the target users may not necessarily choose to send the consignment items, so the consignment items can be filtered out.
  • Step S3032 Determine the user group to which the target user belongs.
  • a group refers to a collection of users obtained by dividing users according to attributes such as age, occupation, gender, region, etc. For example, users can be divided into multiple preset groups according to the city they set in the express logistics software, and then based on the target user's user identity, the preset groups can be queried to obtain the target users. Client. For the convenience of understanding, unless otherwise stated below, it is assumed that the default groups are divided according to regions.
  • Step S3033 Obtain third items whose average shipping volume during the same historical period is less than the second preset frequency threshold from the shipping records of the user group.
  • the electronic device can obtain all users in the user group, extract historical mailing behavior data of all users in the user group, and obtain the third item therefrom.
  • the average shipping volume for the same period in history may refer to the average shipping volume corresponding to the consignment items during the same period in history.
  • the second preset frequency threshold is used to evaluate the average shipping volume during the same historical period, and the specific value can be set according to actual scenario requirements. If the average shipping volume in the same historical period is less than the second preset frequency threshold, it means that in the region corresponding to the user group and in the historical same period at the current time, the shipping volume of the corresponding consignment item in the same period is low, even if the consignment item is It is a seasonal item and is not a popular consignment item in the area, so it can be screened out.
  • Step S3034 Screen out the second items and third items from the first items, and use the first items whose average change rate of mailing volume after screening is less than the preset change rate threshold as candidate items.
  • step S303 The reason for selecting the candidate items based on the average change rate of the shipment volume can be referred to step S303, which will not be described in detail.
  • the electronic device can monitor whether a selection instruction from the target user to be recommended is received, and when the selection instruction is received, fill in the items corresponding to the selection instruction in the target item into the waybill. corresponding area.
  • the method also includes:
  • Step S401 Receive a selection instruction from a target user to be recommended, and obtain the item corresponding to the selection instruction among the target items.
  • step S3031 For the description of the target user, please refer to step S3031, and details will not be described again.
  • the selection instructions may refer to touch screen, voice and other types of instructions.
  • the embodiments of the present application are not limited to this.
  • the user can select an item among the target items by touching the screen to send the message to carry the item.
  • Information selection instructions may refer to touch screen, voice and other types of instructions.
  • Step S402 Enter the items corresponding to the selection instruction into the preset area of the preset waybill.
  • the preset waybill may refer to an unfilled waybill newly created by the target user.
  • the preset area may refer to an area in the preset waybill used to fill in the consignment items.
  • the embodiment of the present application also provides an item recommendation device, as shown in Figure 9, which is the item recommendation device in the embodiment of the present application.
  • the item recommendation device 500 includes:
  • Obtaining unit 501 is used to obtain logistics information
  • the first determination unit 502 is configured to determine the shipping fluctuation value of each candidate item based on the shipping time and shipping volume of each candidate item in the logistics information;
  • the second determination unit 503 is used to determine the global fluctuation value of each candidate item based on the shipping fluctuation value of each candidate item and the average shipping volume of the candidate items in the logistics information;
  • the selection unit 504 is used to select a target item to be recommended from each candidate item according to the global fluctuation value of each candidate item.
  • the first determining unit 502 is also used to:
  • each candidate item in the logistics information divide the shipping volume of each candidate item to obtain the target shipping volume of each candidate item and the historical shipping volume of each candidate item;
  • the shipping fluctuation value of each candidate item is calculated.
  • the second determining unit 503 is also used to:
  • the shipping fluctuation value of each candidate item is weighted to obtain the global fluctuation value of each candidate item.
  • the selection unit 504 is also used to:
  • Candidate items with global fluctuation values greater than the preset score threshold are set as target items to be recommended.
  • the first determining unit 502 is also used to:
  • the first item whose average change rate of shipping volume is less than the preset change rate threshold is set as a candidate item.
  • the first determining unit 502 is also used to:
  • the second items and the third items are screened out from the first items, and the first items whose average change rate of the mailing volume after screening is less than the preset change rate threshold are used as candidate items.
  • the selection unit 504 is also used to:
  • each of the above units can be implemented as an independent entity, or can be combined in any way to be implemented as the same or several entities.
  • each of the above units please refer to the previous method embodiments, and will not be described again here.
  • the item recommendation device can perform the steps of the item recommendation method in any embodiment, it can achieve the beneficial effects that can be achieved by the item recommendation method in any embodiment of the present application. Please refer to the previous description for details and will not be repeated here.
  • Figure 10 shows a schematic structural diagram of the electronic device of the embodiment of the present application.
  • the electronic device provided by the embodiment of the present application includes a processor. 601.
  • the processor 601 is used to implement the steps of the item recommendation method in any embodiment when executing the computer program stored in the memory 602; or, when the processor 601 is used to execute the computer program stored in the memory 602, implement the corresponding implementation as shown in Figure 9 Functions of each unit in the example.
  • a computer program can be divided into one or more modules/units, and one or more modules/units are stored in the memory 602 and executed by the processor 601 to complete the embodiments of the present application.
  • One or more modules/units may be a series of computer program instruction segments capable of completing specific functions. The instruction segments are used to describe the execution process of the computer program in the computer device.
  • the electronic device may include, but is not limited to, a processor 601 and a memory 602 .
  • a processor 601 and a memory 602 .
  • the electronic equipment may include more or less components than those shown in the figures, or some components may be combined, or different components may be used.
  • the processor 601 can be a central processing unit (Central Processing Unit, CPU), or other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), or off-the-shelf processors.
  • Programmable gate array Field-Programmable Gate Array, FPGA
  • a general-purpose processor can be a microprocessor or the processor can be any conventional processor, etc.
  • the processor is the control center of the electronic device and uses various interfaces and lines to connect various parts of the entire electronic device.
  • the memory 602 can be used to store computer programs and/or modules.
  • the processor 601 implements various functions of the computer device by running or executing the computer programs and/or modules stored in the memory 602 and calling data stored in the memory 602.
  • the memory 602 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the storage data area may store a program based on Data created by the use of electronic equipment (such as audio data, video data, etc.), etc.
  • the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart memory card (Smart Media Card, SMC), secure digital (Secure Digital, SD) card , Flash Card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.
  • non-volatile memory such as hard disk, memory, plug-in hard disk, smart memory card (Smart Media Card, SMC), secure digital (Secure Digital, SD) card , Flash Card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.
  • embodiments of the present application provide a storage medium.
  • a computer program is stored on the storage medium.
  • the steps in the item recommendation method in any embodiment of the present application are performed.
  • the storage medium may include: read-only memory (Read Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk, etc.
  • the instructions stored in the storage medium can execute the steps in the item recommendation method in any embodiment of the present application, the beneficial effects that can be achieved by the item recommendation method in any embodiment of the present application can be achieved. For details, see the previous description. , which will not be described in detail here.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • Economics (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Probability & Statistics with Applications (AREA)
  • Development Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Human Resources & Organizations (AREA)
  • Tourism & Hospitality (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Mathematical Physics (AREA)
  • Fuzzy Systems (AREA)
  • Software Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

本申请公开了一种物品推荐方法、装置、电子设备及存储介质，一方面，本申请实施例提供的物品推荐方法可以在用户填写托寄物时，为用户进行托寄物品推荐，减少用户填写运单的时间，并且在确定推荐的目标物品时，通过候选物品平均寄件量对每个候选物品的寄件波动值进行修正，得到更加能够准确表示寄件量随时间变化的全局波动值，解决了小样本候选物品的寄件波动值无法准确表示寄件量随时间变化的问题，进而得到的目标物品更加准确。另一方面，本申请实施例无需根据特定用户的信息即可确定待推荐的目标物品，因此可以实现托寄物品填写的冷启动。

Description

物品推荐方法、装置、电子设备及存储介质 技术领域
本申请涉及物品推荐技术领域，具体涉及一种物品推荐方法、装置、电子设备及存储介质。
发明背景
为了方便用户，目前快递物流公司在自研发的软件中加入了在线填写运单的功能，以减少用户手写运单所需要的时间，并且可以降低用户手写时写错字的概率。
但是，虽然用户可以通过智能手机、个人电脑等电子设备进行运单填写，但是仍然需要通过键盘打字输入托寄物的名称，十分消耗时间。
发明内容
本申请提供一种物品推荐方法、装置、电子设备及存储介质，旨在解决目前的物品推荐方法识别精度不高的问题。
第一方面，本申请提供一种物品推荐方法，包括：
获取物流信息；
根据所述物流信息中每个候选物品的寄件时间和寄件量，确定所述每个候选物品的寄件波动值；
根据所述每个候选物品的寄件波动值和所述物流信息中的候选物品平均寄件量，确定所述每个候选物品的全局波动值；
根据所述每个候选物品的全局波动值，从所述各候选物品中选择待推荐的目标物品。
在本申请一种可能的实现方式中，所述根据所述物流信息中每个候选物品的寄件时间和寄件量，确定所述每个候选物品的寄件波动值，包括：
根据所述物流信息中每个候选物品的寄件时间，对所述每个候选物品的寄件量进行划分，得到所述每个候选物品的目标寄件量和所述每个候选物品的历史寄件量；
根据所述每个候选物品的目标寄件量和所述每个候选物品的历史寄件量，计算得到所述每个候选物品的寄件波动值。
在本申请一种可能的实现方式中，所述根据所述每个候选物品的寄件波动值和所述物流信息中的候选物品平均寄件量，确定所述每个候选物品的全局波动值，包括：
统计所述物流信息中各候选物品的物品数量；
根据所述各候选物品的寄件总量和所述物品数量，计算得到所述物流信息中的候选物品平均寄件量；
根据所述候选物品平均寄件量，以及所述每个候选物品的目标寄件量和所述每个候选物品的历史寄件量，计算得到所述每个候选物品的波动修正系数；
根据所述波动修正系数，对所述每个候选物品的寄件波动值进行加权，得到所述每个候选物品的全局波动值。
在本申请一种可能的实现方式中，所述根据所述每个候选物品的全局波动值，从所述各候选物品中选择待推荐的目标物品，包括：
将所述每个候选物品的全局波动值与预设分数阈值进行对比，得到全局波动值大于预设分数阈值的候选物品；
将全局波动值大于预设分数阈值的候选物品设定为待推荐的目标物品。
在本申请一种可能的实现方式中，所述根据所述物流信息中每个候选物品的寄件时间和寄件量，确定所述每个候选物品的寄件波动值之前，所述方法还包括：
根据所述物流信息中第一物品的寄件时间，统计所述第一物品分别在各预设历史时间段中的寄件量；
根据所述各预设历史时间段中的寄件量，以及所述各预设历史时间段对应的时间间隔，计算得到所述第一物品的寄件量平均变化率；
将寄件量平均变化率小于预设变化率阈值的第一物品设定为候选物品。
在本申请一种可能的实现方式中，所述将寄件量平均变化率小于预设变化率阈值的第一物品设定为候选物品，包括：
从待推荐的目标用户的寄件记录中，获取历史寄件总量小于第一预设次数阈值的第二物品；
确定所述目标用户所属的用户群体；
从所述用户群体的寄件记录中，获取历史同期的平均寄件量小于第二预设次数阈值的第三物品；
从所述第一物品中筛除所述第二物品和所述第三物品，并将筛除后寄件量平均变化率小于预设变化率阈值的第一物品作为候选物品。
在本申请一种可能的实现方式中，所述根据所述每个候选物品的全局波动值，从所述各候选物品中选择待推荐的目标物品之后，所述方法还包括：
接收待推荐的目标用户的选择指令，得到所述目标物品中所述选择指令对应的物品；
将所述选择指令对应的物品输入预设运单的预设区域。
第二方面，本申请提供一种物品推荐装置，包括：
获取单元，用于获取物流信息；
第一确定单元，用于根据所述物流信息中每个候选物品的寄件时间和寄件量，确定所述每个候选物品的寄件波动值；
第二确定单元，用于根据所述每个候选物品的寄件波动值和所述物流信息中的候选物品平均寄件量，确定所述每个候选物品的全局波动值；
选择单元，用于根据所述每个候选物品的全局波动值，从所述各候选物品中选择待推荐的目标物品。
在本申请一种可能的实现方式中，第一确定单元还用于：
根据所述物流信息中每个候选物品的寄件时间，对所述每个候选物品的寄件量进行划分，得到所述每个候选物品的目标寄件量和所述每个候选物品的历史寄件量；
根据所述每个候选物品的目标寄件量和所述每个候选物品的历史寄件量，计算得到所述每个候选物品的寄件波动值。
在本申请一种可能的实现方式中，第二确定单元还用于：
统计所述物流信息中各候选物品的物品数量；
根据所述各候选物品的寄件总量和所述物品数量，计算得到所述物流信息中的候选物品平均寄件量；
根据所述候选物品平均寄件量，以及所述每个候选物品的目标寄件量和所述每个候选物品的历史寄件量，计算得到所述每个候选物品的波动修正系数；
根据所述波动修正系数，对所述每个候选物品的寄件波动值进行加权，得到所述每个候选物品的全局波动值。
在本申请一种可能的实现方式中，选择单元还用于：
将所述每个候选物品的全局波动值与预设分数阈值进行对比，得到全局波动值大于预设分数阈值的候选物品；
将全局波动值大于预设分数阈值的候选物品设定为待推荐的目标物品。
在本申请一种可能的实现方式中，第一确定单元还用于：
根据所述物流信息中第一物品的寄件时间，统计所述第一物品分别在各预设历史时间段中的寄件量；
根据所述各预设历史时间段中的寄件量，以及所述各预设历史时间段对应的时间间隔，计算得到所述第一物品的寄件量平均变化率；
将寄件量平均变化率小于预设变化率阈值的第一物品设定为候选物品。
在本申请一种可能的实现方式中，第一确定单元还用于：
从待推荐的目标用户的寄件记录中，获取历史寄件总量小于第一预设次数阈值的第二物品；
确定所述目标用户所属的用户群体；
从所述用户群体的寄件记录中，获取历史同期的平均寄件量小于第二预设次数阈值的第三物品；
从所述第一物品中筛除所述第二物品和所述第三物品，并将筛除后寄件量平均变化率小于预设变化率阈值的第一物品作为候选物品。
在本申请一种可能的实现方式中，选择单元还用于：
接收待推荐的目标用户的选择指令，得到所述目标物品中所述选择指令对应的物品；
将所述选择指令对应的物品输入预设运单的预设区域。
第三方面，本申请还提供一种电子设备，电子设备包括处理器、存储器以及存储于存储器中并可在处理器上运行的计算机程序，处理器调用存储器中的计算机程序时执行本申请提供的任一种物品推荐方法中的步骤。
第四方面，本申请还提供一种存储介质，存储介质上存储有计算机程序，计算机程序被处理器执行时实现本申请提供的任一种物品推荐方法中的步骤。
综上所述，本申请实施例提供的物品推荐方法包括：获取物流信息；根据所述物流信息中每个候选物品的寄件时间和寄件量，确定所述每个候选物品的寄件波动值；根据所述每个候选物品的寄件波动值和所述物流信息中的候选物品平均寄件量，确定所述每个候选物品的全局波动值；根据所述每个候选物品的全局波动值，从所述各候选物品中选择待推荐的目标物品。
可见，一方面，本申请实施例提供的物品推荐方法可以在用户填写托寄物时，为用户进行托寄物品推荐，减少用户填写运单的时间，并且在确定推荐的目标物品时，通过候选物品平均寄件量对每个候选物品的寄件波动值进行修正，得到更加能够准确表示寄件量随时间变化的全局波动值，解决了小样本候选物品的寄件波动值无法准确表示寄件量随时间变化的问题，进而得到的目标物品更加准确。另一方面，本申请实施例无需根据特定用户的信息即可确定待推荐的目标物品，因此可以实现托寄物品填写的冷启动。
附图简要说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本申请一实施例中提供的物品推荐方法的应用场景示意图；
图2是本申请一实施例中提供的物品推荐方法的流程示意图；
图3是本申请另一实施例中提供的物品推荐方法的流程示意图；
图4是本申请又一实施例中提供的物品推荐方法的流程示意图；
图5是本申请再一实施例中提供的物品推荐方法的流程示意图；
图6是本申请一实施例中提供的获取候选物品的一种流程示意图；
图7是本申请另一实施例中提供的获取候选物品的流程示意图；
图8是本申请另一实施例中提供的物品推荐方法的流程示意图；
图9是本申请一实施例中提供的物品推荐装置的结构示意图；
图10是本申请一实施例中提供的电子设备的结构示意图。
实施本发明的方式
下面将结合本申请实施例中的附图，对本申请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申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
在本申请实施例的描述中，需要理解的是，术语“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含地包括一个或者更多个所述特征。在本申请实施例的描述中，“多个”的含义是两个或两个以上，除非另有明确具体的限定。
为了使本领域任何技术人员能够实现和使用本申请，给出了以下描述。在以下描述中，为了解释的目的而列出了细节。应当明白的是，本领域普通技术人员可以认识到，在不使用这些特定细节的情况下也可以实现本申请。在其它实例中，不会对公知的过程进行详细阐述，以避免不必要的细节使本申请实施例的描述变得晦涩。因此，本申请并非旨在限于所示的实施例，而是与符合本申请实施例所公开的原理和特征的最广范围相一致。
本申请实施例提供一种物品推荐方法、装置、电子设备和存储介质。其中，该物品推荐装置可以集成在电子设备中，该电子设备可以是服务器，也可以是终端等设备。
本申请实施例物品推荐方法的执行主体可以为本申请实施例提供的物品推荐装置，或者集成了该物品推荐装置的服务器设备、物理主机或者用户设备(User Equipment，UE)等不同类型的电子设备，其中，物品推荐装置可以采用硬件或者软件的方式实现，UE具体可以为智能手机、平板电脑、笔记本电脑、掌上电脑、台式电脑或者个人数字助理(Personal Digital Assistant，PDA)等终端设备。
该电子设备可以采用单独运行的工作方式，或者也可以采用设备集群的工作方式。
参见图1，图1是本申请一实施例中提供的物品推荐方法的应用场景示意图。其中，该物品推荐系统可以包括电子设备101，电子设备101中集成有物品推荐装置。
另外，如图1所示，该物品推荐系统还可以包括存储器102，用于存储数据，如存储文本数据。
图1所示的物品推荐系统的场景示意图仅仅是一个示例，本申请实施例描述的物品推荐系统以及场景是为了更加清楚的说明本申请实施例的技术方案，并不构成对于本申请实施例提供的技术方案的限定，本领域普通技术人员可知，随着物品推荐系统的演变和新业务场景的出现，本发明实施例提供的技术方案对于类似的技术问题，同样适用。
下面，开始介绍本申请实施例提供的物品推荐方法，本申请实施例中以电子设备作为执行主体，为了简化与便于描述，后续方法实施例中将省略该执行主体，该物品推荐方法包括：获取物流信息；根据物流信息中每个候选物品的寄件时间和寄件量，确定每个候选物品的寄件波动值；根据每个候选物品的寄件波动值和物流信息中的候选物品平均寄件量，确定每个候选物品的全局波动值；根据每个候选物品的全局波动值，从各候选物品中选择待推荐的目标物品。
参照图2，图2是本申请一实施例提供的物品推荐方法的流程示意图。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。该物品推荐方法具体可以包括以下步骤S201-步骤S204，其中：
步骤S201、获取物流信息。
在本申请实施例中，物品推荐方法可以应用于快递物流软件中，在用户填写运单时，为用户自动推荐托寄物品，减少用户填写运单的时间，提高用户体验感。例如，当用户通过智能手机、个人电脑等终端打开用于寄出快递的APP，并选择待填写的电子运单，进入托寄物品填写页面时，电子设备可以通过本申请实施例提供的物品推荐方法，确定待推荐的托寄物，并在托寄物品填写页面中显示待推荐的托寄物供用户选择，用户无需手动填写运单，而是通过触屏等选择方式即可完成电子订单的填写。
物流信息可以是指物品寄件时产生的信息。示例性地，电子设备可以根据历史运单的数据，得到物流信息，即电子设备将历史运单数据中包含的信息作为物流信息，此时，物流信息中可以包括每一张历史运单对应的时间信息。其中，历史运单可以是指一段历史时间内生成的所有运单。示例性地，可以将当前时间之前的一个月作为上述一段历史时间，将当前时间之前一个月内生成的所有运单作为历史运单。
其中，历史运单的数据可以存储在快递物流软件的后台数据库中，在执行步骤S201时，电子设备可以读取快递物流软件的后台数据库，得到历史运单，并将历史运单数据中包含的信息作为物流信息。
在一些实施例中，物流信息可以包括各历史运单的托寄物品信息，以及各历史运单的寄件时间信息。参考表1，表1中示出了物流信息的一种情况，假设历史运单共有6张，则物流信息可以包括以下信息：
表1
除此之外，物流信息中还可以包括各历史运单对应的快递产品信息，快递产品信息可以理解为快递类型的信息，由快递公司设定。例如，快递产品信息中可以包含“速运”、“今日达”等等快递类型的信息。
步骤S202、根据物流信息中每个候选物品的寄件时间和寄件量，确定每个候选物品的寄件波动值。
候选物品可以包含物流信息中所有的托寄物品。示例性地，当电子设备将历史运单数据中包含的信息作为物流信息时，候选物品可以包括历史运单对应的所有托寄物品。例如物流信息为表1中包含的信息时，候选物品包含苹果、鱼类、橙子，共3种托寄物品。
寄件量可以是指寄件的次数。每个候选物品的寄件量可以是指每个候选物品寄件的次数。示例性地，物流信息中每个候选物品的寄件量可以包括物流信息对应的历史时间内，每个候选物品每天的寄件次数，可以由每天内，每个候选物品对应的历史运单数量得到。例如物流信息为表1中包含的信息，并且假设物流信息对应的历史时间为1月1日和1月2日时，对于候选物品“苹果”，其对应的历史运单数量为3，并且3张历史运单的寄件时间均为1月1日，因此候选物品“苹果”的寄件量为：1月1日共3个，1月2日 共0个。又例如对于候选物品“橙子”，其对应的历史运单数量在1月1日为1，在1月2日为1，因此候选物品“鱼类”的寄件量为：1月1日共1个，1月2日共1个。为了方便理解，下文中若未作特别说明，则认为每个候选物品的寄件量包括物流信息对应的历史时间内，每个候选物品每天的寄件次数。
每个候选物品的寄件时间可以是指每个候选物品每次寄出的时间。示例性地，物流信息中每个候选物品的寄件时间可以包括每个候选物品对应的历史运单中，包含的寄件时间信息。例如物流信息为表1中包含的信息时，对于候选物品“苹果”，其对应的历史运单为历史运单1、4、5，因此候选物品“苹果”的寄件时间可以为1月1日。又例如对于候选物品“橙子”，其对应的历史运单为历史运单3和6，因此候选物品“橙子”的寄件时间包括1月1日和1月2日。
其中，每个候选物品的寄件波动值是根据每个候选物品自身相关的信息，计算得到的寄件量随时间变化的值，用于表示将自身的寄件总量作为衡量标准，每个候选物品的寄件量随时间变化的大小情况。示例性地，每个候选物品的寄件波动值可以用来表示将自身的寄件总量作为衡量标准，每个候选物品的近期寄件量相比历史寄件量的变化大小情况。例如，寄件波动值越大，说明在仅考虑自身寄件量的情况下，对应候选物品的近期寄件量相比历史寄件量的变化越大，在近期对应候选物品的寄件量出现激增。寄件波动值越小，说明在仅考虑自身寄件量的情况下，对应候选物品的近期寄件量相比历史寄件量的变化越小，在近期对应候选物品的寄件量出现激减。
在一些实施例中，对于每个候选物品，电子设备可以根据寄件时间对寄件量进行划分，得到近期寄件量和历史寄件量。例如，若物流信息对应的历史时间为当前时间之前的一个月，则可以将当前时间之前半个月内的寄件量作为近期寄件量，将剩余的寄件量作为历史寄件量，假设当前时间为2月1日，则电子设备可以将1月1日-1月15日内每天的寄件量作为历史寄件量，将1月16日-1月31日内每天的寄件量作为近期寄件量。然后，电子设备计算近期寄件量的第一总值，以及历史寄件量的第二总值，并将第一总值和第二总值之间的总值差，与第一总值和第二总值之间的总值和作比值，以得到寄件量随时间的变化值占历史事件内寄件总量的比值，将该比值作为该候选物品的寄件波动值。其中，每个候选物品的总值差可用于表示近期寄件量的增长值/减少值，每个候选物品的总值和可用于表示候选物品的寄件总量。因此，参考图3，步骤S202“根据物流信息中每个候选物品的寄件时间和寄件量，确定每个候选物品的寄件波动值”可以由以下方式进行：
步骤S2021、根据物流信息中每个候选物品的寄件时间，对每个候选物品的寄件量进行划分，得到每个候选物品的目标寄件量和每个候选物品的历史寄件量。
其中，目标寄件量可以是指近期寄件量。近期寄件量的说明可以参考上文，具体不进行赘述。同样地，历史寄件量的说明也可以参考上文。
步骤S2022、根据每个候选物品的目标寄件量和每个候选物品的历史寄件量，计算得到每个候选物品的寄件波动值。
计算寄件波动值的方法可以参考上文，具体不进行赘述。
可见，通过这种方法计算得到的寄件波动值在近期对应候选物品的寄件量出现激增时，将大于近期对应候选物品的寄件量出现激减时计算得到的寄件波动值，或者近期对应候选物品的寄件量未发生明显变化时计算得到的寄件波动值，可用于表征将自身的寄件总量作为衡量标准，每个候选物品的寄件量随时间变化的大小情况。
进一步地，为了保证每个候选物品的寄件波动值均为正数，方便后续运算，可以采用式子(1)计算得到每个候选物品的寄件波动值：
其中，Si是指第i个候选物品的寄件波动值，是指第i个候选物品的第一总值， 是指第i个候选物品的第二总值。为了方便理解，下文中若未作特别声明，则认为通过式子(1)计算得到寄件波动值，但不能理解为对本申请实施例的限制。
不直接将第一总值和第二总值的差，即近期寄件量的增长值/减少值作为寄件波动值的原因是，对于总值差较大的候选物品，若不考虑其对应的寄件总量，则寄件波动值无法准确表示候选物品近期寄件量的变化程度。为了方便理解，以候选物品A1和A2为例进行说明，假设A1的第一总值为278，第二总值为221，A2的第一总值为50，第二总值为5，可见，A1对应的差大于A2对应的差，然而从A1和A2各自的第一总值和第二总值明显可以看出，A2近期的寄件量出现了激增，若将第一总值和第二总值的差作为寄件波动值，则会判断相比A2，A1近期的寄件量出现了激增。
步骤S203、根据每个候选物品的寄件波动值和物流信息中的候选物品平均寄件量，确定每个候选物品的全局波动值。
候选物品平均寄件量是指所有候选物品的平均寄件值。示例性地，对于每个候选物品，电子设备可以将其对应的寄件量进行求和，得到每个候选物品的寄件总量(相当于每个候选物品的第一总值和第二总值的总值和)，并将每个候选物品的寄件总量进行求和，得到所有候选物品的寄件总量，再将所有候选物品的寄件总量与候选物品的数量之间的比值作为候选物品平均寄件量。以表1为例进行示例性说明，其中，候选物品“苹果”的寄件总量为3，候选物品“鱼类”的寄件总量为1，候选物品“橙子”的寄件总量为2，求和后得到所有候选物品的寄件总量为6，而候选物品的数量为3，因此对于表1，其对应的候选物品平均寄件量为2。
候选物品平均寄件量用于解决小样本候选物品的寄件波动值不准确的问题，小样本候选物品是指对应寄件总量较少，即第一总值和第二总值的和较小的候选物品。由于小样本候选物品对应的寄件总量较少，因此根据式子(1)计算得到寄件波动值时，即使较小，但是同样较小，因此计算得到的Si可能反而会偏大。为了方便理解，以候选物品B1和B2为例进行说明，假设B1的第一总值为1，第二总值为0，B2的第一总值为50，第二总值为5，则对于候选物品B1，计算得到的S1为e，对于候选物品B2，计算得到的S2为S1反而大于S2，电子设备会判定相比B2，B1近期的寄件量出现了激增，然而从B1和B2各自的总值可以看出，B2近期的寄件量出现了激增，而B1近期的寄件量未产生较大的变化，若直接根据式子(1)计算得到的寄件波动值，确定待推荐的物品，则会将未出现激增的候选物品推荐给用户。
全局波动值是通过候选物品平均寄件量对寄件波动值进行修正后，得到的更加准确的波动值，可以理解为通过候选物品平均值对每个候选物品的寄件总量进行扩充，以避免出现小样本候选物品后，对每个候选物品计算得到的用于表示寄件量随时间变化情况的值。
在一些实施例中，可以通过候选物品平均寄件量，每个候选物品的目标寄件量和每个候选物品的历史寄件量，计算得到用于修正寄件波动值的系数，并根据得到的系数和寄件波动值，计算得到全局波动值。此时，参考图4，步骤S203“根据所述每个候选物品的寄件波动值和所述物流信息中的候选物品平均寄件量，确定所述每个候选物品的全局波动值”可以通过以下步骤进行：
步骤S2031、统计物流信息中各候选物品的物品数量。
各候选物品的物品数量可以参考上文中的解释，以表1为例，表1对应的物流信息中的物品数量为3。
步骤S2032、根据各候选物品的寄件总量和所述物品数量，计算得到物流信息中的候选物品平均寄件量。
各候选物品的寄件总量是指上文中对每个候选物品计算第一总值和第二总值的寄件总量之后，对每个候选物品的寄件总量求和后得到的寄件量。以表1为例，表1对应的物流信息中，各候选物品的寄件总量为6。
将各候选物品的寄件总量和物品数量作比值后，即可得到候选拿物品平均寄件量。
步骤S2033、根据候选物品平均寄件量，以及每个候选物品的目标寄件量和每个候选物品的历史寄件量，计算得到每个候选物品的波动修正系数。
示例性地，电子设备在执行步骤S2033时，可以通过上文的方法，根据目标寄件量和历史寄件量，分别计算得到每个候选物品对应的第一总值和第二总值，然后计算每个候选物品对应的总值和，以及总值差，并通过式子(2)计算得到波动修正系数：
其中，Ti是指第i个候选物品的波动修正系数，是指第i个候选物品的第一总值，是指第i个候选物品的第二总值，avg是指候选物品平均寄件量。为了方便理解，下文中若未作特别声明，则认为通过式子(2)计算得到波动修正系数，但不能理解为对本申请实施例的限制。
可见，通过式子(2)，可以通过候选物品平均寄件量对候选物品的寄件总量进行扩充，计算近期寄件量的增长值/减少值在扩充后寄件总量中的占比，进而判断近期寄件量的增长值/减少值的大小程度，波动修正系数越大，说明近期寄件量的增长值/减少值的大小程度越大，波动修正系数越小，说明近期寄件量的增长值/减少值的大小程度越小，进而避免小样本候选物品由于寄件总量较少而导致推荐不准确的问题。
在一些实施例中，还可以通过式子(3)，即在式子(2)中的分母中再添加预设的扩充值，以得到全局波动值，以在各候选物品中包括多个小样本候选物品时，进一步扩充每一个候选物品的寄件总量，避免候选物品平均寄件量仍然较小，导致推荐不准确的问题：
其中，Ti是指第i个候选物品的波动修正系数，是指第i个候选物品的第一总值，是指第i个候选物品的第二总值，avg是指候选物品平均寄件量，W是预设的扩充值，可以根据实际场景的需求进行设置。
步骤S2034、根据所述波动修正系数，对所述每个候选物品的寄件波动值进行加权，得到所述每个候选物品的全局波动值。
示例性地，可以通过式子(4)，计算得到每个候选物品的全局波动值：
Gi＝SiTi    式子(4)
其中，Gi是指第i个候选物品的全局波动值，Si是指第i个候选物品的寄件波动值，Ti是指第i个候选物品的波动修正系数。
步骤S204、根据每个候选物品的全局波动值，从各候选物品中选择待推荐的目标物品。
待推荐的目标物品可以理解为当前热门的托寄物品。
在一些实施例中，电子设备可以选择全局波动值最大的候选物品，将该候选物品作为目标物品。
在另一些实施例中，电子设备可以将各候选物品的全局波动值与预设的波动阈值进行对比，将全局波动值大于波动阈值的候选物品作为目标物品。其中，预设的波动阈值用于评估全局波动值的大小，具体数值可以根据实际场景的需要进行设置。
因此，参考图5，步骤S204“根据所述每个候选物品的全局波动值，从所述各候选物品中选择待推荐的目标物品”可以包括：
步骤S2041、将所述每个候选物品的全局波动值与预设分数阈值进行对比，得到全局波动值大于预设分数阈值的候选物品。
其中，预设分数阈值即为上文中的波动阈值，具体不进行赘述。
步骤S2042、将全局波动值大于预设分数阈值的候选物品设定为待推荐的目标物品。
可见，步骤S201-步骤S204的方法除了可以提高推荐的准确度之外，由于不需要获取特定用户的历史数据即可选择待推荐的目标物品，因此还可以实现寄件时托寄物品填写的冷启 动，即可以推荐用户未填写过的候选物品。例如对于一些具有时令属性的候选物品，比如粽子、月饼等，可以在特定时令之前将其识别为待推荐的目标物品，以解决用户冷启动问题。
综上所述，本申请实施例提供的物品推荐方法包括：获取物流信息；根据物流信息中每个候选物品的寄件时间和寄件量，确定每个候选物品的寄件波动值；根据每个候选物品的寄件波动值和物流信息中的候选物品平均寄件量，确定每个候选物品的全局波动值；根据每个候选物品的全局波动值，从各候选物品中选择待推荐的目标物品。
可见，一方面，本申请实施例提供的物品推荐方法可以在用户填写托寄物时，为用户进行托寄物品推荐，减少用户填写运单的时间，并且在确定推荐的目标物品时，通过候选物品平均寄件量对每个候选物品的寄件波动值进行修正，得到更加能够准确表示寄件量随时间变化的全局波动值，解决了小样本候选物品的寄件波动值无法准确表示寄件量随时间变化的问题，进而得到的目标物品更加准确。另一方面，本申请实施例无需根据特定用户的信息即可确定待推荐的目标物品，因此可以实现托寄物品填写的冷启动。
在一些实施例中，电子设备可以首先对物流信息中包含的托寄物品进行筛选，将其中不容易出现寄件量突变的托寄物品作为候选物品。参考图6，此时，在步骤S202“根据所述物流信息中每个候选物品的寄件时间和寄件量，确定所述每个候选物品的寄件波动值”之前，所述方法还包括：
步骤S301、根据物流信息中第一物品的寄件时间，统计第一物品分别在各预设历史时间段中的寄件量。
示例性地，可以根据预设的时间间隔对物流信息对应的时间范围进行划分，得到各预设历史时间段。例如，若电子设备将当前时间之前一个月内的历史运单信息作为物流信息，即物流信息对应的时间范围是指一个月，并且预设的时间间隔为5天，则可以将一个月划分为多个预设历史时间段，每个预设历史时间段对应该月中的5天。
其中，在步骤S301-步骤S303中，第一物品是指物流信息中包含的所有托寄物品。以表1为例，对于表1对于的物流信息，第一物品包括“苹果”、“鱼类”和“橙子”。
第一物品的寄件时间的说明可以参考候选物品的寄件时间，具体不进行赘述。
第一物品在预设历史时间段中的寄件量是指在该预设历史时间段中，第一物品的寄件次数。以表1为例，假设将表1中物流信息对应的时间范围按预设的时间间隔1天进行划分，得到两个预设历史时间段“1月1日”和“1月2日”，则对于第一物品“苹果”，在预设时间段“1月1日”中的寄件量包括“苹果”在1月1日内的寄件次数，对于第一物品“鱼类”，在预设时间段“1月1日”中的寄件量包括“鱼类”在1月1日内的寄件次数，对于第一物品“橙子”，在预设时间段“1月1日”中的寄件量包括“橙子”在1月1日内的寄件次数。
步骤S302、根据各预设历史时间段中的寄件量，以及各预设历史时间段对应的时间间隔，计算得到第一物品的寄件量平均变化率。
其中，各预设历史时间段对应的时间间隔是指上文中预设的时间间隔，即每个预设历史时间段内包含的时间范围。
示例性地，电子设备可以根据每两个相邻的预设历史事件段中的寄件量，以及时间间隔，计算得到每两个相邻的预设历史事件段对应的寄件量变化率，然后将各预设历史时间段对应的所有寄件量变化率进行平均处理，得到第一物品的寄件量平均变化率。为了方便理解，以下具体举一例进行示例性说明，但不能理解为对本申请实施例的限制：假设共有3个预设历史时间段a、b、c，并且第一物品在预设历史时间段a、b、c中的寄件量分别为10、20、30，并且预设历史时间段对应的时间间隔为5天，则计算后，得到的每两个相邻的预设历史事件段对应的寄件量变化率分别为a和b之间的寄件量变化率2/天，以及b和c之间的寄件量变化率2/天，再进行平均计算后，可以得到第一物品的寄件量平均变化率为2/天。
步骤S303、将寄件量平均变化率小于预设变化率阈值的第一物品设定为候选物品。
其中，预设变化率阈值用于评估寄件量平均变化率的大小，具体的数值可以根据实际场景需求进行设置。
若寄件量平均变化率大于或者等于预设变化率阈值，则说明对应第一物品的寄件量经常发生突变，由于本申请实施例提供的物品推荐方法基于寄件量的变化情况确定目标物品，因此对于寄件量经常发生突变的第一物品，无法通过本申请实施例提供的物品推荐方法进行判断，需要将其筛除，避免出现推荐错误，并且可以减少计算量。
在一些实施例中，还可以根据待推荐用户的寄件记录，以及待推荐用户所属用户群体的寄件记录，进一步筛除待推荐用户寄件概率较低的第一物品，以降低计算量。此时，如图7所示，步骤S303“将寄件量平均变化率小于预设变化率阈值的第一物品设定为候选物品”可以包括：
步骤S3031、从待推荐的目标用户的寄件记录中，获取历史寄件总量小于第一预设次数阈值的第二物品。
其中，待推荐的目标用户可以是指打开快递物流软件的用户，具体可以参考步骤S201中的说明。
寄件记录可以包含用户的历史寄件行为数据，例如，可以包含用户创建的运单数据。因此，目标用户的寄件记录中，包含了目标用户创建的历史运单信息。此处的历史运单信息可以是指目标用户创建过的所有运单，也可以是指在当前时间之前的一段时间内，目标用户创建过的运单。
寄件记录可以存储在快递物流软件的后台数据库中，在执行步骤S3031时，电子设备可以根据目标用户的用户身份标识，查询该后台数据库，得到目标用户的寄件记录。
其中，用户身份标识用于区别不同用户的身份，可以是指用户登录快递物流软件时采用的登录名等等。
历史寄件总量是指在寄件记录中寄件的总次数，在执行步骤S3031时，电子设备可以查询寄件记录中的每一张历史运单，得到寄件记录中的每一张历史运单对应的托寄物品，然后统计目标用户的寄件记录中，不同托寄物品的寄件总次数，得到不同托寄物品各自的历史寄件总量。
其中，第一预设次数阈值用于评估历史寄件总量的大小，具体的数值可以根据实际场景需求进行设置，例如可以将其设置为1。以第一预设次数阈值为1为例，若历史寄件总量小于第一预设次数阈值，则说明目标用户从未寄出对应的托寄物品，即使全局波动值较大，当前寄件量出现激增，目标用户也不一定会选择寄出该托寄物品，因此可以筛除该托寄物品。
步骤S3032、确定目标用户所属的用户群体。
群体是指根据年龄、职业、性别、地域等属性对用户进行划分后得到的用户集合。示例性地，可以根据用户在快递物流软件中自行设置的所属城市，对用户进行划分，得到多个预设群体，然后根据目标用户的用户身份标识，从预设群体中查询得到包含目标用户的用户群体。为了方便理解，在下文中若未作特别声明，则认为预设群体根据地域划分。
步骤S3033、从用户群体的寄件记录中，获取历史同期的平均寄件量小于第二预设次数阈值的第三物品。
寄件记录的说明可以参考上文，具体不进行赘述。在执行步骤S3033时，电子设备可以获取用户群体中的所有用户，并提取用户群体中所有用户的历史寄件行为数据，从中获取第三物品。
历史同期的平均寄件量可以是指历史数年同期，对应托寄物品的平均寄件量。
其中，第二预设次数阈值用于评估历史同期的平均寄件量的大小，具体的数值可以根据实际场景需求进行设置。若历史同期的平均寄件量小于第二预设次数阈值，则说明在用户群体对应的地域，并且在当前时间的历史同期，对应托寄物品的同期寄件量较低，该托寄物品即使为时令物品，在该地域也并非是热门托寄物品，因此可以将其筛除。
步骤S3034、从第一物品中筛除第二物品和第三物品，并将筛除后寄件量平均变化率小于预设变化率阈值的第一物品作为候选物品。
根据寄件量平均变化率选择候选物品的原因可以参考步骤S303，具体不进行赘述。
在一些实施例中，得到待推荐的目标物品后，电子设备可以监听是否接收到待推荐的目标用户的选择指令，在接收到选择指令时，将目标物品中选择指令对应的物品填写至运单的对应区域。参考图8，此时，步骤S204“根据每个候选物品的全局波动值，从各候选物品中选择待推荐的目标物品”之后，所述方法还包括：
步骤S401、接收待推荐的目标用户的选择指令，得到目标物品中选择指令对应的物品。
目标用户的说明可以参考步骤S3031，具体不进行赘述。
其中，选择指令可以是指触屏、语音等等类型的指令，本申请实施例对此不进行限制，例如，用户可以通过触屏的方式，选择目标物品中的某物品，以发出携带该物品信息的选择指令。
步骤S402、将选择指令对应的物品输入预设运单的预设区域。
其中，预设运单可以是指目标用户新创建的未填写运单。
其中，预设区域可以是指预设运单中用于填写托寄物品的区域。
为了更好实施本申请实施例中的物品推荐方法，在物品推荐方法基础之上，本申请实施例中还提供一种物品推荐装置，如图9所示，为本申请实施例中物品推荐装置的一个实施例结构示意图，该物品推荐装置500包括：
获取单元501，用于获取物流信息；
第一确定单元502，用于根据所述物流信息中每个候选物品的寄件时间和寄件量，确定每个候选物品的寄件波动值；
第二确定单元503，用于根据每个候选物品的寄件波动值和物流信息中的候选物品平均寄件量，确定每个候选物品的全局波动值；
选择单元504，用于根据每个候选物品的全局波动值，从各候选物品中选择待推荐的目标物品。
在本申请一种可能的实现方式中，第一确定单元502还用于：
根据物流信息中每个候选物品的寄件时间，对每个候选物品的寄件量进行划分，得到每个候选物品的目标寄件量和每个候选物品的历史寄件量；
根据每个候选物品的目标寄件量和每个候选物品的历史寄件量，计算得到每个候选物品的寄件波动值。
在本申请一种可能的实现方式中，第二确定单元503还用于：
统计物流信息中各候选物品的物品数量；
根据各候选物品的寄件总量和物品数量，计算得到物流信息中的候选物品平均寄件量；
根据候选物品平均寄件量，以及每个候选物品的目标寄件量和每个候选物品的历史寄件量，计算得到每个候选物品的波动修正系数；
根据波动修正系数，对每个候选物品的寄件波动值进行加权，得到每个候选物品的全局波动值。
在本申请一种可能的实现方式中，选择单元504还用于：
将每个候选物品的全局波动值与预设分数阈值进行对比，得到全局波动值大于预设分数阈值的候选物品；
将全局波动值大于预设分数阈值的候选物品设定为待推荐的目标物品。
在本申请一种可能的实现方式中，第一确定单元502还用于：
根据物流信息中第一物品的寄件时间，统计第一物品分别在各预设历史时间段中的寄件量；
根据各预设历史时间段中的寄件量，以及各预设历史时间段对应的时间间隔，计算得到第一物品的寄件量平均变化率；
将寄件量平均变化率小于预设变化率阈值的第一物品设定为候选物品。
在本申请一种可能的实现方式中，第一确定单元502还用于：
从待推荐的目标用户的寄件记录中，获取历史寄件总量小于第一预设次数阈值的第二物 品；
确定目标用户所属的用户群体；
从用户群体的寄件记录中，获取历史同期的平均寄件量小于第二预设次数阈值的第三物品；
从第一物品中筛除第二物品和第三物品，并将筛除后寄件量平均变化率小于预设变化率阈值的第一物品作为候选物品。
在本申请一种可能的实现方式中，选择单元504还用于：
接收待推荐的目标用户的选择指令，得到所述目标物品中所述选择指令对应的物品；
将选择指令对应的物品输入预设运单的预设区域。
具体实施时，以上各个单元可以作为独立的实体来实现，也可以进行任意组合，作为同一或若干个实体来实现，以上各个单元的具体实施可参见前面的方法实施例，在此不再赘述。
由于该物品推荐装置可以执行任意实施例中物品推荐方法中的步骤，因此，可以实现本申请任意实施例中物品推荐方法所能实现的有益效果，详见前面的说明，在此不再赘述。
此外，为了更好实施本申请实施例中物品推荐方法，在物品推荐方法
基础之上，本申请实施例还提供一种电子设备，参阅图10，图10示出了本申请实施例电子设备的一种结构示意图，具体的，本申请实施例提供的电子设备包括处理器601，处理器601用于执行存储器602中存储的计算机程序时实现任意实施例中物品推荐方法的各步骤；或者，处理器601用于执行存储器602中存储的计算机程序时实现如图9对应实施例中各单元的功能。
示例性的，计算机程序可以被分割成一个或多个模块/单元，一个或者多个模块/单元被存储在存储器602中，并由处理器601执行，以完成本申请实施例。一个或多个模块/单元可以是能够完成特定功能的一系列计算机程序指令段，该指令段用于描述计算机程序在计算机装置中的执行过程。
电子设备可包括，但不仅限于处理器601、存储器602。本领域技术人员可以理解，示意仅仅是电子设备的示例，并不构成对电子设备的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件。
处理器601可以是中央处理单元(Central Processing Unit，CPU)，还可以是其他通用处理器、数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等，处理器是电子设备的控制中心，利用各种接口和线路连接整个电子设备的各个部分。
存储器602可用于存储计算机程序和/或模块，处理器601通过运行或执行存储在存储器602内的计算机程序和/或模块，以及调用存储在存储器602内的数据，实现计算机装置的各种功能。存储器602可主要包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序(比如声音播放功能、图像播放功能等)等；存储数据区可存储根据电子设备的使用所创建的数据(比如音频数据、视频数据等)等。此外，存储器可以包括高速随机存取存储器，还可以包括非易失性存储器，例如硬盘、内存、插接式硬盘，智能存储卡(Smart Media Card，SMC)，安全数字(Secure Digital，SD)卡，闪存卡(Flash Card)、至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的物品推荐装置、电子设备及其相应单元的具体工作过程，可以参考任意实施例中物品推荐方法的说明，具体在此不再赘述。
本领域普通技术人员可以理解，上述实施例的各种方法中的全部或部分步骤可以通过指令来完成，或通过指令控制相关的硬件来完成，该指令可以存储于一存储介质中，并由处理器进行加载和执行。
为此，本申请实施例提供一种存储介质，存储介质上存储有计算机程序，该计算机程序被处理器执行时执行本申请任意实施例中物品推荐方法中的步骤，具体操作可参考任意实施例中物品推荐方法的说明，在此不再赘述。
其中，该存储介质可以包括：只读存储器(Read Only Memory,ROM)、随机存取存储器(Random Access Memory,RAM)、磁盘或光盘等。
由于该存储介质中所存储的指令，可以执行本申请任意实施例中物品推荐方法中的步骤，因此，可以实现本申请任意实施例中物品推荐方法所能实现的有益效果，详见前面的说明，在此不再赘述。
以上对本申请实施例所提供的一种物品推荐方法、装置、存储介质及电子设备进行了详细介绍，本文中应用了具体个例对本申请的原理及实施方式进行了阐述，以上实施例的说明只是用于帮助理解本申请的方法及其核心思想；同时，对于本领域的技术人员，依据本申请的思想，在具体实施方式及应用范围上均会有改变之处，综上所述，本说明书内容不应理解为对本申请的限制。

Claims (16)

  1. 一种物品推荐方法，包括：
    获取物流信息；
    根据所述物流信息中每个候选物品的寄件时间和寄件量，确定所述每个候选物品的寄件波动值；
    根据所述每个候选物品的寄件波动值和所述物流信息中的候选物品平均寄件量，确定所述每个候选物品的全局波动值；
    根据所述每个候选物品的全局波动值，从所述各候选物品中选择待推荐的目标物品。
  2. 根据权利要求1所述的物品推荐方法，其中，所述根据所述物流信息中每个候选物品的寄件时间和寄件量，确定所述每个候选物品的寄件波动值，包括：
    根据所述物流信息中每个候选物品的寄件时间，对所述每个候选物品的寄件量进行划分，得到所述每个候选物品的目标寄件量和所述每个候选物品的历史寄件量；
    根据所述每个候选物品的目标寄件量和所述每个候选物品的历史寄件量，计算得到所述每个候选物品的寄件波动值。
  3. 根据权利要求2所述的物品推荐方法，其中，所述根据所述每个候选物品的寄件波动值和所述物流信息中的候选物品平均寄件量，确定所述每个候选物品的全局波动值，包括：
    统计所述物流信息中各候选物品的物品数量；
    根据所述各候选物品的寄件总量和所述物品数量，计算得到所述物流信息中的候选物品平均寄件量；
    根据所述候选物品平均寄件量，以及所述每个候选物品的目标寄件量和所述每个候选物品的历史寄件量，计算得到所述每个候选物品的波动修正系数；
    根据所述波动修正系数，对所述每个候选物品的寄件波动值进行加权，得到所述每个候选物品的全局波动值。
  4. 根据权利要求1至3任一项所述的物品推荐方法，其中，所述根据所述每个候选物品的全局波动值，从所述各候选物品中选择待推荐的目标物品，包括：
    将所述每个候选物品的全局波动值与预设分数阈值进行对比，得到全局波动值大于预设分数阈值的候选物品；
    将全局波动值大于预设分数阈值的候选物品设定为待推荐的目标物品。
  5. 根据权利要求1至4任一项所述的物品推荐方法，其中，所述根据所述物流信息中每个候选物品的寄件时间和寄件量，确定所述每个候选物品的寄件波动值之前，所述方法还包括：
    根据所述物流信息中第一物品的寄件时间，统计所述第一物品分别在各预设历史时间段中的寄件量；
    根据所述各预设历史时间段中的寄件量，以及所述各预设历史时间段对应的时间间隔，计算得到所述第一物品的寄件量平均变化率；
    将寄件量平均变化率小于预设变化率阈值的第一物品设定为候选物品。
  6. 根据权利要求5所述的物品推荐方法，其中，所述将寄件量平均变化率小于预设变化率阈值的第一物品设定为候选物品，包括：
    从待推荐的目标用户的寄件记录中，获取历史寄件总量小于第一预设次数阈值的第二物品；
    确定所述目标用户所属的用户群体；
    从所述用户群体的寄件记录中，获取历史同期的平均寄件量小于第二预设次数阈值的第三物品；
    从所述第一物品中筛除所述第二物品和所述第三物品，并将筛除后寄件量平均变化率小于预设变化率阈值的第一物品作为候选物品。
  7. 根据权利要求1至6任一项所述的物品推荐方法，其中，所述根据所述每个候选物品的全局波动值，从所述各候选物品中选择待推荐的目标物品之后，所述方法还包括：
    接收待推荐的目标用户的选择指令，得到所述目标物品中所述选择指令对应的物品；
    将所述选择指令对应的物品输入预设运单的预设区域。
  8. 一种物品推荐装置，包括：
    获取单元，用于获取物流信息；
    第一确定单元，用于根据所述物流信息中每个候选物品的寄件时间和寄件量，确定所述每个候选物品的寄件波动值；
    第二确定单元，用于根据所述每个候选物品的寄件波动值和所述物流信息中的候选物品平均寄件量，确定所述每个候选物品的全局波动值；
    选择单元，用于根据所述每个候选物品的全局波动值，从所述各候选物品中选择待推荐的目标物品。
  9. 根据权利要求8所述的物品推荐装置，其中，所述第一确定单元还用于：
    根据所述物流信息中每个候选物品的寄件时间，对所述每个候选物品的寄件量进行划分，得到所述每个候选物品的目标寄件量和所述每个候选物品的历史寄件量；
    根据所述每个候选物品的目标寄件量和所述每个候选物品的历史寄件量，计算得到所述每个候选物品的寄件波动值。
  10. 根据权利要求9所述的物品推荐装置，其中，所述第二确定单元还用于：
    统计所述物流信息中各候选物品的物品数量；
    根据所述各候选物品的寄件总量和所述物品数量，计算得到所述物流信息中的候选物品平均寄件量；
    根据所述候选物品平均寄件量，以及所述每个候选物品的目标寄件量和所述每个候选物品的历史寄件量，计算得到所述每个候选物品的波动修正系数；
    根据所述波动修正系数，对所述每个候选物品的寄件波动值进行加权，得到所述每个候选物品的全局波动值。
  11. 根据权利要求8至10中任一项所述的物品推荐装置，其中，所述选择单元还用于：
    将所述每个候选物品的全局波动值与预设分数阈值进行对比，得到全局波动值大于预设分数阈值的候选物品；
    将全局波动值大于预设分数阈值的候选物品设定为待推荐的目标物品。
  12. 根据权利要求8至11中任一项所述的物品推荐装置，其中，所述第一确定单元还用于：
    根据所述物流信息中第一物品的寄件时间，统计所述第一物品分别在各预设历史时间段中的寄件量；
    根据所述各预设历史时间段中的寄件量，以及所述各预设历史时间段对应的时间间隔，计算得到所述第一物品的寄件量平均变化率；
    将寄件量平均变化率小于预设变化率阈值的第一物品设定为候选物品。
  13. 根据权利要求12所述的物品推荐装置，其中，所述第一确定单元还用于：
    从待推荐的目标用户的寄件记录中，获取历史寄件总量小于第一预设次数阈值的第二物品；
    确定所述目标用户所属的用户群体；
    从所述用户群体的寄件记录中，获取历史同期的平均寄件量小于第二预设次数阈值的第三物品；
    从所述第一物品中筛除所述第二物品和所述第三物品，并将筛除后寄件量平均变化率小 于预设变化率阈值的第一物品作为候选物品。
  14. 根据权利要求8至13任一项所述的物品推荐装置，其中，所述选择单元还用于：
    接收待推荐的目标用户的选择指令，得到所述目标物品中所述选择指令对应的物品；
    将选择指令对应的物品输入预设运单的预设区域。
  15. 一种电子设备，所述电子设备包括处理器、存储器以及存储于所述存储器中并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现如权利要求1至7任一项所述的物品推荐方法中的步骤。
  16. 一种存储介质，所述存储介质上存储有计算机程序，所述计算机程序被处理器执行时实现权利要求1至7任一项所述的物品推荐方法中的步骤。
PCT/CN2023/114022 2022-06-24 2023-08-21 物品推荐方法、装置、电子设备及存储介质 WO2023246954A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210738747.1A CN117331980A (zh) 2022-06-24 2022-06-24 物品推荐方法、装置、电子设备及存储介质
CN202210738747.1 2022-06-24

Publications (1)

Publication Number Publication Date
WO2023246954A1 true WO2023246954A1 (zh) 2023-12-28

Family

ID=89277938

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/114022 WO2023246954A1 (zh) 2022-06-24 2023-08-21 物品推荐方法、装置、电子设备及存储介质

Country Status (2)

Country Link
CN (1) CN117331980A (zh)
WO (1) WO2023246954A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020073846A1 (zh) * 2018-10-08 2020-04-16 菜鸟智能物流控股有限公司 一种物流对象信息管理方法和系统
CN112766995A (zh) * 2019-10-21 2021-05-07 招商证券股份有限公司 物品推荐方法、装置、终端设备及存储介质
CN112950241A (zh) * 2019-12-10 2021-06-11 顺丰科技有限公司 虚拟资源的发放方法、装置、计算机设备和存储介质
CN113806621A (zh) * 2020-06-11 2021-12-17 顺丰科技有限公司 信息推荐方法、装置、计算机设备和存储介质
CN113869943A (zh) * 2021-09-16 2021-12-31 北京三快在线科技有限公司 物品推荐方法、装置、设备以及存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020073846A1 (zh) * 2018-10-08 2020-04-16 菜鸟智能物流控股有限公司 一种物流对象信息管理方法和系统
CN112766995A (zh) * 2019-10-21 2021-05-07 招商证券股份有限公司 物品推荐方法、装置、终端设备及存储介质
CN112950241A (zh) * 2019-12-10 2021-06-11 顺丰科技有限公司 虚拟资源的发放方法、装置、计算机设备和存储介质
CN113806621A (zh) * 2020-06-11 2021-12-17 顺丰科技有限公司 信息推荐方法、装置、计算机设备和存储介质
CN113869943A (zh) * 2021-09-16 2021-12-31 北京三快在线科技有限公司 物品推荐方法、装置、设备以及存储介质

Also Published As

Publication number Publication date
CN117331980A (zh) 2024-01-02

Similar Documents

Publication Publication Date Title
US9348898B2 (en) Recommendation system with dual collaborative filter usage matrix
US20200294111A1 (en) Determining target user group
CN104834731B (zh) 一种自媒体信息的推荐方法及装置
WO2020048084A1 (zh) 资源推荐方法、装置、计算机设备及计算机可读存储介质
CN107305637B (zh) 基于K-Means算法的数据聚类方法和装置
CN109408665A (zh) 一种信息推荐方法及装置、存储介质
CN109242612B (zh) 一种产品推荐的方法及设备
WO2018053966A1 (zh) 点击率预估
CN109033165B (zh) 一种数据展示方法、计算机可读存储介质及终端设备
US12079438B2 (en) Interactive display of data distributions
US11669566B2 (en) Multi-resolution color-based image search
WO2022007626A1 (zh) 视频内容推荐方法、装置及计算机设备
CN112488863A (zh) 一种在用户冷启动场景下的险种推荐方法及相关设备
CN111160699A (zh) 一种专家推荐方法及系统
WO2023246954A1 (zh) 物品推荐方法、装置、电子设备及存储介质
JP2014222474A (ja) 情報処理装置、方法及びプログラム
CN109783175B (zh) 应用程序图标管理方法、装置、可读存储介质及终端设备
WO2017118439A1 (zh) 信息发布方法、信息发布装置和存储介质
CN117112942A (zh) 用户界面调整方法、装置、计算机设备和存储介质
WO2019237534A1 (zh) 一种系统组件显示方法、可读存储介质、终端设备及装置
CN113986932B (zh) 数据处理方法、装置、计算机设备及可读存储介质
CN113342762B (zh) 针对商户好友的共享消息的智能处理方法及相关装置
US20230052619A1 (en) Real-time error prevention during invoice creation
CN110298679A (zh) 一种计算样本数据之间的距离的方法及设备
WO2017219317A1 (zh) 一种基于搜索内容的信息推送方法及装置

Legal Events

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

Ref document number: 23826598

Country of ref document: EP

Kind code of ref document: A1