Method and system for acquiring receiving preference of consumer based on natural language processing model

Abstract

The invention provides a method and system for obtaining a consumer receiving preference based on a natural language processing model, and relates to the technical field of express logistics transportation, and the method comprises the steps: obtaining the sentence pattern content of a target order in a preset region; splitting the sentence pattern content according to an express industry lexicon to obtain a plurality of segmented words; matching the character representation corresponding to each segmented word from an express industry lexicon, and combining to obtain a part-of-speech expression corresponding to the sentence pattern content; and identifying the part-of-speech expression according to a preset matching rule, and determining an order sending willingness label corresponding to the target order. By implementing the method, the express company is helped to automatically identify and extract the receiving preference of the consumer by means of the receiving address information under the background of the private express sheet, the personalized delivery service capability of the express company is improved, and the receiving experience and satisfaction of the consumer are improved.

Description

Technical Field

[0001] The present application relates to the field of express logistics and transportation technology, and in particular to a method and system for acquiring consumer delivery preferences based on a natural language processing model. Background Art

[0002] With the rapid development of the Internet, e-commerce has become an important way for people to shop in their daily lives. In order to regulate the postal and express delivery service industries and protect the rights and interests of consumers, the state has continuously introduced relevant new postal regulations. These new regulations emphasize the importance of consumer experience, requiring e-commerce platforms and express delivery companies to pay more attention to consumers' shopping experience in their services. Especially in terms of commodity delivery services, the demand has gradually changed from a single fast service to a diversified and personalized service.

[0003] In related technologies, in order to protect consumers' personal privacy information, merchants or e-commerce platforms generally hide consumers' sensitive information (such as name, phone number, and detailed address) on express delivery notes and generate privacy labels.

[0004] However, one of the important measures to improve consumer experience is on-demand delivery. However, in the context of privacy labels, there are e-commerce platforms and merchants between consumers and express delivery companies, which leads to the disconnection between consumers and express delivery companies, making it difficult for express delivery companies to meet consumers' delivery needs. Summary of the invention

[0005] The present application provides a method and system for acquiring consumer delivery preferences based on a natural language processing model, which is used to solve the problem that express delivery companies are unable to determine consumer delivery preferences based on the privacy delivery bill of lading.

[0006] In a first aspect, the present application provides a method for obtaining consumer delivery preferences based on a natural language processing model, which is applied to an express delivery management system. The method includes: Obtain the sentence content of the target order in a preset area, where the preset area is determined based on the location of the delivery address in the target order; The sentence content is split according to the express industry vocabulary to obtain multiple word segments, and the express industry vocabulary includes a translation engine and multiple word segmentation sets; Match the character representation of each word segment from the express industry vocabulary, and combine them to obtain the part-of-speech expression corresponding to the sentence content; The part-of-speech expression is identified according to a preset matching rule to determine the order dispatch intention label corresponding to the target order. The preset matching rule is related to the sentence type corresponding to the sentence content.

[0007] Through the above embodiments, the express management system overcomes the problem in the prior art that it is difficult to obtain consumer delivery needs, and helps express companies to automatically identify and extract consumers' delivery preferences based on delivery address information under the background of privacy labels, thereby improving the personalized delivery service capabilities of express companies and enhancing consumers' delivery experience and satisfaction.

[0008] In some embodiments, the step of splitting the sentence content according to the express industry vocabulary to obtain multiple word segments specifically includes: Load online or offline courier industry word library. When the courier industry word library is online, it supports user editing. The sentence content is preprocessed using the express industry vocabulary, which includes cleaning special characters, simplifying traditional Chinese characters, and converting variant words; The preprocessed sentence content is split according to the express industry vocabulary to obtain multiple word segmentation combinations.

[0009] Through the above embodiment, the express management system can split the order address through the express industry vocabulary to obtain multiple word segmentation combinations, thereby improving the accuracy of determining the consumer's delivery preference based on word segmentation.

[0010] In some embodiments, the step of matching the character representation corresponding to each word segment from the express industry vocabulary and combining to obtain the part-of-speech expression corresponding to the sentence content specifically includes: Detect whether the sentence content conforms to a special sentence type, which includes a preset regular expression type, a combination type, and a special processing type; If so, the sentence content is simplified using the processing rules corresponding to the special sentence type; The character representation corresponding to each word segment is matched from the express industry vocabulary, and the part-of-speech expression corresponding to the sentence content is obtained by combining them.

[0011] Through the above embodiments, the express management system uses targeted simplification rules to normalize special sentence types, reducing the complexity of the part-of-speech expression. Finally, based on simplification, part-of-speech tagging is performed in combination with the position and attribute information of the segmentation, so that the part-of-speech expression can more accurately reflect the semantic structure of the sentence, thereby improving the accuracy of determining the consumer's delivery preference based on the part-of-speech expression.

[0012] In some embodiments, after the step of identifying the part-of-speech expression according to the preset matching rule and determining the order dispatch intention label corresponding to the target order, the method further includes: Obtain complaint data and negative review data uploaded by consumers; Build a consumer complaint profile based on complaint data and negative review data.

[0013] Through the above embodiment, the express management system obtains the complaint records and negative review information of consumers in historical orders, determines the main dissatisfaction points of consumers with express services, and forms a user profile with key elements such as the object of complaint, reason for complaint, and frequency of complaint. This facilitates express companies to make targeted improvements to the pain points of different user groups and improve service quality.

[0014] In some embodiments, after the step of constructing a complaint profile of a consumer based on the complaint data and the negative review data, the step further includes: Build a consumer demand profile based on order dispatch intention labels; Construct a consumer delivery demand form based on the consumer’s personal information, complaint profile and demand profile.

[0015] Through the above embodiment, the express management system integrates consumers' personal information, complaint portraits and demand portraits to generate a multi-dimensional and comprehensive delivery demand table, covering the entire process data such as user attributes, historical behaviors and current demands, providing guidance for the express company's precise marketing and service operations.

[0016] In some embodiments, after the step of identifying the part-of-speech expression according to the preset matching rule and determining the order dispatch intention label corresponding to the target order, the method further includes: The dispatch intention labels are classified and counted, and the distribution of dispatch intention labels corresponding to different dispatch intention requirements is updated. The dispatch intention requirements include behavioral requirements, spatial requirements, and time requirements.

[0017] Dynamically display the distribution of order dispatch intention labels.

[0018] Through the above embodiments, the express management system performs cluster analysis on the dispatch intention labels, summarizes several typical dispatch intention requirements, including behavioral requirements (such as the way of signing for delivery), spatial requirements (such as the place of receipt), time requirements (such as the delivery period), etc., and then counts the distribution of each type of dispatch intention label, quantifies the popularity and trend of different requirements, and forms a dynamic display of the distribution of dispatch intention labels. Through an intuitive visual interface, express companies can dynamically monitor market changes, continuously improve service strategies, and continuously consolidate and enhance user stickiness.

[0019] In some embodiments, after the step of identifying the part-of-speech expression according to the preset matching rule and determining the order dispatch intention label corresponding to the target order, the method further includes: The dispatch intention label is marked and displayed on the AOI area data map.

[0020] Through the above embodiments, the express management system directly presents the spatiotemporal characteristics of dispatching demand by marking the distribution of receiving preferences of consumers in different regions on the geographic information layer. It can assist express companies in formulating differentiated value-added services based on regional user portraits to meet the personalized needs of different consumer groups.

[0021] In a second aspect, the present application provides a courier management system, the courier management system comprising: one or more processors and a memory; The memory is coupled to the one or more processors, and the memory is used to store computer program code, which includes computer instructions. The one or more processors call the computer instructions so that the express management system can implement a method for obtaining consumer delivery preferences based on an NLP natural language processing model provided in the above embodiment, which will not be repeated here.

[0022] In a third aspect, the present application provides a computer-readable storage medium, including instructions. When the instructions are executed on an express management system, the express management system can implement a method for obtaining consumer delivery preferences based on an NLP natural language processing model provided in the above embodiment, which will not be repeated here.

[0023] In a fourth aspect, the present application provides a computer program product. When the computer program product runs on an express management system, the express management system can implement a method for obtaining consumer delivery preferences based on an NLP natural language processing model provided in the above embodiment, which will not be repeated here.

[0024] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. The express delivery management system overcomes the problem of difficulty in obtaining consumer delivery needs under the background of privacy labels. Through intelligent analysis of order delivery address information, it realizes automatic identification and extraction of consumer delivery preferences. In this process, the express industry vocabulary is used to segment the address, and part-of-speech is tagged based on the segmentation position and attributes. The part-of-speech expressions are matched according to preset rules, so as to accurately judge the consumer's willingness to dispatch orders, greatly improving the express company's personalized delivery service capabilities and effectively improving consumers' delivery experience and satisfaction.

[0025] 2. The express delivery management system obtains complaint records and negative review information from consumers' historical orders, analyzes and forms a user complaint profile with the object of complaint, reason, and frequency as key elements, and builds a consumer demand profile in combination with the dispatch willingness label, and finally generates a multi-dimensional delivery demand table covering user attributes, historical behaviors, and current needs. Using this demand table, express delivery companies can formulate targeted service improvement strategies and marketing plans based on the pain points and needs of different user groups to achieve precise operations.

[0026] 3. The express delivery management system uses cluster analysis and statistical methods to summarize typical demand types such as behavior, space, and time of dispatch willingness, quantify the popularity trend of different demands, and dynamically display the distribution of dispatch willingness labels through a visual interface to help express delivery companies monitor market changes in real time and continuously optimize services. In addition, the system also combines dispatch willingness labels with AOI area maps to intuitively present the spatiotemporal characteristics of dispatch demand in different regions, providing decision support for express delivery companies to formulate differentiated value-added services tailored to local conditions and meet the personalized needs of different consumer groups. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is a flow chart of a method for obtaining consumer delivery preferences based on an NLP natural language processing model in an embodiment of the present application; Figure 2 This is another flowchart of a method for obtaining consumer delivery preferences based on an NLP natural language processing model in an embodiment of the present application; Figure 3 This is a schematic diagram of an exemplary scenario in which the express management system generates a dispatch intention label based on an express order in an embodiment of the present application; Figure 4 This is a schematic diagram of an exemplary scenario in which the express management system in the embodiment of the present application analyzes and extracts the order address and determines the order dispatch intention label; Figure 5 This is a schematic diagram of an exemplary scenario in which the express management system in the embodiment of the present application dynamically displays the distribution of order dispatch intention labels; Figure 6 It is a schematic diagram of the structure of a physical device of the express delivery management system in an embodiment of the present application. DETAILED DESCRIPTION

[0028] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification and appended claims of the present application, the singular expressions "one", "a kind of", "said", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations comprising one or more listed items.

[0029] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0030] In the case of privacy-preserving express delivery bills, it is difficult for consumers to obtain their delivery preferences. The technical solution provided by this application can parse consumers’ delivery preferences from the delivery addresses filled in by consumers based on the NLP natural language processing model. Figure 3 The figure is a schematic diagram of an exemplary scenario in which an express management system generates a dispatch intention label based on an express order in an embodiment of the present application.

[0031] The express management system provided in the embodiment of the present application identifies the delivery address in the express order and inputs it into the NLP natural language processing model. The NLP natural language processing model performs word segmentation and part-of-speech tagging on the delivery address, and outputs the dispatch intention label to assist the express company in quickly understanding the consumer's delivery preferences.

[0032] For ease of understanding, the following is a flow chart of the method provided in this implementation. Figure 1 The figure is a flow chart of a method for obtaining consumer delivery preferences based on an NLP natural language processing model in an embodiment of the present application.

[0033] S101, obtaining sentence content of a target order in a preset area.

[0034] Specifically, the express management system scans the address information of the preset area in the target express order through a camera or other equipment to obtain the sentence content used to analyze the label of the willingness to dispatch the order. Among them, the preset area generally refers to the area where the consumer's delivery address is printed in the express order, and the consumer's demand information is generally printed at the end of the delivery address. The demand information is the sentence content used to analyze the label of the willingness to dispatch the order.

[0035] For example, in a specific embodiment, the printed content of the address information in a courier order is: "Please deliver to Fengchao locker (rejection station!) at No. *, Building *, XX community, XX road ...

[0036] S102, splitting the sentence content according to the express industry vocabulary to obtain multiple word segments.

[0037] Among them, the express industry vocabulary contains a large number of express industry related word segments, including but not limited to express industry specific terms (such as common express company names, service types, express document fields, etc.), address element related vocabulary (such as administrative district names at all levels, common road names, building names, community names, place name suffixes, etc.), delivery service related vocabulary (such as delivery, delivery, signing for, collection on behalf of, self-pickup, post station, Fengchao, express cabinet, etc.), variant words (such as "Caiwu Post Station" (a variant of Cainiao Post Station)), common words (such as "CN" (the first letter of a certain brand)), as well as time and number related vocabulary (such as today, tomorrow, Monday, 12 o'clock, Fengchao cabinet number XXX, etc.).

[0038] The express management system will split the sentence content obtained in step S101, match it with the existing word segments in the express industry vocabulary, and output multiple successfully matched word segments.

[0039] For example, in a specific embodiment, the sentence content is known to be "Please deliver to Fengchao cabinet (rejection station!)", then the segmented words obtained by splitting according to the express industry vocabulary are "please deliver, Fengchao cabinet, rejectment, station".

[0040] It should be noted that the express industry vocabulary includes two parts: an online vocabulary and an offline vocabulary. When the electronic device equipped with the express management system is connected to the Internet, you can choose to use the online vocabulary or the offline vocabulary according to your needs. When the electronic device equipped with the express management system is offline, the express management system automatically uses the offline vocabulary. Among them, the online vocabulary is equipped with an online search engine, which can support users to search online for word combinations such as addresses and service sites, and supports adding new searched words to the express industry vocabulary to expand the express industry vocabulary. In addition, users can also modify existing word segments according to actual needs, which is not limited here.

[0041] S103, matching the character representation corresponding to each word segment from the express industry vocabulary, and combining to obtain a part-of-speech expression corresponding to the sentence content.

[0042] Specific as Figure 4 The figure is a schematic diagram of an exemplary scenario in which the express management system in an embodiment of the present application analyzes and extracts the order address to determine the order dispatch intention label.

[0043] The express management system splits the sample address "Please deliver to Fengchao Cabinet (rejection station!) in Room *, Building *, XXX Community, XXX Road, XXX Road" into multiple participles "Please deliver, Fengchao Cabinet, Rejection, Station". Then, the character representation corresponding to each participle is matched from the express industry vocabulary. For example, "Please deliver" belongs to the positive prefix (represented by "PP"), "Fengchao Cabinet" belongs to the self-collection cabinet noun (represented by "ZT"), "Rejection" belongs to the negative prefix (represented by "NS"), and "Station" belongs to the station name (represented by "YZ"). Then, each participle is combined to obtain the part-of-speech expression "PP&ZT&NS&YZ".

[0044] It should be noted that the above-mentioned self-collection cabinet nouns and post station names are stored in the express industry vocabulary, and can be directly matched and queried. The character representations corresponding to different attributes such as forward prefixes and self-collection cabinet nouns are also stored in the express industry vocabulary, and the express management system can also directly match and query when performing part-of-speech tagging.

[0045] S104: Identify the part-of-speech expression according to the preset matching rules to determine the order dispatch intention label corresponding to the target order.

[0046] Specific as Figure 4 The figure is a schematic diagram of an exemplary scenario in which the express management system in an embodiment of the present application analyzes and extracts the order address to determine the order dispatch intention label.

[0047] After the express management system generates the part-of-speech expression corresponding to the sentence content, it splits the part-of-speech expression based on the built-in splitting rules to obtain the dispatch intention label.

[0048] Let's continue with the example in step S103. The express management system splits the part-of-speech expression "PP&ZT&NS&YZ" according to the splitting rules of "positive prefix + self-pickup cabinet noun" and "negative prefix + post station name", and obtains two groups of part-of-speech expressions "PP&ZT (put in the self-pickup cabinet)" and "NS&YZ (do not put in the post station)". The two groups of part-of-speech expressions are then translated to obtain the dispatch intention labels "put in the self-pickup cabinet" and "do not put in the post station".

[0049] In the above embodiment, the express management system overcomes the problem in the prior art that it is difficult to obtain consumer delivery needs, and helps express companies to automatically identify and extract consumers' delivery preferences based on delivery address information under the background of privacy labels, thereby greatly improving the personalized delivery service capabilities of express companies and enhancing consumers' delivery experience and satisfaction.

[0050] The following is a more detailed description of the process of the method provided by this embodiment. Figure 2As shown, it is another flowchart of a method for obtaining consumer delivery preferences based on an NLP natural language processing model in an embodiment of the present application.

[0051] S201, pre-processing the sentence content through the express industry vocabulary.

[0052] The preprocessing process includes but is not limited to cleaning special characters, simplifying traditional Chinese characters and escaping variant words. Cleaning special characters means that the order address may contain some special characters, such as punctuation marks, emoticons, etc., which need to be cleaned first to avoid affecting the subsequent word segmentation effect; simplifying traditional Chinese characters means that some merchants and consumers may use traditional Chinese characters, and in order to process them uniformly, traditional Chinese characters need to be converted into corresponding simplified Chinese characters; escaping variant words means that some words may have variant forms, such as "Express Box" and "Express Delivery Box", etc., and variant words need to be uniformly converted into standard forms.

[0053] For example, the original sentence content is: "Please deliver to Fengchao locker (rejection station!), THX~[smile]". After cleaning special characters and converting traditional and simplified Chinese to "Please deliver to Fengchao locker (rejection station), thank you". The sentence structure after preprocessing is more standardized, which is convenient for subsequent word segmentation and analysis.

[0054] It should be noted that the express industry vocabulary has built-in common special characters, traditional and simplified comparison tables, variant word mapping tables, etc., which can perform fast and accurate normalization preprocessing on sentence content, and are not limited here.

[0055] In the above embodiment, the express management system can split the order address through the express industry vocabulary to obtain multiple word segmentation combinations, thereby improving the accuracy of determining the consumer's delivery preference based on word segmentation.

[0056] S202: split the preprocessed sentence pattern according to the express industry vocabulary to obtain multiple word segmentation combinations.

[0057] This step is the same as step S102 and will not be repeated here.

[0058] S203. Whether the sentence conforms to the special sentence type.

[0059] In the embodiment of the present application, special sentence types include but are not limited to preset regular expression types, combination types, and special processing types. For example, the sentence "When you arrive, call and no one answers, put it at the 302 door" meets the preset regular expression type ".* put it at the [0-9]{3,4} door"; the sentence "put it next to the hospital next to the school" meets the combination type, that is, the place name combination; the sentence "put it at the Cainiao Post Station" meets the special processing type, that is, the brand name needs to be identified.

[0060] After detecting that the current sentence conforms to the special sentence type, the express management system enters step S204 to first simplify the sentence content according to the processing rules corresponding to the special sentence type. Otherwise, it directly enters step S205 to perform part-of-speech tagging on the sentence.

[0061] It should be noted that the preset regular expression types, combination types, and special processing types provided in the above examples are only examples. In addition, there may be other special sentence types and specific matching rules corresponding to special sentence types, which are not limited here.

[0062] S204: Simplify the sentence content using the processing rules corresponding to the special sentence type.

[0063] For example, the sentence "If no one answers the phone when I get there, put it at door 302" can be simplified to "put it at door 302" according to the processing rule of the preset regular expression type ".*put it at [0-9]{3,4}". Similarly, the processing rules corresponding to other special sentence types simplify the sentences in a similar way, which is not repeated here.

[0064] In the above embodiment, the express management system uses targeted simplification rules to normalize special sentence types, reducing the complexity of the part-of-speech expression. Finally, based on simplification, part-of-speech tagging is performed in combination with the position and attribute information of the segmentation, so that the part-of-speech expression can more accurately reflect the semantic structure of the sentence, thereby improving the accuracy of determining the consumer's delivery preference based on the part-of-speech expression.

[0065] S205. Match the character representation corresponding to each word segment from the express industry vocabulary, and combine to obtain a part-of-speech expression corresponding to the sentence content.

[0066] This step is the same as step S103 and will not be repeated here.

[0067] S206: Identify the part-of-speech expression according to the preset matching rules to determine the order dispatch intention label corresponding to the target order.

[0068] This step is the same as step S104 and will not be repeated here.

[0069] S207, classify and count the order dispatching intention labels, and update the distribution of order dispatching intention labels corresponding to different order dispatching intention requirements.

[0070] After determining the dispatch intention label corresponding to the target order, the express management system will update the classification statistics of the dispatch intention label for the day. Among them, the classification statistics include the number of express orders and the number of consumers corresponding to each dispatch intention label. Of course, the dispatch intention labels can be classified (such as divided into behavioral needs, spatial needs, and time needs, etc.) or divided in more detail. For example, the label "place collection point" can be divided into "place at the door / gate post", "place at the property", "place at the security room", and "place at the gate of the community", etc., which are not limited here.

[0071] S208. Dynamically display the distribution of order dispatch intention labels.

[0072] After step S207, the express management system dynamically displays the distribution of order dispatch intention labels. Figure 5 The figure is a schematic diagram of an exemplary scenario in which the express management system dynamically displays the distribution of dispatch intention tags in an embodiment of the present application. The express management system updates the table data in real time according to the identified dispatch intention tags to realize the dynamic display of the distribution of dispatch intention tags.

[0073] In the above embodiment, the express management system integrates the consumer's personal information, complaint profile and demand profile to generate a multi-dimensional and comprehensive delivery demand table, covering the entire process data such as user attributes, historical behavior and current demand, providing guidance for the express company's precise marketing and service operations.

[0074] In the above embodiment, the express management system performs cluster analysis on the dispatch intention labels, summarizes several typical dispatch intention requirements, including behavioral requirements (such as the way of signing for delivery), spatial requirements (such as the place of delivery), time requirements (such as the delivery time period), etc., and then counts the distribution of each type of dispatch intention label, quantifies the popularity and trend of different requirements, and forms a dynamic display of the distribution of dispatch intention labels. Through an intuitive visual interface, express companies can dynamically monitor market changes, continuously improve service strategies, and continuously consolidate and enhance user stickiness.

[0075] S209: Mark and display the order dispatch intention label on the AOI area data map.

[0076] After obtaining the consumer's order dispatching intention label, the express delivery management system can further combine these label information with the geographic information system (GIS) to intuitively mark and display the electronic map of the area of ​​interest (AOI). The purpose of this is to help express delivery companies intuitively understand the distribution characteristics of consumers' delivery preferences in different regions, so as to formulate more refined and differentiated delivery strategies.

[0077] For example, the express management system can determine the AOI area where the consumer is located based on the province, city, and district information in the consumer's address. Then, the consumer's order dispatching intention label (such as "delivery to express lockers", "home delivery", "weekend delivery", etc.) is marked on the data map of the corresponding AOI area. After accumulating a certain amount of order data, the system can statistically analyze the distribution ratio and quantity scale of different order dispatching intention labels in each AOI area, generate visualization effects such as heat maps or bubble maps that reflect the characteristics of regional delivery demand, and intuitively present the differences in consumers' delivery preferences in the region.

[0078] In the above embodiment, the express management system directly presents the spatiotemporal characteristics of dispatch demand by marking the distribution of consumers' delivery preferences in different regions on the geographic information layer. It can assist express companies in formulating differentiated value-added services based on regional user portraits to meet the personalized needs of different consumer groups.

[0079] S210. Construct a complaint profile based on the complaint data and negative review data uploaded by consumers.

[0080] The express delivery management system collects and analyzes the complaint data and negative review data actively uploaded by consumers in historical orders, summarizes the main reasons for consumers' dissatisfaction with express delivery services, and thus constructs a user complaint portrait with the complaint object, complaint reason, and complaint frequency as core elements.

[0081] Specifically, the express delivery management system must first obtain the complaint records and negative review information submitted by consumers through online order evaluation, telephone / online customer service, official apps, etc. Then, it uses natural language processing technologies such as text mining and sentiment analysis to perform semantic understanding and extract key information from the complaint content, and identify the objects of the complaint (such as the courier's service attitude, damaged parcels, delivery time, etc.), causes (such as courier's rude words, serious deformation of the package, overdue delivery, etc.), and frequency (such as the number of times such complaints occur and the time distribution pattern).

[0082] On this basis, the express management system can characterize consumers' complaint behavior in multiple dimensions and construct a complaint portrait that reflects their concerns and dissatisfaction with express services, such as "concerned about the courier's service attitude, with an average of two complaints about rude language per month" and "concerned about the integrity of the express delivery, with two complaints about serious damage in half a year".

[0083] In the above embodiment, the express management system obtains the complaint records and negative review information of consumers in historical orders, determines the main dissatisfaction points of consumers with express services, and forms a user profile with key elements such as the object of complaint, reason for complaint, and frequency of complaint. This facilitates express companies to make targeted improvements to the pain points of different user groups and improve service quality.

[0084] S211. Build a consumer demand profile based on order dispatching intention labels.

[0085] In some implementations, the express delivery management system clusters and groups the labels of dispatch willingness according to different dimensions such as behavioral attributes, time attributes, and spatial attributes. Taking behavioral attributes as an example, dispatch method preferences such as "delivery to express lockers" and "home delivery" can be classified as the theme of "receiving behavior habits"; taking time attributes as an example, delivery time requirements such as "weekend delivery" and "holiday delivery" can be classified as the theme of "receiving time preference". As a result, the system has initially clarified several main characteristic dimensions of consumer demand.

[0086] On the basis of clustering and grouping, the express delivery management system combines other consumer portrait tags (such as demographic attributes, consumption behavior, etc.) to deeply explore the key driving factors under each demand theme. For example, the behavioral habit of "delivery in express lockers" may be due to the fact that consumers do not want to meet couriers face to face (such as during the epidemic), are often away from home and it is difficult to sign for the package in person, and enjoy the flexibility of 24-hour self-pickup; the time preference of "weekend delivery" may be due to the fact that consumers are busy at work on weekdays and only have time to receive express delivery at ease on weekends.

[0087] Through the above analysis, the express delivery management system finally formed a set of structured consumer demand portraits, covering a series of standardized labels such as receiving behavior habits (and the reasons behind them), receiving time preferences (and the reasons behind them), receiving space preferences (and the reasons behind them), etc., revealing the deep-seated motivations of consumers for expressing their willingness to place corresponding orders at multiple levels.

[0088] S212. Construct a consumer delivery demand table based on the consumer's personal information, complaint profile and demand profile.

[0089] After obtaining the consumer's complaint and demand profiles, the express delivery management system can further integrate the consumer's personal information (such as name, contact information, delivery address, identity type, etc.) to build a comprehensive "consumer delivery demand table". This demand table provides a panoramic and diversified portrayal of the consumer's delivery characteristics from three perspectives: user attributes, historical complaints, and current needs, and strives to provide express delivery companies with the richest and most detailed user insights possible.

[0090] In the above embodiment, the express management system integrates the consumer's personal information, complaint profile and demand profile to generate a multi-dimensional and comprehensive delivery demand table, covering the entire process data such as user attributes, historical behavior and current demand, providing guidance for the express company's precise marketing and service operations.

[0091] The express delivery management system according to the embodiment of the present invention is applied to electronic equipment. Figure 6 A schematic diagram of the architecture of an electronic device suitable for implementing an embodiment of the present invention is shown.

[0092] It should be noted that Figure 6 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0093] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments can be completed by instructions (computer programs), or by controlling related hardware through instructions (computer programs), and the instructions can be stored in a computer-readable storage medium and loaded and executed by a processor. The electronic device of this embodiment includes a storage medium and a processor, wherein a plurality of instructions are stored in the storage medium, and the instructions can be loaded by the processor to execute any step of the method provided in the embodiment of the present invention.

[0094] Specifically, the storage medium and the processor are electrically connected directly or indirectly to realize data transmission or interaction. For example, these elements can be electrically connected to each other through one or more signal lines. The storage medium stores computer execution instructions for implementing the data access control method, including at least one software function module that can be stored in the storage medium in the form of software or firmware. The processor executes various functional applications and data processing by running the software program and module stored in the storage medium. The storage medium can be, but is not limited to, random access storage medium (Random Access Memory, referred to as: RAM), read-only storage medium (Read Only Memory, referred to as: ROM), programmable read-only storage medium (Programmable Read-Only Memory, referred to as: PROM), erasable read-only storage medium (Erasable Programmable Read-Only Memory, referred to as: EPROM), electrically erasable read-only storage medium (Electric Erasable Programmable Read-Only Memory, referred to as: EEPROM), etc. Among them, the storage medium is used to store programs, and the processor executes the program after receiving the execution instruction.

[0095] Furthermore, the software programs and modules in the above-mentioned storage medium may also include an operating system, which may include various software components and / or drivers for managing system tasks (such as memory management, storage device control, power management, etc.), and may communicate with various hardware or software components to provide an operating environment for other software components. The processor may be an integrated circuit chip having signal processing capabilities. The above-mentioned processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc., which may implement or execute the various methods, steps, and logic flow diagrams disclosed in this embodiment. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0096] Since the instructions stored in the storage medium can execute the steps in any method provided in the embodiments of the present invention, the beneficial effects of any method provided in the embodiments of the present invention can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0097] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for obtaining consumer delivery preferences based on a natural language processing model, applied to an express delivery management system, characterized in that: The method comprises: Acquire sentence content of a target order in a preset area, where the preset area is determined according to the location of a delivery address in the target order; Splitting the sentence content according to the express industry vocabulary to obtain a plurality of segmented words, wherein the express industry vocabulary includes a translation engine and a plurality of segmented word sets; Matching the character representation corresponding to each segmented word from the express industry vocabulary, combining to obtain a part-of-speech expression corresponding to the sentence content, wherein the mapping relationship between the segmented word and the character representation is stored in the express industry vocabulary; The part-of-speech expression is identified according to a preset matching rule to determine the order dispatch intention label corresponding to the target order, and the preset matching rule is related to the sentence type corresponding to the sentence content.

2. The method according to claim 1, characterized in that: The step of splitting the sentence content according to the express industry vocabulary to obtain multiple word segments specifically includes: Loading an online or offline courier industry vocabulary library, wherein the courier industry vocabulary library supports user editing when it is online; Preprocessing the sentence content through the express industry vocabulary, wherein the preprocessing includes cleaning special characters, simplifying traditional Chinese characters, and converting variant words; The preprocessed sentence content is split according to the express industry vocabulary to obtain a plurality of word segmentation combinations.

3. The method according to claim 1, characterized in that The step of matching the character representation corresponding to each word segment from the express industry vocabulary to obtain the part-of-speech expression corresponding to the sentence content specifically includes: Detecting whether the sentence content conforms to a special sentence type, wherein the special sentence type includes a preset regular expression type, a combination type, and a special processing type; If so, simplify the sentence content using the processing rules corresponding to the special sentence type; The character representation corresponding to each word segment is matched from the express industry vocabulary, and the part-of-speech expression corresponding to the sentence content is obtained by combining them.

4. The method according to claim 1, characterized in that After the step of identifying the part-of-speech expression according to the preset matching rule and determining the order dispatch intention label corresponding to the target order, the method further includes: Obtain complaint data and negative review data uploaded by consumers; A complaint profile of the consumer is constructed based on the complaint data and negative review data.

5. The method according to claim 4, characterized in that After the step of constructing the consumer's complaint profile based on the complaint data and the negative review data, the step further includes: Constructing a demand profile of the consumer based on the order dispatching intention label; A consumer delivery demand table is constructed based on the consumer's personal information, the complaint profile and the demand profile.

6. The method according to claim 1, characterized in that After the step of identifying the part-of-speech expression according to the preset matching rule and determining the order dispatch intention label corresponding to the target order, the method further includes: Classify and count the order dispatching intention labels, and update the distribution of order dispatching intention labels corresponding to different order dispatching intention requirements, wherein the order dispatching intention requirements include behavioral requirements, spatial requirements, and time requirements; The distribution of the order dispatching intention labels is dynamically displayed.

7. The method according to claim 1, characterized in that After the step of identifying the part-of-speech expression according to the preset matching rule and determining the order dispatch intention label corresponding to the target order, the method further includes: The order dispatch intention label is marked and displayed on the AOI area data map.

8. A courier management system, characterized in that: The express management system includes: one or more processors and memory; The memory is coupled to the one or more processors, and the memory is used to store computer program codes, wherein the computer program codes include computer instructions, and the one or more processors call the computer instructions to enable the express management system to execute the method according to any one of claims 1 to 7.

9. A computer-readable storage medium comprising instructions, characterized in that: When the instruction is executed on the express management system, the express management system executes the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that When the computer program product is executed on an express management system, the express management system is enabled to execute the method according to any one of claims 1 to 7.

Images