Member preference-based express delivery method, device and equipment and storage medium

Abstract

The invention relates to the technical field of express delivery, in particular to an express delivery method and device based on member preferences, equipment and a storage medium. The member preference-based express delivery method comprises the following steps of adopting a preference classification model, generating member receiving recommendation information and a classification result according to historical purchase data, and outputting the receiving recommendation information to obtain a member first selection result; adopting a delivery recommendation model to generate recommended delivery information according to the receiving information and the classification result, and outputting the recommended delivery information to obtain a second member selection result; adjusting labels of the members according to the visual identification data to obtain label information; and generating delivery information according to the delivery instruction, the first member selection result, the second member selection result and the label information. According to the express delivery method, a plurality of factors are considered, so that the delivery information better conforms to the preferences of members to the greatest extent, and the satisfaction degree of the members to the delivery service is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of express delivery, and particularly to an express delivery method, device, equipment and storage medium based on member preferences. Background Art

[0002] At present, with the booming development of the express delivery industry, the market competition is becoming increasingly fierce. Improving the user experience has become the key for each express delivery enterprise to stand out. Especially for the member group, it is particularly important to meet their diverse needs, which makes precise express delivery management the core issue in the industry development.

[0003] At present, the needs of members in express delivery show significant diversification characteristics. In terms of delivery methods, some members are persistent in door-to-door delivery and enjoy the convenience of having the express delivered directly to their doorstep. While some members, due to busy work or special living environments, prefer designated location delivery, such as delivering to the company front desk, the property management of the community, etc. There are also many members who prefer to have the express delivered to a designated express cabinet or collection point, which gives them greater time flexibility and is convenient for them to pick up the package when they are free. In terms of delivery time, some members hope that the express can be delivered during specific time periods on weekdays to avoid affecting work; while some members prefer weekend or evening delivery to ensure that they can receive it in time when they are at home. In addition, for pre-delivery phone calls, some members think it is a necessary communication link and can make preparations for receiving in advance; while some members think it will disturb them and hope to reduce such calls. However, the current logistics industry is difficult to provide precise services, resulting in the failure to meet the internal requirements of the member group.

[0004] Meanwhile, the label data of members comes from a wide range of sources, including the shopping records of members on e-commerce platforms, operation behaviors on the express delivery enterprise APP, historical delivery feedback, etc. There are many problems in the management and application of the label data of members from these different sources. On the one hand, the data formats are not unified, some are recorded in text form, and some are presented in digital codes, which brings great difficulties to the integration and analysis of the data. On the other hand, there are situations of data duplication and conflicts. For example, the definitions of the consumption ability labels of the same member from different channels are different, resulting in the inability to accurately judge the true consumption level of the member. In addition, the data is not updated in a timely manner and cannot reflect the latest needs and preference changes of members in real time, making the delivery strategies formulated based on these data lag behind.

[0005] It can be seen that the existing technology still needs to be improved. Summary of the Invention

[0006] In view of the above deficiencies of the existing technology, the purpose of the present invention is to provide an express delivery method, device, equipment and storage medium based on member preferences, aiming to solve the problem that the express delivery service in the existing technology fails to accurately target member preferences.

[0007] To achieve the above object, the present invention adopts the following technical solutions:

[0008] In a first aspect of the present invention, there is provided a courier delivery method based on member preferences, including the following steps: obtaining historical purchase data of a member, using a preference classification model, generating receiving recommendation information and a classification result of the member according to the historical purchase data, and outputting the receiving recommendation information to obtain a first selection result of the member; obtaining receiving information specified by the member, using a delivery recommendation model, generating recommended delivery information according to the receiving information and the classification result, and outputting the recommended delivery information to obtain a second selection result of the member; obtaining visual identification data of the member, adjusting the label of the member according to the visual identification data to obtain label information; obtaining a delivery instruction, and generating delivery information according to the delivery instruction, the first selection result of the member, the second selection result of the member, and the label information.

[0009] Optionally, in a first implementation manner of the first aspect of the present invention, the step of obtaining receiving information specified by the member, using a delivery recommendation model, generating recommended delivery information according to the receiving information and the classification result, and outputting the recommended delivery information to obtain a second selection result of the member specifically includes: obtaining receiving information specified by the member, where the receiving information includes required receiving date data and required receiving address data; constructing a delivery recommendation model, where the delivery recommendation model includes a time series analysis sub-model and a location understanding sub-model; using the time series analysis sub-model to generate a recommended delivery time according to the required receiving date data and the classification result; using the location understanding sub-model to generate a recommended delivery location according to the required receiving address data; outputting the recommended delivery time and the recommended delivery location to obtain a second selection result of the member.

[0010] Optionally, in a second implementation manner of the first aspect of the present invention, the step of using the time series analysis sub-model to generate a recommended delivery time according to the required receiving date data and the classification result specifically includes: obtaining the required receiving date data, preprocessing the required receiving date data to obtain preprocessed data; obtaining multiple member receiving data, classifying and labeling the member receiving data to construct a member receiving database; using the time series analysis sub-model to generate a recommended delivery time according to the preprocessed data, the classification result, and the actual delivery time of the courier in the member receiving database.

[0011] Optionally, in the third implementation manner of the first aspect of the present invention, the method of using the location understanding sub-model to generate a recommended delivery location according to the required delivery address data specifically includes: obtaining the required delivery address data, using the location understanding sub-model to preprocess the required delivery address data according to natural language processing technology to obtain the first address information; using the location understanding sub-model to identify the location type of the first address information according to the semantic understanding algorithm to obtain the second address information; obtaining map data, and using the location understanding sub-model to perform fuzzy matching on the second address information in combination with the map data to generate an accurate recommended delivery location.

[0012] Optionally, in the fourth implementation manner of the first aspect of the present invention, the method of obtaining the visual identification data of the member, adjusting the label of the member according to the visual identification data, and obtaining the label information specifically includes: obtaining the visual identification data of the member, using the image feature extraction model to identify the features in the visual identification data to obtain the first source information; obtaining the data file related to the member, extracting the features in the data file to obtain the second source information; using the association analysis model to determine the original data source of the label of the member according to the first source information and the second source information; using the rule model constructed by the decision tree algorithm to adjust the label of the member according to the preset rules and the original data source to obtain the label information.

[0013] Optionally, in the fifth implementation manner of the first aspect of the present invention, the method of obtaining the delivery instruction, and generating the delivery information according to the delivery instruction, the first selection result of the member, the second selection result of the member, and the label information specifically includes: obtaining the delivery instruction, using the location generation algorithm to generate the delivery location according to the first selection result of the member and the second selection result of the member; obtaining the current traffic condition, using the Dijkstra algorithm to generate the optimal path according to the current traffic condition and the delivery location, and generating the delivery information with the delivery location, the optimal path, and the label information; continuously obtaining the latest weather condition information and the latest traffic condition information, and adjusting the delivery information according to the latest weather condition information and the latest traffic condition information.

[0014] Optionally, in the sixth implementation manner of the first aspect of the present invention, continuously obtaining the latest weather condition information and the latest traffic condition information, and adjusting the delivery information according to the latest weather condition information and the latest traffic condition information specifically includes: obtaining historical delivery record data, training a machine learning algorithm using the historical delivery record data to construct a delivery status prediction model; obtaining the latest weather condition information and the latest traffic condition information, using the delivery status prediction model, predicting the delivery information according to the latest weather condition information and the latest traffic condition information, generating a warning information according to the prediction result; formulating a warning response rule, adjusting the delivery information according to the warning information and the warning response rule, and outputting the adjusted delivery information to the delivery staff.

[0015] The second aspect of the present invention provides an express delivery device based on member preferences, including: a classification module, configured to obtain the historical purchase data of a member, use a preference classification model to generate a receiving recommendation information and a classification result of the member according to the historical purchase data, and output the receiving recommendation information to obtain a first selection result of the member; an analysis module, configured to obtain the receiving information specified by the member, use a delivery recommendation model to generate a recommended delivery information according to the receiving information and the classification result, and output the recommended delivery information to obtain a second selection result of the member; a label module, configured to obtain the visual identification data of the member, and adjust the label of the member according to the visual identification data to obtain label information; a delivery module, configured to obtain a delivery instruction, and generate a delivery information according to the delivery instruction, the first selection result of the member, the second selection result of the member, and the label information.

[0016] Optionally, in the first implementation manner of the second aspect of the present invention, the analysis module includes: an obtaining unit, configured to obtain the receiving information specified by the member, where the receiving information includes a required receiving date data and a required receiving address data; a constructing unit, configured to construct a delivery recommendation model, where the delivery recommendation model includes a time series analysis sub-model and a location understanding sub-model; a first generating unit, configured to use the time series analysis sub-model to generate a recommended delivery time according to the required receiving date data and the classification result; a second generating unit, configured to use the location understanding sub-model to generate a recommended delivery location according to the required receiving address data; and an output unit, configured to output the recommended delivery time and the recommended delivery location to obtain a second selection result of the member.

[0017] Optionally, in the second implementation manner of the second aspect of the present invention, the first generation unit includes: a preprocessing subunit, configured to obtain the required delivery date data and perform preprocessing on the required delivery date data to obtain preprocessed data; a labeling subunit, configured to obtain a plurality of member receiving data, classify and label the member receiving data to construct a member receiving database; a generation subunit, configured to use a time series analysis submodel to generate a recommended delivery time according to the preprocessed data, the classification result, and the actual delivery time of the express in the member receiving database.

[0018] Optionally, in the third implementation manner of the second aspect of the present invention, the second generation unit includes: an address processing subunit, configured to obtain the required delivery address data, and use a location understanding submodel to perform preprocessing on the required delivery address data according to natural language processing technology to obtain first address information; an identification subunit, configured to use a location understanding submodel to identify the location type of the first address information according to a semantic understanding algorithm to obtain second address information; a matching subunit, configured to obtain map data, and use a location understanding submodel to perform fuzzy matching on the second address information in combination with the map data to generate an accurate recommended delivery location.

[0019] Optionally, in the fourth implementation manner of the second aspect of the present invention, the label module includes: a first extraction unit, configured to obtain the visual identification data of the member, and use an image feature extraction model to identify the features in the visual identification data to obtain first source information; a second extraction unit, configured to obtain the data file related to the member and extract the features in the data file to obtain second source information; an association analysis unit, configured to use an association analysis model to determine the original data source of the member's label according to the first source information and the second source information; an adjustment unit, configured to use a rule model constructed by a decision tree algorithm to adjust the member's label according to preset rules and the original data source to obtain label information.

[0020] Optionally, in the fifth implementation manner of the second aspect of the present invention, the delivery module includes: a location generation unit, configured to obtain a delivery instruction, and use a location generation algorithm to generate a receiving location according to the first selection result and the second selection result of the member; a delivery generation unit, configured to obtain the current traffic condition, and use the Dijkstra algorithm to generate an optimal path according to the current traffic condition and the receiving location, and generate delivery information with the receiving location, the optimal path, and the label information; a scheduling unit, configured to continuously obtain the latest weather condition information and the latest traffic condition information, and adjust the delivery information according to the latest weather condition information and the latest traffic condition information.

[0021] Optionally, in the sixth implementation manner of the second aspect of the present invention, the scheduling unit includes: a construction subunit, configured to obtain historical delivery record data and train a machine learning algorithm using the historical delivery record data to construct a delivery status prediction model; a prediction subunit, configured to obtain the latest weather condition information and the latest traffic condition information, use the delivery status prediction model to predict the delivery information according to the latest weather condition information and the latest traffic condition information, and generate a warning information according to the prediction result; a scheduling subunit, configured to formulate a warning response rule, adjust the delivery information according to the warning information and the warning response rule, and output the adjusted delivery information to the deliveryman.

[0022] The third aspect of the present invention provides an express delivery device based on member preferences, including a memory and at least one processor, wherein computer-readable instructions are stored in the memory; the at least one processor calls the computer-readable instructions in the memory to execute each step of the above-mentioned express delivery method based on member preferences.

[0023] The fourth aspect of the present invention provides a computer-readable storage medium, on which computer-readable instructions are stored, and when the computer-readable instructions are executed by a processor, each step of the above-mentioned express delivery method based on member preferences is implemented.

[0024] Beneficial effects: The present invention provides an express delivery method based on member preferences. The express delivery method first uses a preference classification model to generate a member's receiving recommendation information and classification result according to historical purchase data, outputs the receiving recommendation information to obtain the member's first selection result, and obtains the member's basic preference information; then uses a delivery recommendation model to generate recommended delivery information according to the receiving information and classification result, outputs the recommended delivery information to obtain the member's second selection result, and obtains the member's preference information for specific express delivery; then obtains the member's visual identification data, adjusts the member's label according to the visual identification data to obtain label information, and clarifies the source of the member information; finally, generates delivery information that meets the member's preferences according to the delivery instruction, the member's first selection result, the member's second selection result, and the label information, thereby improving the quality of the express company's service. Description of the Drawings

[0025] Figure 1 It is the first flowchart of the express delivery method based on member preferences provided by the embodiment of the present invention;

[0026] Figure 2 It is the second flowchart of the express delivery method based on member preferences provided by the embodiment of the present invention;

[0027] Figure 3The third flowchart of the express delivery method based on member preferences provided by the embodiments of the present invention;

[0028] Figure 4 The fourth flowchart of the express delivery method based on member preferences provided by the embodiments of the present invention;

[0029] Figure 5 The fifth flowchart of the express delivery method based on member preferences provided by the embodiments of the present invention;

[0030] Figure 6 The sixth flowchart of the express delivery method based on member preferences provided by the embodiments of the present invention;

[0031] Figure 7 The seventh flowchart of the express delivery method based on member preferences provided by the embodiments of the present invention;

[0032] Figure 8 A schematic structural diagram of an express delivery device based on member preferences provided by the embodiments of the present invention;

[0033] Figure 9 Another schematic structural diagram of an express delivery device based on member preferences provided by the embodiments of the present invention;

[0034] Figure 10 A schematic structural diagram of an express delivery device based on member preferences provided by the embodiments of the present invention. Detailed implementation manners

[0035] The present invention provides an express delivery method, device, equipment and storage medium based on member preferences. First, by using a preference classification model, the present invention generates receiving recommendation information and classification results for members according to historical purchase data, outputs the receiving recommendation information to obtain the first selection result of the members, and the first selection result corresponds to the basic preference information of the members; then, by using a delivery recommendation model, the present invention generates recommended delivery information according to the receiving information and classification results, outputs the recommended delivery information to obtain the second selection result of the members, and obtains the preference information of the members for each specific express delivery, so as to accurately meet the delivery preferences of the members in terms of address and time dimensions. Then, by obtaining the visual identification data of the members, the present invention adjusts the tags of the members according to the visual identification data to obtain tag information, clarifies the source of the member information, and optimizes data management; finally, according to the delivery instructions, the first selection result of the members, the second selection result of the members and the tag information, the present invention generates delivery information that meets the member preferences, so that the delivery personnel can perform deliveries according to the clear preference information of the members, reducing delivery errors and delays caused by unclear information, thereby improving the overall efficiency of express delivery.

[0036] In the description of the present invention, the claims and the above-mentioned drawings, the terms "first", "second", "third", "fourth", etc. (if any) are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments described herein can be implemented in an order other than that shown or described herein. In addition, the term "comprising" or "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily limit to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0037] For ease of understanding, the specific process of the embodiments of the present invention will be described below. Please refer to Figure 1 , the first embodiment of the express delivery method based on member preferences in the embodiments of the present invention includes:

[0038] S101. Obtain the historical purchase data of the member, use the preference classification model, generate the receiving recommendation information and classification result of the member according to the historical purchase data, and output the receiving recommendation information to obtain the first selection result of the member;

[0039] Specifically, the historical purchase data of the member includes the categories, price ranges, purchase frequencies, etc. of the goods purchased in the past period. The preference classification model based on machine learning can analyze and predict according to the historical purchase data of the member. For example, if a member often buys fresh food and is older, the model may give priority to recommending home delivery and delivery time during the day. At this time, the member can select the receiving recommendation information generated by the preference classification model in the system or independently select other receiving methods and times, so as to obtain the first selection result of the member. The first selection result is a regular preference selection, such as the preferred delivery time period within a day and the delivery method. In addition, the preference classification model can also classify the member according to the historical purchase data to form a classification result.

[0040] S102. Obtain the receiving information designated by the member, use the delivery recommendation model, generate the recommended delivery information according to the receiving information and the classification result, and output the recommended delivery information to obtain the second selection result of the member;

[0041] After a member purchases a product, they need to fill in the delivery information. For each delivery information, the delivery recommendation model generates recommended delivery information based on the delivery information and the classification result. In addition to the usual delivery address, the delivery information can also include the delivery on weekdays or rest days set by the member. Based on the classification result, the delivery recommendation model can find the preferences of similar members, such as more specific delivery times. For example, for members classified as office workers, recommend the rest days selected by the members, and recommend a more accurate location nearby based on the delivery address for the member to confirm. Similarly, the member can directly select the delivery time and specific delivery location recommended by the system, or manually modify them to obtain the member's second-choice result;

[0042] S103. Obtain the visual identification data of the member, adjust the tags of the member according to the visual identification data, and obtain tag information;

[0043] The tags of members often have multiple source channels, and adjusting the tags of members can help sort out the tags, which is conducive to the management and utilization of data. In this embodiment, the visual identification data of the member can be obtained through the system. If there is relevant visual identification in the portrait background, the source of the tag data can be clarified and the tag data of the member can be adjusted.

[0044] S104. Obtain a delivery instruction, and generate delivery information according to the delivery instruction, the member's first-choice result, the member's second-choice result, and the tag information.

[0045] When a delivery instruction is received, it means that the express package has entered the network point and can be delivered. At this time, the system forms specific delivery information according to the delivery instruction, the member's first-choice result, the member's second-choice result, and the tag information. The member's first-choice result includes the preferred delivery method and the preferred delivery time period of the member, which will also be considered at this time. The member's second-choice result includes preferences for weekdays or rest days, delivery location preferences, etc. The tag information can include information such as whether the member needs to be contacted by phone in advance. By comprehensively considering the above factors, accurate delivery information that meets the customer's preferences can be generated for the courier to execute.

[0046] Please refer to Figure 2 , the second embodiment of the express delivery method based on member preferences in the embodiment of the present invention includes:

[0047] S201. Obtain the delivery information specified by the member, where the delivery information includes required delivery date data and required delivery address data;

[0048] Specifically, before receiving the express delivery, the member can specify the delivery date and the delivery address. The delivery date can be specific days or a certain time period within a day, such as from 8:00 to 10:00. The delivery date requested by the member here is only for reference, and the specific arrangement will be made in combination with the actual delivery allocation.

[0049] S202. Construct a delivery recommendation model, where the delivery recommendation model includes a time series analysis sub-model and a location understanding sub-model;

[0050] S203. Use the time series analysis sub-model to generate a recommended delivery time according to the required delivery date data and the classification result;

[0051] The classification result includes the pre-classified member types. For office workers, the time series analysis sub-model can further generate a more suitable recommended delivery time for office workers according to the delivery date data and the classification result. For example, if there are multiple delivery dates specified by office workers, the time series analysis sub-model will give priority to recommending rest days and recommend the time period when the member usually picks up the package among the rest days, so that the recommended delivery time is more in line with the needs of office workers. If only working days are available, the nearest day will be recommended and delivery to the door will be recommended after the member gets off work.

[0052] S204. Use the location understanding sub-model to generate a recommended delivery location according to the required delivery address data;

[0053] Through the location understanding sub-model, the address entered by the member can be quickly converted into an accurate address. For example, some members may enter misspelled words or vague terms (such as "nearby", "next to"). The location understanding sub-model can process these errors or unclear contents to ensure the generation of an accurate address.

[0054] S205. Output the recommended delivery time and the recommended delivery location to obtain the member's second-choice result.

[0055] When the system forms the recommended delivery time and the recommended delivery location, it will output them. After the member confirms, the second-choice result is obtained. The member can also make manual adjustments according to the recommended delivery time and the recommended delivery location. The recommended delivery time and the recommended delivery location can reduce the member's operations and improve the convenience of system operations.

[0056] Please refer to Figure 3 , the third embodiment of the express delivery method based on member preferences in the embodiments of the present invention includes:

[0057] S301. Obtain the required delivery date data, and preprocess the required delivery date data to obtain preprocessed data;

[0058] The time series analysis sub-model first checks the correctness of the time format. If the time format is incorrect, correction processing is performed.

[0059] S302. Obtain multiple member receiving data, classify and label the member receiving data to construct a member receiving database;

[0060] The receiving data of members includes multiple contents, such as the type of goods, the receiving time, the receiving location, the member type (such as office workers, stay-at-home people who don't need to go to work, the elderly, college students), etc. By classifying and re-labeling the receiving data, it can be used for comparison by the time series analysis sub-model, so as to generate a delivery time that better meets the member's preferences for new express orders.

[0061] S303. Use the time series analysis sub-model to generate a recommended delivery time according to the preprocessed data, the classification result, and the actual delivery time of the express in the member receiving database.

[0062] The time series analysis sub-model will comprehensively consider the preprocessed data, the classification result, and the actual delivery time of the express in the member receiving database to generate a recommended delivery time. For example, for members whose classification result is office workers, the time series analysis sub-model will compare with the actual delivery time of the express of similar members in the member receiving database, so as to initially form a time range that conforms to the preferences of this member, and then further combine the preprocessed data to obtain the recommended delivery time. By comprehensively considering, the recommended delivery time can be made not to deviate too much from the delivery time expected by the member, and at the same time it also conforms to the actual delivery capacity of the network.

[0063] Please refer to Figure 4 In the fourth embodiment of the express delivery method based on member preferences in the embodiments of the present invention, it includes:

[0064] S401. Obtain the required receiving address data, and use the location understanding sub-model to preprocess the required receiving address data according to natural language processing technology to obtain the first address information;

[0065] When the member designates the delivery location, the system obtains the required receiving address data, and then the location understanding model of the system uses natural language processing technology for parsing. The location understanding sub-model first performs preprocessing operations such as word segmentation and part-of-speech tagging on the input text to obtain key vocabulary;

[0066] S402. Use the location understanding sub-model to identify the location type of the first address information according to the semantic understanding algorithm to obtain the second address information;

[0067] Then the location understanding sub-model will use semantic understanding algorithms to identify the location type of the first address information, such as building names, landmarks, orientation descriptions, etc., so as to initially determine the address location;

[0068] S403. Obtain map data, and use the location understanding sub-model to perform fuzzy matching on the second address information in combination with the map data to generate an accurate recommended delivery location.

[0069] To further improve the accuracy of the recommended delivery location, the location understanding sub-model performs fuzzy matching on the second address information in combination with the map data. Even if the location description contains fuzzy terms, under the fuzzy matching of the model in combination with the map data, possible accurate locations can be generated. For example, when a member inputs "the second mailbox on the left at the entrance of the community", the location understanding sub-model identifies "the entrance of the community" as the key location and "the second mailbox on the left" as the relative location description. By combining with the geographical coordinate information of the entrance of the community in the map data, the accurate location coordinates of the mailbox are determined and displayed on the map for the member to confirm. If the member is not satisfied with the parsing result, the system provides functions for manual adjustment or further description until accurate location information is obtained.

[0070] Please refer to Figure 5 , the fifth embodiment of the express delivery method based on member preferences in the embodiments of the present invention includes:

[0071] S501. Obtain the visual identification data of the member, and use the image feature extraction model to identify the features in the visual identification data to obtain the first source information;

[0072] Specifically, the system regularly starts a deep learning-based image recognition algorithm (such as a convolutional neural network) to scan and recognize the visual identification data of the member, and identify the features in the image, such as specific icons, text identifiers, etc., so as to obtain the first source information;

[0073] S502. Obtain the data file related to the member, and extract the features in the data file to obtain the second source information;

[0074] In addition, data files related to the member will also be stored in the system. The naming rules of the data files, keywords in the storage path, etc. can all be used to trace the original source of the member label.

[0075] S503. Use the association analysis model to determine the original data source of the member's label according to the first source information and the second source information;

[0076] The association analysis model can perform association analysis on the image features in the first source information and information such as keywords in the second source information, and then match them with the pre-constructed list of original sources of tags, so as to more accurately analyze and determine the original data sources of the member tags. Members of the system may be referred by multiple different platforms or customer systems. For customers from different sources, different additional services may be generated. By determining the original data sources of the member tags, member services can be better provided.

[0077] S504. The rule model constructed by using the decision tree algorithm adjusts the member tags according to the preset rules and the original data sources to obtain tag information.

[0078] Once the source is determined, the rule model constructed by using the decision tree algorithm operates according to the preset rules. For example, if the source is determined to be "data related to false signature", the rule model modifies the source name of "A tag" to "false signature" according to the rules, re-sets the push source name, and updates the interface document name provided to the source platform. During the whole process, the system details and records log information such as operation steps, time, and operators for subsequent auditing and problem tracing.

[0079] Please refer to Figure 6 , the sixth embodiment of the express delivery method based on member preferences in the embodiments of the present invention includes:

[0080] S601. Obtain a delivery instruction, and use the location generation algorithm to generate a delivery address according to the member's first selection result and the member's second selection result;

[0081] After obtaining the delivery instruction, the express order enters the delivery link. The member's first selection result includes the preferred delivery method of the member, such as door-to-door delivery or collection at the post station. The member's second selection result includes the preferred delivery address of the member. The location generation algorithm forms a specific address by comprehensively considering the two selection results.

[0082] S602. Obtain the current traffic conditions, and use the Dijkstra algorithm to generate an optimal path according to the current traffic conditions and the delivery address, and generate delivery information including the delivery address, the optimal path, and the tag information;

[0083] In the path planning algorithm based on the Geographic Information System (GIS), the Dijkstra algorithm can be used for improvement to calculate the shortest driving path from one location to another. The delivery information sent to the delivery person should not only include the delivery address, the optimal transportation path, and the estimated delivery time, but also often include the tag information of the customer. For example, for certain types of customers or members, a phone call is required for confirmation before delivery. And some members can be directly placed at the post station or other designated collection points.

[0084] S603. Continuously obtain the latest weather condition information and the latest traffic condition information, and adjust the delivery information according to the latest weather condition information and the latest traffic condition information.

[0085] By continuously obtaining the latest weather condition information and the latest traffic condition information, the system can adjust and update the delivery information in real time to ensure that the express parcels can be delivered within the specified time as much as possible. For the predicted possible delayed delivery (such as the expected arrival time exceeding the member's expected time due to traffic congestion) or abnormal situations (such as bad weather may affect the delivery safety) based on the latest weather condition information and the latest traffic condition information, the system timely adjusts the delivery information, such as adjusting the delivery route or modifying the delivery time. At the same time, the system will send the delay message to the member and communicate and negotiate with the member to improve the member's satisfaction.

[0086] Please refer to Figure 7 , the seventh embodiment of the express delivery method based on member preferences in the embodiments of the present invention includes:

[0087] S701. Obtain historical delivery record data, and use the historical delivery record data to train a machine learning algorithm to construct a delivery status prediction model;

[0088] Specifically, machine learning algorithms such as the random forest algorithm can be used to train with these data to construct a delivery status prediction model;

[0089] S702. Obtain the latest weather condition information and the latest traffic condition information, use the delivery status prediction model, predict the delivery information according to the latest weather condition information and the latest traffic condition information, and generate a warning message according to the prediction result;

[0090] During the delivery process, the delivery status prediction model based on big data analysis can analyze and predict the delivery task information (including delivery location, time, etc.) and historical delivery data predicted according to the latest weather condition information and the latest traffic condition information, and form a warning message according to the prediction result. The warning message includes information such as whether there is a delay, the delay duration, or whether to reschedule the delivery, etc.

[0091] S703. Formulate a warning response rule, adjust the delivery information according to the warning message and the warning response rule, and output the adjusted delivery information to the deliveryman.

[0092] In the warning response rule, when the predicted delay duration is greater than the set value, the delivery is rescheduled, and the member is contacted to communicate and negotiate a new delivery time. If the delay is caused by traffic congestion, the delivery route is re-planned and adjusted.

[0093] The above has described the express delivery method based on member preferences in the embodiments of the present invention. Next, the express delivery device based on member preferences in the embodiments of the present invention will be described. Please refer to Figure 8 An embodiment of the express delivery device based on member preferences in the embodiments of the present invention includes:

[0094] A classification module 10, configured to obtain historical purchase data of a member, adopt a preference classification model, generate receiving recommendation information and a classification result for the member according to the historical purchase data, and output the receiving recommendation information to obtain a first selection result of the member;

[0095] An analysis module 20, configured to obtain receiving information specified by a member, adopt a delivery recommendation model, generate recommended delivery information according to the receiving information and the classification result, and output the recommended delivery information to obtain a second selection result of the member;

[0096] A label module 30, configured to obtain visual identification data of a member, adjust the label of the member according to the visual identification data, and obtain label information;

[0097] A delivery module 40, configured to obtain a delivery instruction, and generate delivery information according to the delivery instruction, the first selection result of the member, the second selection result of the member, and the label information.

[0098] Please refer to Figure 9 An embodiment of the express delivery device based on member preferences in the embodiments of the present invention includes:

[0099] A classification module 10, configured to obtain historical purchase data of a member, adopt a preference classification model, generate receiving recommendation information and a classification result for the member according to the historical purchase data, and output the receiving recommendation information to obtain a first selection result of the member;

[0100] An analysis module 20, configured to obtain receiving information specified by a member, adopt a delivery recommendation model, generate recommended delivery information according to the receiving information and the classification result, and output the recommended delivery information to obtain a second selection result of the member;

[0101] A label module 30, configured to obtain visual identification data of a member, adjust the label of the member according to the visual identification data, and obtain label information;

[0102] A delivery module 40, configured to obtain a delivery instruction, and generate delivery information according to the delivery instruction, the first selection result of the member, the second selection result of the member, and the label information;

[0103] In this embodiment, the analysis module 20 includes:

[0104] An acquisition unit 21 for acquiring the receiving information specified by a member, where the receiving information includes required delivery date data and required delivery address data;

[0105] A construction unit 22 for constructing a delivery recommendation model, where the delivery recommendation model includes a time series analysis sub-model and a location understanding sub-model;

[0106] A first generation unit 23 for using the time series analysis sub-model to generate a recommended delivery time according to the required delivery date data and the classification result;

[0107] A second generation unit 24 for using the location understanding sub-model to generate a recommended delivery location according to the required delivery address data;

[0108] An output unit 25 for outputting the recommended delivery time and the recommended delivery location to obtain the second selection result of the member;

[0109] In this embodiment, the first generation unit 23 includes:

[0110] A preprocessing sub-unit 231 for acquiring the required delivery date data and preprocessing the required delivery date data to obtain preprocessed data;

[0111] A labeling sub-unit 232 for acquiring multiple member receiving data and classifying and labeling the member receiving data to construct a member receiving database;

[0112] A generation sub-unit 233 for using the time series analysis sub-model to generate a recommended delivery time according to the preprocessed data, the classification result, and the actual delivery time of the express in the member receiving database;

[0113] In this embodiment, the second generation unit 24 includes:

[0114] An address processing sub-unit 241 for acquiring the required delivery address data and using the location understanding sub-model to preprocess the required delivery address data according to natural language processing technology to obtain first address information;

[0115] An identification sub-unit 242 for using the location understanding sub-model to identify the location type of the first address information according to a semantic understanding algorithm to obtain second address information;

[0116] A matching sub-unit 243 for acquiring map data and using the location understanding sub-model to perform fuzzy matching on the second address information in combination with the map data to generate an accurate recommended delivery location;

[0117] In this embodiment, the label module 30 includes:

[0118] The first extraction unit 31 is configured to obtain the visual identification data of the member, and use an image feature extraction model to identify the features in the visual identification data, so as to obtain the first source information;

[0119] The second extraction unit 32 is configured to obtain the data file related to the member, and extract the features in the data file, so as to obtain the second source information;

[0120] The association analysis unit 33 is configured to use an association analysis model to determine the original data source of the member's label according to the first source information and the second source information;

[0121] The adjustment unit 34 is configured to use a rule model constructed by a decision tree algorithm to adjust the member's label according to the preset rules and the original data source, so as to obtain the label information;

[0122] In this embodiment, the delivery module 40 includes:

[0123] The location generation unit 41 is configured to obtain a delivery instruction, and use a location generation algorithm to generate a delivery address according to the member's first selection result and the member's second selection result;

[0124] The delivery generation unit 42 is configured to obtain the current traffic condition, and use the Dijkstra algorithm to generate an optimal path according to the current traffic condition and the delivery address, and generate delivery information by combining the delivery address, the optimal path, and the label information;

[0125] The scheduling unit 43 is configured to continuously obtain the latest weather condition information and the latest traffic condition information, and adjust the delivery information according to the latest weather condition information and the latest traffic condition information;

[0126] In this embodiment, the scheduling unit 43 includes:

[0127] The construction subunit 431 is configured to obtain historical delivery record data, and use the historical delivery record data to train a machine learning algorithm, so as to construct a delivery status prediction model;

[0128] The prediction subunit 432 is configured to obtain the latest weather condition information and the latest traffic condition information, use the delivery status prediction model to predict the delivery information according to the latest weather condition information and the latest traffic condition information, and generate a warning information according to the prediction result;

[0129] The scheduling subunit 433 is configured to formulate a warning response rule, adjust the delivery information according to the warning information and the warning response rule, and output the adjusted delivery information to the deliveryman.

[0130] The express delivery device based on member preferences provided by the present invention first forms receiving recommendation information according to the historical purchase data of members, and further obtains the first selection result of members based on this; then forms recommended delivery information according to the receiving information specified by members, and further obtains the second selection result of members based on this. Then, it traces the original label source according to the relevant original data of members and makes adjustments. Finally, by combining the delivery instruction, the first selection result of members, the second selection result of members, and the label information to generate the delivery information, the obtained delivery information conforms to the preferences of members to the greatest extent, and within the delivery capacity range of the express delivery enterprise, it can better improve the satisfaction of members with the services of the express delivery company.

[0131] The above is a detailed description of the express delivery device based on member preferences in the embodiments of the present invention from the perspective of modular functional entities. The following is a detailed description of the express delivery device based on member preferences in the embodiments of the present invention from the perspective of hardware processing.

[0132] Figure 10 FIG. is a schematic structural diagram of an express delivery device based on member preferences provided by an embodiment of the present invention. The express delivery device 900 based on member preferences may vary greatly due to configuration or performance differences, and may include one or more processors (central processing units, CPU) 910 (for example, one or more processors) and a memory 920, and one or more storage media 930 for storing application programs 933 or data 932 (for example, one or more mass storage devices). Among them, the memory 920 and the storage media 930 may be transient storage or persistent storage. The program stored in the storage media 930 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations on the express delivery device 900 based on member preferences. Further, the processor 910 may be configured to communicate with the storage media 930 and execute a series of instruction operations in the storage media 930 on the express delivery device 900 to implement the steps of the express delivery method based on member preferences provided by the above method embodiments.

[0133] The express delivery device 900 based on member preferences may further include one or more power supplies 940, one or more wired or wireless network interfaces 950, one or more input / output interfaces 960, and / or one or more operating systems 931, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, and so on. Those skilled in the art can understand that Figure 10 The shown structural diagram of the express delivery device based on member preferences does not constitute a limitation on the express delivery device based on member preferences, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0134] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are run on a computer, the computer is caused to execute the steps of the express delivery method based on member preferences.

[0135] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described devices or apparatuses can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0136] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The foregoing storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc that can store program codes.

[0137] It can be understood that for those of ordinary skill in the art, equivalent substitutions or changes can be made according to the technical solution of the present invention and its inventive concept, and all such changes or substitutions should fall within the protection scope of the claims appended to the present invention.

Claims

1. A method for express delivery based on member preferences, characterized in that: The steps include: Obtain the member's historical purchase data, adopt a preference classification model, generate the member's delivery recommendation information and classification results based on the historical purchase data, and output the delivery recommendation information to obtain the member's first choice result; Obtaining the delivery information specified by the member, using the delivery recommendation model to generate recommended delivery information according to the delivery information and the classification result, and outputting the recommended delivery information to obtain the member's second choice result; Obtaining visual identification data of the member, adjusting the member's label according to the visual identification data, and obtaining label information; A delivery instruction is obtained, and delivery information is generated according to the delivery instruction, the member's first selection result, the member's second selection result, and the tag information.

2. The express delivery method based on member preferences according to claim 1, characterized in that: The obtaining of the delivery information specified by the member, using the delivery recommendation model, generating recommended delivery information according to the delivery information and the classification result, and outputting the recommended delivery information to obtain the member's second selection result specifically includes: Obtaining the delivery information specified by the member, the delivery information including the required delivery date data and the required delivery address data; Constructing a delivery recommendation model, wherein the delivery recommendation model includes a time series analysis sub-model and a location understanding sub-model; A time series analysis sub-model is used to generate recommended delivery times based on the required delivery date data and classification results; Use the location understanding sub-model to generate recommended delivery locations based on the required delivery address data; Output the recommended delivery time and recommended delivery location to obtain the member's second choice result.

3. The express delivery method based on member preferences according to claim 2, characterized in that: The time series analysis sub-model is used to generate the recommended delivery time according to the required delivery date data and classification results, specifically including: Obtaining required delivery date data, and preprocessing the required delivery date data to obtain preprocessed data; Acquire multiple member receipt data, and classify and annotate the member receipt data to build a member receipt database; The time series analysis sub-model is used to generate the recommended delivery time based on the preprocessed data, classification results and the actual delivery time of the express in the member receipt database.

4. The express delivery method based on member preferences according to claim 2, characterized in that: The location understanding sub-model is used to generate a recommended delivery location based on the required delivery address data, specifically including: Obtaining required delivery address data, using the location understanding sub-model, and pre-processing the required delivery address data according to natural language processing technology to obtain first address information; Using the location understanding sub-model, the location type of the first address information is identified according to the semantic understanding algorithm to obtain the second address information; Obtain map data, use the location understanding sub-model, and perform fuzzy matching on the second address information in combination with the map data to generate an accurate recommended delivery location.

5. The express delivery method based on member preferences according to claim 1, characterized in that: The obtaining of the member's visual identification data, adjusting the member's label according to the visual identification data, and obtaining label information specifically includes: Obtaining visual identification data of the member, and using an image feature extraction model to identify features in the visual identification data to obtain first source information; Obtain data files related to members and extract features in the data files to obtain second source information; Adopting the association analysis model, the original data source of the member's tag is determined based on the first source information and the second source information; The rule model constructed using the decision tree algorithm adjusts members’ labels according to preset rules and original data sources to obtain label information.

6. The express delivery method based on member preferences according to claim 1, characterized in that: The obtaining of the delivery instruction and generating the delivery information according to the delivery instruction, the member's first selection result, the member's second selection result and the tag information specifically include: Obtain delivery instructions, use the location generation algorithm, and generate a delivery location based on the member's first choice result and the member's second choice result; Obtain the current traffic conditions, use the Dijkstra algorithm to generate the optimal path based on the current traffic conditions and the delivery location, and generate delivery information based on the delivery location, optimal path, and label information; Continuously obtain the latest weather information and the latest traffic information, and adjust the delivery information based on the latest weather information and the latest traffic information.

7. The express delivery method based on member preferences according to claim 6, characterized in that: The continuously obtaining the latest weather information and the latest traffic information, and adjusting the delivery information according to the latest weather information and the latest traffic information, specifically includes: Obtain historical delivery record data and use it to train machine learning algorithms to build a delivery status prediction model; Obtain the latest weather information and the latest traffic information, use the delivery status prediction model to predict the delivery information based on the latest weather information and the latest traffic information, and generate warning information based on the prediction results; Formulate early warning response rules, adjust the delivery information according to the early warning information and the early warning response rules, and output the adjusted delivery information to the delivery person.

8. A courier delivery device based on member preferences, characterized in that: include: A classification module is used to obtain the member's historical purchase data, adopt a preference classification model, generate the member's delivery recommendation information and classification results based on the historical purchase data, and output the delivery recommendation information to obtain the member's first choice result; An analysis module is used to obtain the delivery information specified by the member, use the delivery recommendation model to generate recommended delivery information according to the delivery information and the classification result, and output the recommended delivery information to obtain the member's second choice result; The label module is used to obtain the member's visual identification data, adjust the member's label according to the visual identification data, and obtain label information; The delivery module is used to obtain the delivery instruction and generate the delivery information according to the delivery instruction, the member's first selection result, the member's second selection result and the label information.

9. A courier delivery device based on member preferences, characterized in that: comprising a memory and at least one processor, wherein the memory has computer-readable instructions stored therein; The at least one processor calls the computer-readable instructions in the memory to execute the various steps of the express delivery method based on member preferences as described in any one of claims 1-7.

10. A computer-readable storage medium having computer-readable instructions stored thereon, characterized in that: When the computer-readable instructions are executed by a processor, the steps of the express delivery method based on member preferences as described in any one of claims 1 to 7 are implemented.

Images