Distribution range determination method and device

Abstract

The invention discloses a distribution range determination method and device, relates to the field of data processing, and is used for solving the problems of high cost, long processing period and low efficiency of searching an expandable area of a front warehouse distribution service in a manual mode at present. The method comprises the steps of obtaining feedback information of a user, if the feedback information indicates that a first address does not have a delivery service, determining a storage bin closest to the first address and the distance between the storage bin and the first address, and if the distance is smaller than a preset threshold value, bringing the first address into a delivery range of the storage bin. Through the above scheme, the distribution range of the storage warehouse, especially the distribution range of the front warehouse, can be automatically and instantly adjusted so as to ensure the distribution efficiency and the service quality.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular, to a method and device for determining a delivery range. Background Art

[0002] The front - warehouse model is a fresh - food retail delivery model. In this model, front - warehouses are densely deployed in populated areas to provide instant delivery services to users within a certain area range (such as a five - kilometer range) nearby. Compared with the traditional e - commerce model that provides services to users with the entire county as the smallest unit, the front - warehouse model can provide fast delivery services in densely populated areas, is closer to users, has stronger timeliness, and can ensure the freshness of goods when they are delivered to users.

[0003] However, due to the limitation of the front - warehouse deployment cost, the retail coverage of the front - warehouse model is relatively limited and usually can only serve densely populated areas. But with the construction of new communities and the flow of urban population, fresh - food e - commerce needs to continuously adjust the delivery range to adapt to the new population distribution. Currently, the adjustment method usually involves manually finding areas where the front - warehouse delivery service can be expanded, unable to adjust the storage warehouse immediately, and unable to ensure delivery efficiency and service quality.

[0004] Therefore, how to immediately adjust the delivery range of a storage warehouse (such as a front - warehouse) to ensure delivery efficiency and service quality is an urgent problem to be solved currently. Summary of the Invention

[0005] The present invention provides a method and device for determining a delivery range, which is used to immediately adjust the delivery range of a storage warehouse to ensure delivery efficiency and service quality.

[0006] In a first aspect, the present invention provides a method for determining a delivery range. The method includes: obtaining feedback information of a user, where the feedback information is used to indicate that a first address does not have a delivery service, determining the storage warehouse closest to the first address and the distance from the storage warehouse to the first address, and if the distance is less than a preset threshold, including the first address in the delivery range of the storage warehouse.

[0007] Through the above solution, the first address not within the delivery range can be determined from the user's feedback information. By judging the distance between the first address and the closest storage warehouse, it is determined whether the first address can be included in the delivery range. Compared with the prior art where opportunities for expanding the delivery area are manually searched, the present application can automatically obtain the first address that needs to expand the area and automatically judge the feasibility of including the first address in the delivery range, and can automatically obtain opportunities for expanding the area.

[0008] Optionally, the feedback information is sourced from customer service feedback on the e-commerce platform and / or information feedback on the media platform. Through the above solution, feedback information on users' requirements for the delivery service can be obtained from multiple platforms. Compared with the prior art that lacks reliable data at the community level, this application can automatically mine user feedback from customer service and comments on social platforms to obtain opportunities for expanding the delivery area.

[0009] Optionally, the feedback information is sourced from the login information of users on the application (APP). Before obtaining the feedback information of users, it further includes: determining that the login address included in the login information does not have the delivery service. In this way, the user addresses outside the delivery service range can be obtained through the APP login information, and thus potential opportunities for expanding the delivery service range can be obtained.

[0010] Optionally, determining the storage warehouse closest to the first address includes: obtaining the community address to which the login address belongs and determining the storage warehouse closest to the community address. In this way, the login address can be associated with the community information. Since the amount of user login data is very large and there are some invalid data (such as login data in rivers, lakes or parks), it is necessary to associate the user login data with the known community data.

[0011] Optionally, before determining the storage warehouse closest to the first address, it further includes: using a first large language model to classify the feedback information to obtain a classification result, and determining that the classification result is the first type of feedback information, where the first type of feedback information is the feedback information that the user expects to have the delivery service at the first address. In this way, the large language model can be used to quickly identify comments related to expanding the delivery area (i.e., feedback information) on each platform, improve the data quality, and reduce the cost of manual classification.

[0012] Optionally, before determining the storage warehouse closest to the first address, it further includes: using a second large language model to extract the first address in the feedback information, where the number of parameters of the second large language model is more than that of the first large language model. In this way, by using a large language model with a larger number of parameters, the first address in the feedback information can be quickly inferred and mined.

[0013] Optionally, after obtaining the user's feedback information, it further includes: extracting the first address in the feedback information, encoding the first address if the first address belongs to a complete address, and if the first address belongs to an incomplete address, obtaining information on one or more service cities to which the first address belongs, and jointly encoding the information on one or more service cities and the first address. In this way, the first address is associated with the service city, and then encoded to obtain the first address-encoding information (such as location coordinates) data pair most relevant to the service scope, which can ensure that no potential expansion opportunities are missed, while ensuring the integrity and accuracy of the address information and filtering out redundant information.

[0014] Optionally, determine the distance from the storage warehouse to the first address. If the distance is less than a preset threshold, include the first address in the delivery scope of the storage warehouse, including: determining the first distance from the storage warehouse to the first address, and the second distance from the farthest delivery location of the storage warehouse to the first address. If the first distance is less than the first threshold, and / or the second distance is less than the second threshold, include the first address in the delivery scope of the storage warehouse. In this way, by calculating the distance between the existing storage warehouse and the first address, the feasibility of including the first address in the delivery service scope of the storage warehouse can be evaluated, providing a strong basis for decision-making.

[0015] In a second aspect, the present invention provides a delivery scope determination device, which includes:

[0016] An acquisition module for acquiring the user's feedback information, where the feedback information is used to indicate that the first address does not have a delivery service;

[0017] A determination module for determining the storage warehouse closest to the first address and the distance from the storage warehouse to the first address. If the distance is less than a preset threshold, include the first address in the delivery scope of the storage warehouse.

[0018] In a possible implementation manner, the feedback information is derived from the customer service feedback of an e-commerce platform and / or the information feedback of a media platform.

[0019] In a possible implementation manner, the feedback information is derived from the login information of the user on the application program APP, and the acquisition module is further used to: determine that the login address included in the login information does not have a delivery service.

[0020] In a possible implementation manner, it further includes a classification module for classifying the feedback information using a first large language model to obtain a classification result, and determining that the classification result is a first type of feedback information, where the first type of feedback information is the feedback information that the user expects to have a delivery service at the first address.

[0021] In a possible implementation manner, the determination module is specifically used to: obtain the community address to which the login address belongs, and determine the storage warehouse closest to the community address.

[0022] In a possible implementation, it further includes an extraction module for extracting the first address in the feedback information using a second large language model, where the number of parameters of the second large language model is more than that of the first large language model.

[0023] In a possible implementation, the extraction module is further configured to: extract the first address in the feedback information, if the first address belongs to a complete address, encode the first address, if the first address belongs to an incomplete address, obtain information of one or more service cities to which the first address belongs, and jointly encode the information of the one or more service cities and the first address.

[0024] In a possible implementation, the determination module is specifically configured to: determine a first distance from the storage bin to the first address, and a second distance from the farthest delivery location of the storage bin to the first address, if the first distance is less than a first threshold, and / or, the second distance is less than a second threshold, then include the first address in the delivery range of the storage bin.

[0025] In a third aspect, the present invention further provides a delivery range determination device, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the methods described in various possible designs of the first aspect are implemented.

[0026] In a fourth aspect, the present invention further provides a computer-readable storage medium, on which a computer program or instruction is stored. When the computer program or instruction is executed by a processor, the methods described in various possible designs of the first aspect are implemented.

[0027] In a fifth aspect, the present invention further provides a computer program product, which when running on a computer causes the computer to execute the method described in any one of the above first aspects.

[0028] These implementation manners or other implementation manners of the present application will be more clearly understood in the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0030] Figure 1 It is a flowchart of a delivery range determination method provided by an embodiment of the present invention;

[0031] Figure 2A schematic flow chart of using a large language model to extract the first address in feedback information provided by an embodiment of the present invention;

[0032] Figure 3 A schematic diagram of the distance from the first address to the storage warehouse provided by an embodiment of the present invention;

[0033] Figure 4 A schematic flow chart of another method for determining the delivery range provided by an embodiment of the present invention;

[0034] Figure 5 A schematic diagram of a device for determining the delivery range provided by an embodiment of the present invention;

[0035] Figure 6 A schematic diagram of another device for determining the delivery range provided by an embodiment of the present invention. Detailed implementation manners

[0036] In order to make the objectives, technical solutions and beneficial effects of the present application clearer and more understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0037] As described in the background art, the existing storage warehouse adjustment method cannot achieve instant adjustment. This is mainly because the existing population living conditions lack reliable real-time data at the community level, and only manual methods can be used to find areas where the front-end warehouse delivery service can be expanded.

[0038] The manual methods mainly include on-site inspections offline and screening of online user feedback. However, no matter which one, they all face a large amount of investigation and data cleaning work, with a long processing cycle and low efficiency. Therefore, how to instantaneously adjust the delivery range of the storage warehouse (especially the front-end warehouse) to ensure the delivery efficiency and service quality is an urgent problem to be solved in the current field of storage warehouse adjustment.

[0039] In view of this, the present invention provides a method for determining the delivery range. This method automatically searches for the nearest storage warehouse according to the user's feedback signal, and when the distance between the location of the storage warehouse and the address feedback by the user is relatively close, automatically expands the address feedback by the user to the delivery range of the storage warehouse, so as to instantaneously adjust the delivery range of the storage warehouse and ensure the delivery efficiency and service quality.

[0040] The following will introduce the solution of the present application in detail with reference to the accompanying drawings.

[0041] Please refer to Figure 1 , which shows a schematic flow chart of a method for determining the delivery range. This method is applicable to computing devices, such as the server of a delivery platform, simply referred to as a delivery server.

[0042] AsFigure 1 As shown in the figure, the method may specifically include the following steps:

[0043] Step 110: Obtain the feedback information of the user.

[0044] Here, the feedback information is used to indicate that the first address does not have a delivery service.

[0045] The feedback information may be directly or indirectly fed back to the delivery server by the user when determining that a certain address does not have a delivery service. For example, if the user currently lives in Yulan Xiangyuan but finds that there is no delivery service in Yulan Xiangyuan, in this case, the user can send feedback information to the delivery server or the corresponding customer service. As an example, the feedback information may be: "I am an old customer of your company. I just moved to Yulan Xiangyuan. Why is there no delivery service here?"

[0046] Optionally, the feedback information may come from multiple channels, including but not limited to the following three channels:

[0047] Channel 1: Feedback from the customer service of the e-commerce platform. For example, when the user finds that the first address does not have a delivery service, the user can feedback to the customer service on the e-commerce platform and inform the customer service to enable the delivery service for the first address. That is to say, it is possible to automatically mine the user feedback received on the customer service side as an opportunity point for expanding the delivery area;

[0048] Channel 2: Information feedback from the media platform. For example, when the user finds that the first address does not have a delivery service, the user can give feedback on the social media platform. For example, apply for enabling the delivery service for the first address on the social platform, or make comments on the social platform. That is to say, it is possible to automatically obtain and mine the comments on the social platform to obtain the opportunity points for expanding the delivery area from the feedback of customers on the social platform.

[0049] Channel 3: The login information of the user on the application APP. For example, the delivery server can automatically obtain the login information of each user in the delivery APP. For each piece of login information, extract the login address included in the piece of login information. If it is determined that the login address does not have a delivery service, then this piece of login information can be used as a piece of feedback information.

[0050] For example, when the user logs in to the delivery APP outside the delivery service area of the front warehouse, the delivery APP automatically pushes the user's login information to the delivery server. The delivery server extracts the login address in this piece of login information and finds that the login address is outside the delivery service range, indicating that the user is currently not within the delivery service range. In this case, the current login address of the user can be used as the first address. In this way, it is possible to provide the opportunity of delivery service for new addresses to old users using the delivery APP.

[0051] Through the above solution, feedback information on users' requirements for the delivery service can be obtained from multiple platforms. Compared with the existing solution where reliable data at the community level is lacking and the delivery scope cannot be immediately expanded, this acquisition method can obtain multi-dimensional reliable data to support the immediate expansion of the delivery scope.

[0052] Optionally, if the feedback information comes from Channel 3 above, after the delivery server extracts the login address included in the login information, it also needs to determine whether the login address is a valid address. If it is an invalid address, such as the login address being in a river, lake, or park, the delivery server can first obtain the community address to which the login address belongs, and then determine whether the community address has a delivery service. If not, the community address is used as the first address. In this way, the invalid address in the login information can be converted into community information, that is, after filtering some invalid data in the user's login information (such as login data in a river, lake, or park), the association of the delivery scope is performed.

[0053] Further, optionally, if the feedback information comes from Channel 3 above, the login information can also be associated with the community address. Specifically, the communities can be classified in advance, one type being fenced communities and the other type being unfenced communities. Based on this classification, after obtaining the community address, if the community is a fenced community, it is determined whether the login address is within the community. If the login address is within the community, the community address can be determined as the community address to which the login address belongs. For an unfenced community, it is necessary to determine whether the login address is within a certain radius of the community location. If the login address is within a certain radius of the community, the community address can be determined as the community address to which the login address belongs.

[0054] Further, optionally, the login information that cannot obtain the community is discarded. For example, the login information can be associated with the community data and clustered, and then the irrelevant login data is filtered out to obtain the login information that is most valuable for the fresh food delivery service. By filtering out the login data that cannot be associated with a specific community, a large number of locations with little value for fresh food consumption, such as lakes, rivers, parks, and factories, can be cleaned, significantly reducing the data scale and greatly reducing the number of login points that need to be checked, ensuring the quality and relevance of the data.

[0055] In the above content, the community address comes from the community data, and the community data can be obtained by interacting with third-party devices such as the real estate trading platform and the map service provider.

[0056] It is understandable that there is a lot of noise in the feedback information obtained through the above solutions. For example, there are many types of information that users feedback through customer service on e-commerce platforms and on social media platforms. Among these information, the information indicating that the first address does not have delivery services only accounts for a very small proportion. Moreover, there is a lot of information without valid address information in the information feedback through customer service on e-commerce platforms and on social media platforms, so the first address cannot be provided. In addition, the user login data volume is very large and there are a large number of invalid login addresses. For example, some login addresses come from web crawlers, etc.

[0057] Based on this, in one example, after obtaining the user's feedback information, the noise in the feedback information can also be removed to improve the quality of the information. It should be noted that this cleaning operation can be performed before determining the community address. For example, it can be immediately performed after obtaining the user's delivery information to reduce the amount of data to be processed subsequently.

[0058] Furthermore, optionally, the feedback information can also be preprocessed. For example, the data after removing noise is preprocessed, including but not limited to: removing special characters, converting case, correcting typos, etc.

[0059] Optionally, a large language model can be used to extract the first address in the feedback information. Specifically, please refer to Figure 2 which shows a schematic flow diagram of using a large language model to extract the first address in the feedback information. Step 110 specifically includes the following steps:

[0060] Step 210: Use the first large language model to classify the feedback information to obtain a classification result, and confirm that the classification result is the first type of feedback information.

[0061] Here, the first type of feedback information is the feedback information that the user expects to open delivery services at the first address. A relatively small model can be used for fine-tuning to classify the feedback information. For example, a small model below 7B is used. In this way, a large number of irrelevant comments can be quickly filtered out.

[0062] Among them, the classification result can include the intention of expanding the area and the intention of not expanding the area. That is, the feedback information is divided into information with the intention of expanding the area and information without the intention of expanding the area. Among them, the information with the intention of expanding the area means that the user expects to open delivery services at the first address, and the information without the intention of expanding the area is other information except the information with the intention of expanding the area. By using the first large language model to classify all the obtained information, the feedback information with the intention of expanding the area can be obtained. In this way, the large language model can be used to quickly identify the comments (i.e., feedback information) related to expanding the area on each platform, improve the data quality, and reduce the cost of manual classification.

[0063] Optionally, before using the first large language model to classify the feedback information, the first large language model can be fine-tuned first. Fine-tuning of a large model refers to the process of performing additional training on a pre-trained large language model to make it better adapt to specific tasks. For example, by annotating a set number (such as about 100) of datasets to fine-tune the open-source base model, the model can identify customer service information related to expansion areas and feedback information on social platforms. In this way, the relatively small fine-tuned large language model can quickly identify the feedback information related to expansion areas in customer service information and comment information on social platforms.

[0064] Step 220: Use the second large language model to extract the first address in the feedback information.

[0065] Here, the number of parameters of the second large language model is more than that of the first large language model. For example, a model with more than 14B parameters can be selected. Since this step is more complex than step 210, models with more than 14B parameters perform significantly better than models with 7B or less. By using a large language model with a larger number of parameters, the first address in the feedback information can be quickly inferred and mined. It can be understood that the number of parameters of the second large language model should not be too large, as it will cause latency problems.

[0066] Optionally, the second large language model can be used to extract the address in the feedback information and determine whether the address in the feedback information is a valid address. If the feedback information contains valid address information, the second large language model outputs the valid first address. For example, "Yulan Xiangyuan" in the above feedback information example is a valid address. If there is no address information in the feedback information, or the address information is invalid (such as the address information range is too large), a null value is output. For example, if the feedback information is: "Why is there no delivery in Xishi", the second large language model identifies the address information as "Xi'an City". Since the address information range is too large and it is invalid address information, a null value is output.

[0067] It can be understood that for feedback information from customer service feedback on e-commerce platforms and / or information feedback on media platforms, the second large language model can be used to extract the first address in the feedback information; for feedback information from the login information of users on the application APP, the community address corresponding to the login information can be directly used as the first address.

[0068] Optionally, a high-performance large language model inference framework (versatile large language model, vLLM) can be selected as the second large language model to extract the first address in the feedback information.

[0069] Further, optionally, named entity recognition (NER) based on a large language model can be used to extract the first address in the feedback information.

[0070] Optionally, before using the second large language model to extract the first address, the large language model can be trained using a set dataset. For example, a set number of feedback messages can be selected for annotation (such as about 600 feedback messages), that is, the first address in each feedback message is annotated. If there is no valid address in the feedback message, it is annotated as "none". In this way, by annotating a set number of feedback messages and using the annotated feedback messages to train the large model, it can be ensured that the large model can accurately identify the address information in the feedback message.

[0071] Further, optionally, after fine-tuning the second large language model, activation-aware weight quantization (AWQ) needs to be performed on the second large language model, and then key-value cache (KV Cache) operation is performed on the second large language model to improve the running efficiency of the model (such as increasing the model inference time by ten times). Finally, the PagedAttention mechanism is used to manage the video memory fragmentation of the second large language model to reduce the video memory occupancy of the second large language model and ensure that there is no out of memory (OOM) situation.

[0072] Through the above solutions, the input and output, preprocessing, fine-tuning dataset, and inference model deployment of the first large language model and the second large language model are designed respectively, and the first address in the feedback information can be accurately extracted.

[0073] Step 230: Encode the first address.

[0074] Here, for the extracted first address, it needs to be encoded to facilitate finding the corresponding forward warehouse according to the encoded first address later.

[0075] Optionally, the first address can be encoded by a map service provider. Specifically, the first address is converted into longitude and latitude coordinates through the address encoding provided by the map service provider. In this way, the coordinates corresponding to the first address mined from user comments and login information are associated with the coordinates of the already deployed forward warehouses through the longitude and latitude coordinates, which is beneficial for obtaining the nearest forward warehouse later.

[0076] Optionally, if the first address belongs to a complete address, encode the first address. If the first address belongs to an incomplete address, obtain information about one or more service cities to which the first address belongs, and jointly encode the information about the one or more service cities and the first address. For example, the feedback message is: "I am an old customer of your company. I just moved to Yulan Xiangyuan. Why is there no delivery service here?" The first address in this feedback message is "Yulan Xiangyuan". However, there may be many communities with the name "Yulan Xiangyuan" in many cities. For example, there are communities named "Yulan Xiangyuan" in both Beijing and Shanghai, that is, the first address "Yulan Xiangyuan" is an incomplete address. At this time, information about all service cities of the front-end warehouse can be added to the address text of "Yulan Xiangyuan", and then address encoding is performed on each complete address, that is, encode "Yulan Xiangyuan, Beijing" and "Yulan Xiangyuan, Shanghai" respectively. In this way, the first address is associated with the service city, and then encoding is performed to obtain the first address-encoding information (such as location coordinates) data pair that is most relevant to the service scope, which can ensure that no potential expansion opportunities are missed, while ensuring the integrity and accuracy of the address information and filtering out redundant information.

[0077] Step 120: Determine the storage warehouse closest to the first address and the distance from the storage warehouse to the first address.

[0078] Here, the first address can be compared with the addresses of the already deployed storage warehouses (such as front-end warehouses). For example, compare the longitude and latitude coordinates of the first address with the longitude and latitude coordinates of the already deployed storage warehouses to find the storage warehouse (such as front-end warehouse) closest to the first address, and then determine the distance from the storage warehouse (such as front-end warehouse) to the first address.

[0079] Step 130: Determine whether the distance from the storage warehouse to the first address is less than a preset threshold. If so, execute Step 140; if not, execute Step 150.

[0080] Here, if the distance between the first address and the storage warehouse (such as front-end warehouse) is less than the preset threshold, the first address can be included in the delivery service scope of the storage warehouse (such as front-end warehouse).

[0081] Optionally, the first distance from the storage warehouse to the first address and the second distance from the farthest delivery location of the storage warehouse to the first address can be determined. If the first distance is less than the first threshold and / or the second distance is less than the second threshold, the first address is included in the delivery scope of the storage warehouse.

[0082] It can be understood that the first address can be regarded as a point. If the range of the first address is large, the distance from the boundary of the range of the first address to the storage warehouse can be calculated, or the distance from the center point of the first address to the storage warehouse can be calculated, without specific restrictions.

[0083] Specifically, please refer to Figure 3 , which shows a schematic diagram of the distance from the first address to the storage bin. Figure 3 (A) shows the first distance d from the first address to the storage bin 1 , Figure 3 (B) shows the second distance d from the farthest delivery position of the storage bin at the first address to the first address 2 , that is, the distance from the first address to the boundary of the delivery range of the village storage bin. If d 1 is less than the first threshold, and / or d 2 is less than the second threshold, then the first address is included in the delivery range of the storage bin. In this way, by calculating the distance between the existing storage bin and the first address, the feasibility of including the first address in the delivery service range of the storage bin can be evaluated, providing a strong basis for decision-making.

[0084] Step 140: Include the first address in the delivery range of the storage bin.

[0085] Here, the distance from the storage bin to the first address is less than the preset threshold, indicating that the first address is very close to the storage bin, and it can be considered to include the first address in the delivery range of the storage bin.

[0086] Step 150: Ignore the feedback information.

[0087] Optionally, after step 140 and step 150, the location information of the first address and the corresponding storage bin (such as a front-end warehouse) can be displayed at the front end for the operation staff to decide whether to expand the delivery range to the first address.

[0088] The method for determining the delivery range is introduced above. Next, a complete example is used to illustrate the above method.

[0089] Please refer to Figure 4 , which shows a flow schematic diagram of another method for determining the delivery range, specifically including the following steps:

[0090] Step 410: Obtain the feedback information of the user.

[0091] Here, the feedback information of the user can be obtained from the customer service feedback of the e-commerce platform and / or the information feedback of the media platform.

[0092] Step 420: Extract the first address in the feedback information.

[0093] Step 430: Encode the first address.

[0094] Here, the first address in the feedback information can be associated with the city information where the delivery service has been launched, and at the same time, the first address and the city information are encoded.

[0095] Step 440: Obtain the user's login address, and use the login address as the first address and encode it.

[0096] Here, the login address of the user on the application APP can be obtained, and the login address is not within the scope of the front warehouse service.

[0097] It can be understood that steps 410 - 430 and step 440 can be carried out in parallel.

[0098] Step 450: Associate the encoded first address with the front warehouse information.

[0099] Step 460: Display the associated information and decide whether to include the first address in the delivery scope of the storage warehouse.

[0100] It can be understood that steps 410 - 460 can refer to the above steps 110 - 130, which will not be elaborated here.

[0101] Through the above solution, the first address not within the delivery scope can be determined from the user's feedback information. By judging the distance between the first address and the nearest storage warehouse, it is determined whether the first address can be included in the delivery scope. Compared with the prior art in which manual search for the opportunity points for expanding the delivery service, this application can automatically obtain the first address that needs to expand the area and automatically judge the feasibility of including the first address in the delivery scope, and can automatically obtain the opportunity points for expanding the area.

[0102] Based on the same concept, the embodiment of this application also provides a delivery scope determination device, and this delivery scope determination device can execute the delivery scope determination method introduced above.

[0103] Please refer to Figure 5 , and a structural schematic diagram of a delivery scope determination device provided by the embodiment of this application is given. As Figure 5 shown, this delivery scope determination device includes:

[0104] An acquisition module 501, configured to acquire the user's feedback information, where the feedback information is used to indicate that the first address does not have a delivery service;

[0105] A determination module 502, configured to determine the storage warehouse closest to the first address and the distance from the storage warehouse to the first address. If the distance is less than a preset threshold, the first address is included in the delivery scope of the storage warehouse.

[0106] In a possible implementation manner, the feedback information is from the customer service feedback of the e - commerce platform and / or the information feedback of the media platform.

[0107] In a possible implementation, the feedback information is derived from the user's login information on the application APP, and the obtaining module 501 is further configured to: determine that the login address included in the login information does not have a delivery service.

[0108] In a possible implementation, it further includes a classification module 503, which is used to classify the feedback information using a first large language model to obtain a classification result, and determine that the classification result is a first type of feedback information, where the first type of feedback information is feedback information that the user expects to enable a delivery service at a first address.

[0109] In a possible implementation, the determining module 502 is specifically configured to: obtain the community address to which the login address belongs, and determine the storage warehouse closest to the community address.

[0110] In a possible implementation, it further includes an extraction module 504, which is used to extract the first address in the feedback information using a second large language model, and the number of parameters of the second large language model is more than that of the first large language model.

[0111] In a possible implementation, the extraction module 504 is further configured to: extract the first address in the feedback information. If the first address belongs to a complete address, encode the first address. If the first address belongs to an incomplete address, obtain the information of one or more service cities to which the first address belongs, and jointly encode the information of one or more service cities and the first address.

[0112] In a possible implementation, the determining module 502 is specifically configured to: determine a first distance from the storage warehouse to the first address, and a second distance from the farthest delivery location of the storage warehouse to the first address. If the first distance is less than a first threshold, and / or the second distance is less than a second threshold, then include the first address in the delivery range of the storage warehouse.

[0113] Please refer to Figure 6 , which shows a schematic structural diagram of another delivery range determination device 600 provided by an embodiment of the present application. As Figure 6 shown, the delivery range determination device includes: a memory 601 and a processor 602, and the processor 602 is coupled to the memory 601. The memory 601 is used to store program instructions, and the processor 602 is used to call the program instructions stored in the memory 601 and execute the above-mentioned delivery range determination method according to the obtained program.

[0114] Optionally, the delivery scope determination device may further include an interface circuit 603, which may be a transceiver or an input / output interface. The input / output interface is used for inputting and / or outputting information, where output can be understood as sending, and input can be understood as receiving. The processor 602 may communicate with other components in the delivery scope determination device or other devices outside the delivery scope determination device through the interface circuit 603 to obtain the information required for performing the above delivery scope determination method.

[0115] When the delivery scope determination device 600 is used to implement Figure 1 the method shown, the processor 602 is used to implement the functions of the above-mentioned acquisition module 501, determination module 502, classification module 503, and extraction module 504.

[0116] It can be understood that the processor in the embodiments of the present application may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. The general-purpose processor may be a microprocessor or any conventional processor.

[0117] The memory in the embodiments of the present application may be a random access memory, flash memory, read-only memory, programmable read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, register, hard disk, removable hard disk, compact disc read-only memory (CD-ROM), or any other form of storage medium well-known in the art. An exemplary storage medium is coupled to the processor, enabling the processor to read information from and write information to the storage medium. The storage medium may also be a component of the processor.

[0118] Based on the same technical concept, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program or instruction is stored. When the computer program or instruction is executed by a processor, the computer is caused to execute the above delivery scope determination method.

[0119] Based on the same technical concept, an embodiment of the present invention further provides a computer-readable program product, which when executed, causes a computer to execute the above delivery scope determination method.

[0120] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0121] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0122] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0123] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0124] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these modifications and variations.

Claims

1. A method for determining a delivery range, characterized in that: include: Obtaining feedback information from a user, where the feedback information is used to indicate that the first address does not have a delivery service; Determine the storage bin closest to the first address and the distance from the storage bin to the first address; If the distance is less than a preset threshold, the first address is included in the delivery range of the storage warehouse.

2. The method according to claim 1, characterized in that The feedback information comes from customer service feedback from the e-commerce platform and / or information feedback from the media platform.

3. The method according to claim 1, characterized in that The feedback information comes from the user's login information on the application APP; Before obtaining the user's feedback information, the method further includes: It is determined that the registration address included in the registration information does not have a delivery service.

4. The method according to claim 3, characterized in that The determining of the storage bin closest to the first address includes: Obtaining the cell address to which the login address belongs; The storage bin closest to the cell address is determined.

5. The method according to claim 1, characterized in that Before determining the storage bin closest to the first address, the method further includes: Using the first language model to classify the feedback information, to obtain a classification result; The classification result is determined to be first type feedback information, where the first type feedback information is feedback information that the user expects to activate a delivery service at the first address.

6. The method according to claim 5, characterized in that Before determining the storage bin closest to the first address, the method further includes: The first address in the feedback information is extracted using a second largest language model, wherein the second largest language model has more parameters than the first largest language model.

7. The method according to claim 1, characterized in that After obtaining the user's feedback information, the method further includes: Extracting the first address in the feedback information; If the first address is a complete address, encoding the first address; If the first address is an incomplete address, information of one or more service cities to which the first address belongs is obtained, and the information of the one or more service cities and the first address are jointly encoded.

8. The method according to any one of claims 1 to 7, characterized in that: The determining the distance from the storage warehouse to the first address, and if the distance is less than a preset threshold, including the first address in the delivery range of the storage warehouse, includes: Determining a first distance from the storage bin to the first address, and a second distance from the farthest delivery location of the storage bin to the first address; If the first distance is smaller than a first threshold value, and / or the second distance is smaller than a second threshold value, the first address is included in the delivery range of the storage warehouse.

9. A delivery range determination device, characterized in that: The device comprises: An acquisition module, used for acquiring feedback information from a user, wherein the feedback information is used for indicating that the first address does not have a delivery service; A determination module is used to determine the storage warehouse closest to the first address and the distance from the storage warehouse to the first address; if the distance is less than a preset threshold, the first address is included in the distribution range of the storage warehouse.

10. A delivery range determination device, characterized in that: include: A processor, wherein the processor is coupled to a memory, the memory is used to store a computer program or an instruction, and the processor is used to execute the computer program or the instruction to implement the method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that: It stores a computer program executable by a computer device, and when the program is run on the computer device, the computer device executes the steps of any method described in claims 1 to 8.

12. A computer program product, characterized in that When the method is executed on a computer, the computer is enabled to execute the method according to any one of claims 1 to 8.

Images