A Multi-Region Transportation Service Mode Matching Method and System

Abstract

The present invention discloses a multi - area transportation service mode matching method and system, specifically relating to the technical field of transportation services, which includes the following steps: Divide user identities into order - placing users and order - receiving users, and respectively obtain the order - placing requirement set of the target order - placing user and the service sets of all order - receiving users in a non - order - receiving state; First, perform hard - condition screening to retain the order - receiving users that meet the hard conditions, and then perform flexible - condition screening; If there are order - receiving users that meet all the flexible conditions, directly match; Otherwise, conduct portrait analysis using K - means clustering based on the historical records of the target order - placing user, and evaluate the matching failure rate; According to the failure - rate result, successively establish negotiation communication channels until a consistent match is achieved; The present invention can deeply understand the demand characteristics and behavior patterns of order - placing users, provide highly personalized matching services, improve the service quality of the platform, and further enhance user loyalty and satisfaction.

Description

Technical Field

[0001] The present invention relates to the technical field of transportation services. More specifically, the present invention relates to a multi-region transportation service mode matching method and system. Background Art

[0002] In existing service trading platforms, logistics scheduling systems, and e-commerce platforms, the matching between user demands and service providers is a key link. Traditional matching methods are difficult to cope with complex and changeable user demands, especially when users have different priorities or flexibilities for multiple requirement items. These traditional methods often fail to effectively solve the problem of needing to readjust or negotiate after a matching failure, resulting in a decline in user experience and resource waste. After a matching failure, users usually can only manually adjust their requirements or wait for new service providers to appear. This process is both time-consuming and inefficient, wasting users' time and platform resources, and also reducing users' usage experience and satisfaction. Therefore, a multi-region transportation service mode matching method and system are proposed here to solve the above problems. Summary of the Invention

[0003] To achieve the above object, the present invention provides the following technical solutions:

[0004] A multi-region transportation service mode matching method, comprising the following steps:

[0005] Divide all user identities, set two roles: order-issuing users and order-receiving users, then respectively obtain the order-issuing requirement set of the target order-issuing user, and at the same time obtain the order-receiving service sets of all order-receiving users in the non-order-receiving state;

[0006] Based on the order-issuing requirement set and the order-receiving service sets, perform hard condition screening, retain the order-receiving users who meet the hard conditions, and then perform flexible condition screening. If there are order-receiving users who meet all the flexible conditions, obtain the target order-receiving user according to the preset satisfaction strategy and match with the target order-issuing user;

[0007] If there are no order-receiving users who meet all the flexible conditions, obtain the historical order-issuing records of the target order-issuing user and perform portrait analysis using the K-means clustering algorithm, and evaluate the matching failure rate based on the results of the portrait analysis;

[0008] Based on the matching failure rate evaluation results, successively establish negotiation communication channels for the target order-issuing user until an agreement is reached. At this time, the order-receiving user corresponding to the negotiation communication channel matches with the target order-issuing user.

[0009] In a preferred embodiment, the order - issuing requirement set includes a hard - requirement subset and a flexible - requirement subset. The hard - requirement subset refers to the set composed of all the hard requirements that the target order - issuing user does not accept changes to, and the flexible - requirement subset refers to the set composed of all the flexible requirements that the target order - issuing user can accept changes to;

[0010] The order - receiving service set includes a hard - service subset and a flexible - service subset. The hard - service subset refers to the set composed of all the hard services that the order - receiving user does not accept changes to, and the flexible - service subset refers to the set composed of all the flexible services that the order - receiving user can accept changes to.

[0011] In a preferred embodiment, performing hard - condition screening means that: the order - receiving service set meets all the hard requirements that are not accepted for change in the hard - requirement subset.

[0012] In a preferred embodiment, performing flexible - condition screening means that:

[0013] Obtain the flexible - requirement subset in the order - issuing requirement set, and compare each flexible requirement in the flexible - requirement subset one by one with the order - receiving service set of the remaining order - receiving users. If all the flexible requirements are met, mark it as an accepted matching object; if not all the flexible requirements are met, mark it as a rejected matching object.

[0014] In a preferred embodiment, obtaining the target order - receiving user according to a preset satisfaction strategy means that:

[0015] Obtain the historical service scores of all the order - receiving users marked as accepted matching objects, and then sort them in descending order. The order - receiving user at the top is the target order - receiving user.

[0016] In a preferred embodiment, obtaining the historical order - issuing records of the target order - issuing user and performing portrait analysis using the K - means clustering algorithm means that:

[0017] Obtain the flexible - requirement subset in the current order - issuing requirement set of the target order - issuing user, as well as the historical order - issuing records of the target order - issuing user, and extract information according to the flexible - requirement items in the current flexible - requirement subset:

[0018] Mark the flexible - requirement items in the current flexible - requirement subset as TXi respectively, set the target flexible - requirement item, obtain the transaction data at the time of transaction and the submission data at the time of issuance of all successfully - matched orders in the historical order - issuing records of the target order - issuing user, and then calculate the modification amplitude value corresponding to the target flexible - requirement item for each successfully - matched order. The acquisition logic of the modification amplitude value is:

[0019] First, calculate the difference between the transaction data at the time of transaction and the submission data at the time of issuance for the target elasticity requirement item, then divide this difference by the submission data at the time of issuance, and take the absolute value of the resulting ratio to obtain the modified change ratio value;

[0020] Next, calculate the average value of the modified change ratio values corresponding to the target elasticity requirement items of all successfully matched orders to obtain the modified amplitude value and mark it as Mi;

[0021] Use the modified amplitude value Mi of each elasticity requirement item TXi as the feature input for clustering, use the K-means clustering algorithm to perform clustering analysis on the modified amplitude values of all elasticity requirement items, use the elbow method to determine the optimal number of clusters K, the elasticity requirement items in each cluster form a level, and correspond each elasticity requirement item with its divided classification level to obtain the user portrait of the target order-issuing user.

[0022] In a preferred embodiment, the evaluation of the matching failure rate based on the result of portrait analysis means:

[0023] Obtain the user portrait of the target order-issuing user, and then set the weight coefficient for the elasticity requirement items in the current elasticity requirement subset, ; The level of the elasticity requirement item is expressed as , represents the level of the elasticity requirement item corresponding weight coefficient, and K represents the total number of levels;

[0024] Next, use the Hamming distance to represent the matching result of each elasticity requirement item:

[0025] ; represents the Hamming distance corresponding to the elasticity requirement item, and the difference amplitude value is obtained by first subtracting the data corresponding to the elasticity requirement item in the order-receiving service set of the order-receiving user from the requirement data corresponding to the elasticity requirement item in the elasticity requirement subset, then dividing the resulting difference by the requirement data in the elasticity requirement subset, and taking the absolute value of the resulting ratio;

[0026] Calculate the overall difference value D: ; n represents the total number of elasticity requirement items;

[0027] Then substitute it into the pre-trained comprehensive matching degree model formula:

[0028] ; is the parameter for adjusting the sensitivity, is the offset used to adjust the starting point of the matching degree, is the comprehensive matching degree;

[0029] The matching failure rate is obtained by subtracting the comprehensive matching degree from the value 1. It is obtained.

[0030] In a preferred embodiment, a multi-region transportation service mode matching system includes:

[0031] An identity information collection module divides all user identities, sets two roles: the order-issuing user and the order-receiving user, then respectively obtains the order-issuing requirement set of the target order-issuing user, and at the same time obtains the order-receiving service sets of all order-receiving users in the non-order-receiving state;

[0032] A screening module performs hard condition screening based on the order-issuing requirement set and the order-receiving service sets, retains the order-receiving users that meet the hard conditions, and then performs flexible condition screening;

[0033] A negotiation module, if there are no order-receiving users that meet all the flexible conditions, obtains the historical order-issuing records of the target order-issuing user and uses the K-means clustering algorithm for portrait analysis, and based on the results of the portrait analysis, evaluates the matching failure rate. Based on the matching failure rate evaluation results, a negotiation communication channel is successively established for the target order-issuing user;

[0034] A matching module, if there are order-receiving users that meet all the flexible conditions, obtains the target order-receiving user according to the preset satisfaction strategy and reaches a match with the target order-issuing user, and when the negotiation is consistent, the order-receiving user corresponding to the negotiation communication channel reaches a match with the target order-issuing user.

[0035] The technical effects and advantages of the present invention:

[0036] The present invention can intelligently sort all potential order-receiving users based on detailed historical data and real-time matching failure rate evaluation, and preferentially select the order-receiving user with the highest matching success rate for negotiation. It can gradually eliminate the factors that may lead to matching failure, continuously optimize the matching conditions, and finally ensure that the order-issuing user reaches a match with the most suitable order-receiving user. Compared with the traditional static matching method, the present invention significantly improves the matching success rate, enables complex and variable requirements to be met more efficiently, and greatly improves the matching quality of the platform.

[0037] Through the intelligent sorting and negotiation mechanism, the present invention reduces the unnecessary communication and negotiation time between users. The system will automatically preferentially select those order-receiving users who are most likely to reach an agreement with the order-issuing user for communication, which can avoid a large number of ineffective attempts and repeated communications, and greatly reduce the communication cost during the matching process. In this way, the system not only improves the matching efficiency but also optimizes the use of resources, enabling the platform to better handle a large number of matching requests while ensuring the improvement of the user experience.

[0038] The present invention allows the order - issuing user and the order - receiving user to adjust their respective requirements and service conditions in real - time during the negotiation process. Moreover, the system can help users clearly understand the actual effects of each adjustment by dynamically adjusting the matching conditions and recalculating the matching failure rate. This flexible matching mechanism enables users to explore different matching solutions more freely. Especially when facing complex and variable requirements, they can quickly find the most suitable solution through negotiation. This not only increases the matching success rate but also enables the system to better adapt to the personalized needs of different users.

[0039] The present invention significantly reduces the trial - and - error cost of blind matching by preferentially selecting order - receiving users with a high matching success rate for negotiation. In traditional systems, users may need to try multiple times to find a suitable matching partner. However, through intelligent sorting and multi - round negotiation mechanisms, the present invention greatly reduces the number of ineffective attempts. Users can find suitable matching partners more quickly, avoiding unnecessary time waste and resource consumption, thus making the operation of the platform more efficient and significantly improving user satisfaction.

[0040] By using the K - means clustering algorithm to analyze the historical data of order - issuing users and combining with the priority division of flexible requirements, the system can deeply understand the demand characteristics and behavior patterns of order - issuing users. This precise user profiling analysis enables the system to provide highly personalized matching services for each order - issuing user. According to the specific needs and priorities of users, the system can intelligently adjust the matching strategy, thereby improving the accuracy and success rate of matching while meeting user needs. This personalized matching service not only improves the service quality of the platform but also further enhances user loyalty and satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] For the convenience of those skilled in the art to understand, the present invention will be further described below in conjunction with the accompanying drawings;

[0042] Figure 1 It is a schematic diagram of a multi - region transportation service mode matching method in the present invention.

[0043] Figure 2 It is a schematic diagram of a multi - region transportation service mode matching system in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0044] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0045] Refer to Figure 1 - Figure 2 The following embodiments are obtained:

[0046] Embodiment 1: A method for matching multi-region transportation service modes, comprising the following steps:

[0047] Divide all user identities, set two roles: order-issuing users and order-receiving users, then respectively obtain the order-issuing requirement sets of target order-issuing users, and at the same time obtain the order-receiving service sets of all order-receiving users in a non-order-receiving state; the order-issuing requirement set includes a hard requirement subset and a flexible requirement subset. The hard requirement subset refers to the set composed of all hard requirements that the target order-issuing user does not accept changes, and the flexible requirement subset refers to the set composed of all flexible requirements that the target order-issuing user can accept changes;

[0048] The order-receiving service set includes a hard service subset and a flexible service subset. The basic service subset refers to the set composed of all hard services that the order-receiving user does not accept changes, and the flexible service subset refers to the set composed of all flexible services that the order-receiving user can accept changes.

[0049] Suppose there is a logistics platform, order-issuing user A needs to transport a batch of goods, and there are three order-receiving users B, C, and D in a non-order-receiving state on the platform. The following steps are used to elaborate in detail:

[0050] User role division: Order-issuing user A: Initiate a transportation request. Order-receiving users B, C, and D: Provide transportation services.

[0051] Obtain the order-issuing requirement set and the order-receiving service set: Requirement set of order-issuing user A: Hard requirement subset: Transportation method: Air transportation; Departure place: Beijing; Destination: Shanghai; Flexible requirement subset: Transportation time: Within 2 days, Price: Within 2000 yuan.

[0052] Service set of order-receiving user: Order-receiving user B: Hard service subset: Air transportation, Beijing to Shanghai; Flexible service subset: Within 3 days, 1900 yuan.

[0053] Order-receiving user C: Hard service subset: Air transportation, Beijing to Shanghai; Flexible service subset: Within 2 days, 1800 yuan.

[0054] Order-receiving user D: Hard service subset: Sea transportation, Tianjin to Shanghai; Flexible service subset: Within 5 days, 1500 yuan.

[0055] Based on the order-issuing requirement set and the order-receiving service set, perform hard condition screening, retain the order-receiving users who meet the hard conditions, and then perform flexible condition screening. If there are order-receiving users who meet all the flexible conditions, obtain the target order-receiving users according to the preset satisfaction strategy and match them with the target order-issuing users; performing hard condition screening means that the order-receiving service set meets all the non-changeable hard requirements in the hard requirement subset, and compare the hard requirement subset of the order-issuing user A with the hard service subsets of each order-receiving user.

[0056] Order-receiving user B: Air transportation (complied), Beijing to Shanghai (complied), passed the hard condition screening. Order-receiving user C: Air transportation (complied), Beijing to Shanghai (complied), passed the hard condition screening. Order-receiving user D: Sea transportation (not complied), Tianjin to Shanghai (not complied), did not pass the hard condition screening. Screening result: Order-receiving users B and C passed the hard condition screening, and order-receiving user D was excluded.

[0057] Perform flexible condition screening on order-receiving users B and C who passed the hard condition screening: Performing flexible condition screening means that: obtain the flexible requirement subset in the order-issuing requirement set, and compare each flexible requirement in the flexible requirement subset with the order-receiving service sets of the retained order-receiving users one by one. If all the flexible requirements are met, mark it as an accepted matching object. If not all the flexible requirements are met, mark it as a rejected matching object.

[0058] Order-receiving user B: Within 3 days (not fully complied), 1900 yuan (complied), met all the flexible conditions, and was marked as an accepted matching object. Order-receiving user C: Within 2 days (complied), 1800 yuan (complied), fully met all the flexible conditions, and was marked as an accepted matching object. Screening result: Both order-receiving users B and C fully met the flexible conditions of order-issuing user A. Therefore, according to the preset strategy, obtain the historical service scores of all order-receiving users marked as accepted matching objects, and then sort them in descending order. The order-receiving user ranked first is the target order-receiving user. The historical service score of order-receiving user B is 97.65, and the historical service score of order-receiving user C is 95.32. Then order-receiving user B ranks first, and order-receiving user B is successfully matched with order-issuing user A.

[0059] If there are no order-receiving users who meet all the flexible conditions, obtain the historical order-issuing records of the target order-issuing user and perform portrait analysis using the K-means clustering algorithm, and evaluate the matching failure rate based on the results of the portrait analysis. Obtain the flexible requirement subset in the current order-issuing requirement set of the target order-issuing user, and the historical order-issuing records of the target order-issuing user, and extract information according to the flexible requirement items in the current flexible requirement subset:

[0060] Mark the elasticity requirement items in the current subset of elasticity requirements as TXi respectively, set the target elasticity requirement item, obtain the transaction data at the time of transaction and the submission data at the time of issuance of all successfully matched orders in the historical order issuance records of the target order-issuing user, and then calculate the modification amplitude value corresponding to the target elasticity requirement item for each successfully matched order. The acquisition logic of the modification amplitude value is as follows:

[0061] First, calculate the difference between the transaction data at the time of transaction of the target elasticity requirement item minus the submission data at the time of issuance, and then divide this difference by the submission data at the time of issuance. The ratio obtained is then processed by taking the absolute value to obtain the modification change ratio value;

[0062] Then, calculate the average value of the modification change ratio values corresponding to the target elasticity requirement items of all successfully matched orders to obtain the modification amplitude value, which is marked as Mi;

[0063] Use the modification amplitude value Mi of each elasticity requirement item TXi as the input of the clustering feature, use the K-means clustering algorithm to perform clustering analysis on the modification amplitude values of all elasticity requirement items, use the elbow method to determine the optimal number of clusters K, and the elasticity requirement items in each cluster form a level. Corresponding each elasticity requirement item to its divided classification level to obtain the user portrait of the target order-issuing user.

[0064] The specific steps of the clustering analysis are as follows:

[0065] Collect the modification amplitude values: Extract the modification amplitude value Mi of each elasticity requirement item TXi. These values will be used as the input features of the K-means algorithm. For example, if there are 10 different elasticity requirement items TX1, TX2,..., TX10, and the modification amplitude values M1, M2,..., M10 of each item have been calculated, these values will form a feature vector for clustering.

[0066] K-means is a commonly used unsupervised clustering algorithm. By dividing the data into K clusters, each cluster is represented by a center (centroid). The algorithm performs clustering by minimizing the squared distance between the data points and their corresponding cluster centers. Use the elbow method: Determine the optimal number of clusters K through the elbow method. The specific steps are as follows:

[0067] Select multiple different K values (such as from 1 to 10), perform K-means clustering for each K value, calculate the within-cluster sum of squared errors, draw the relationship diagram between the within-cluster sum of squared errors and the K value, find the "elbow" position, that is, the point where the curve begins to flatten significantly, and use the corresponding K value as the optimal number of clusters.

[0068] Initialize cluster centers: Randomly select K initial centroids. The initial centroids can be randomly selected from the data points. Assign data points: Assign the modification magnitude value Mi of each flexible requirement item to the cluster center (centroid) closest to it, calculated as the Euclidean distance. Update cluster centers: Calculate the average value of all data points in each cluster and use this average as the new cluster center. Repeat iteration: Continue the process of assigning data points and updating cluster centers until the cluster centers no longer change significantly (i.e., the cluster assignments and centroid positions are stable). After the algorithm converges, all flexible requirement items TXi will be assigned to K different clusters, and each cluster represents items with similar modification magnitudes.

[0069] Level division: Analyze according to the center value (centroid) of each cluster. Usually, clusters with smaller centroids may represent requirement items with smaller modification magnitudes (i.e., higher-priority items), while clusters with larger centroids represent requirement items with larger modification magnitudes (lower-priority items).

[0070] Sort the clusters according to the centroid size and divide them into different levels. For example: Level 1 (high priority): The cluster with the smallest modification magnitude. Level 2 (medium priority): The cluster with medium modification magnitude. Level 3 (low priority): The cluster with the largest modification magnitude.

[0071] Obtain the user profile of the target order-issuing user, and then set weight coefficients for the flexible requirement items in this subset of flexible requirements, satisfying: ; The level of the flexible requirement item is represented as , represents the level of the flexible requirement item The corresponding weight coefficient, K represents the total number of levels; Through cluster analysis, the flexible requirement items of the target order-issuing user can be grouped according to their historical modification magnitudes. The items in each group have similar modification magnitude characteristics. Optimize the matching process: The clustered flexible requirement items can better reflect the actual flexibility of the order-issuing user for these requirements. By dividing these items into different priorities, resources can be more effectively allocated and strategies adjusted during the matching process. Priority assignment: Through the levels divided after clustering, different weight coefficients can be set for each level. Different weight coefficients indicate that the order-issuing user has different degrees of acceptance of changes for different flexible requirements during the process of achieving a match.

[0072] Then use the Hamming distance to represent the matching result of each flexible requirement item:

[0073] ; It represents the Hamming distance corresponding to the elastic requirement item. The difference amplitude value is obtained by first subtracting the data corresponding to the elastic requirement item in the order-receiving user's order-receiving service set from the corresponding requirement data in the elastic requirement subset, then dividing the obtained difference by the requirement data in the elastic requirement subset, and finally taking the absolute value of the obtained ratio. Calculated from the historical data above, so as to distinguish whether the order-issuing user is likely to accept. For example, the elastic requirement items of the order-issuing user include: the delivery time is acceptable within 3 days, and the service item provided by the order-receiving user is the delivery time within 4 days. The calculation is 0.33, and the one calculated from the historical data is 0.5. Then there is a high probability that this elastic requirement will be accepted. Therefore, the Hamming distance calculation result is zero.

[0074] Calculate the overall difference value D: ; n represents the total number of elastic requirement items; the overall difference value D is a comprehensive index calculated based on the difference result di of each elastic requirement item and the corresponding weight wi. D represents the sum of the weighted difference values between all elastic requirement items, reflecting the overall mismatch degree between the requirements of the order-issuing user and the services of the order-receiving user. The larger the value, the greater the difference between the two users and the higher the mismatch degree.

[0075] Then substitute it into the pre-trained comprehensive matching degree model formula:

[0076] ; is a parameter for adjusting the sensitivity, is the offset, used to adjust the starting point of the matching degree, is the comprehensive matching degree; the comprehensive matching degree S is a normalized value, usually between 0 and 1, used to quantify the matching degree between the order-issuing user and the order-receiving user. The closer the S value is to 1, the higher the probability of successful matching; the closer it is to 0, the lower the probability of successful matching. Reflecting the matching quality: Through the calculation of S, the quality of the matching can be quantitatively represented, providing an intuitive matching score for the user, thereby helping the user make a decision. The parameter for adjusting the sensitivity of the matching degree controls the influence degree of D on S, The larger the value, the more significant the influence of D on S, and the more sensitive the matching degree. is the offset, used to adjust the starting point of the matching degree calculation, affecting the baseline level of S, adjusting the starting position of the matching degree, ensuring that the comprehensive matching degree model reaches the expected accuracy during training. The matching failure rate is obtained by subtracting the comprehensive matching degree from the value 1. It is obtained that the matching failure rate is used to quantify the risk of matching failure. If the failure rate is close to 1, it indicates that the success probability of matching is very low; Decision support: By calculating the matching failure rate, suggestions can be provided to users during the matching process to help them evaluate the feasibility of the current matching and whether negotiation or readjustment of matching conditions is required in the future.

[0077] Based on the evaluation result of the matching failure rate, negotiation communication channels are successively established for the target order - placing users until an agreement is reached. At this time, the order - receiving user corresponding to the negotiation communication channel for this time matches with the target order - placing user. "Successively" means sorting based on the matching failure rate, and each time the order - receiving user with the highest success rate is selected to establish a negotiation communication channel between the target order - placing user and this order - receiving user. This can be achieved through the chat function, phone, email, or other real - time communication tools within the platform. The two parties can discuss possible adjustments through the negotiation communication channel. For example, the order - placing user may be willing to relax the time requirement, while the order - receiving user may be willing to make a certain concession in price. Moreover, during the negotiation process, the matching conditions can be updated in real - time, and the matching failure rate can be recalculated to help both parties understand the effect of the current adjustment. If no agreement is reached in this negotiation, continue to select the next order - receiving user for negotiation until an agreement is reached. At this time, the order - receiving user corresponding to the negotiation communication channel for this time matches with the target order - placing user.

[0078] By sorting based on the matching failure rate, each time the order - receiving user with the highest matching success rate is selected to negotiate with the order - placing user. This "successive" process means that the system will give priority to processing the most likely successful matches, gradually eliminating the possibility of failure until the best matching solution is found. Reducing ineffective communication: By preferentially selecting order - receiving users with higher matching success rates for negotiation, unnecessary communication and negotiation time can be reduced, improving efficiency. During the negotiation process, both parties can adjust their respective requirements and service conditions, and can also update the matching conditions in real - time and recalculate the matching failure rate. Users can clearly understand the effect of each adjustment, which makes it easier for both parties to reach an agreement. If an agreement is not reached in one negotiation, the system can quickly select the next optimal order - receiving user for negotiation until the match is successful. This multi - round negotiation mechanism ensures that even if the initial match fails, a most suitable matching solution can be found through continuous adjustment. By preferentially selecting the order - receiving user with the highest success rate for negotiation, the trial - and - error cost of blind matching is reduced. Users can find a suitable matching partner faster and avoid unnecessary waste of time and resources.

[0079] Embodiment 2: A multi - region transportation service mode matching system, including:

[0080] An identity information collection module that divides all user identities, sets two roles: order - placing users and order - receiving users, and then respectively obtains the order - placing requirement set of the target order - placing user, and at the same time obtains the order - receiving service sets of all order - receiving users in the non - order - receiving state;

[0081] A screening module, based on the order-issuing requirement set and the order-receiving service set, performs hard condition screening, retains the order-receiving users who meet the hard conditions, and then performs flexible condition screening;

[0082] A negotiation module, if there are no order-receiving users who meet all the flexible conditions, obtains the historical order-issuing records of the target order-issuing user and performs portrait analysis using the K-means clustering algorithm, and based on the results of the portrait analysis, evaluates the matching failure rate. Based on the matching failure rate evaluation results, successively establish a negotiation communication channel for the target order-issuing user;

[0083] A matching module, if there are order-receiving users who meet all the flexible conditions, obtains the target order-receiving user according to the preset satisfaction strategy and reaches a match with the target order-issuing user, and when the negotiation is successful, the order-receiving user corresponding to the negotiation communication channel in this negotiation reaches a match with the target order-issuing user;

[0084] A service feedback module, after the service is completed, the target order-issuing user gives a service score to the order-receiving user who has reached a match in this time.

[0085] The multi-region can refer to multiple regions geographically, multiple regions in the market environment, and the diversity of user groups. The order-receiving users include, but are not limited to, the following groups: taxis (or other urban transportation tools, such as tricycles, motorcycles, vans, buses, etc.); food delivery personnel (regardless of the transportation tools they use); truck drivers, logistics companies, air freight companies, shipping companies, etc., who provide land, water and air transportation services, or companies.

[0086] The order-issuing users include, but are not limited to, the following groups: passengers taking taxis (or other urban transportation tools, such as tricycles, motorcycles, vans, buses, etc.); merchants or consumers using food delivery services; people or companies using land, water and air transportation services.

[0087] The above formulas are all dimensionless and take their numerical values for calculation. The formula is a formula obtained by software simulation with a large amount of data to approximate the real situation as much as possible. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0088] It should be understood that in various embodiments of the present application, the magnitudes of the serial numbers of the above processes do not mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0089] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of this application.

[0090] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0091] As described above, the above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed by this application can easily think of changes or substitutions, and all should be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

1. A multi-regional transportation service mode matching method, characterized in that: The following steps are involved: All user identities are divided into two roles: order-issuing users and order-receiving users. Then, the order-issuing requirement sets of the target order-issuing users are obtained respectively, and the order-receiving service sets of all order-receiving users who are not in the order-receiving state are obtained at the same time. Based on the order placing requirement set and order receiving service set, hard condition screening is performed to retain order receiving users that meet the hard conditions. Then, flexible condition screening is performed. If there is an order receiving user that meets all flexible conditions, the target order receiving user is obtained according to the preset satisfaction strategy and matched with the target order placing user. If there is no order-taking user that meets all elastic conditions, the historical order-issuing records of the target order-issuing user are obtained and the K-means clustering algorithm is used for portrait analysis, and the matching failure rate is evaluated based on the results of the portrait analysis; Based on the matching failure rate evaluation result, negotiation communication channels are successively established for the target order-issuing user until a consensus is reached, and the order-receiving user corresponding to the negotiation communication channel is matched with the target order-issuing user; Obtaining the historical order records of the target order-issuing user and using the K-means clustering algorithm for portrait analysis means: Get the elastic requirement subset of the current order requirement set of the target order user, as well as the historical order records of the target order user, and extract information according to the elastic requirement items in the current elastic requirement subset: Mark the elastic requirement items in this elastic requirement subset as TXi, set the target elastic requirement item, obtain the transaction data of all successfully matched orders in the historical order records of the target order user, and the submission data when they were issued, and then calculate the modification amplitude value corresponding to the target elastic requirement item of each successfully matched order. The logic for obtaining the modification amplitude value is: First, calculate the difference between the transaction data at the time of transaction of the target elasticity requirement project and the submitted data at the time of issuance, then divide the difference by the submitted data at the time of issuance, and then perform absolute value processing on the obtained ratio to obtain the modified change ratio value; Then, the modification change ratio values ​​corresponding to the target elasticity requirement items of all successfully matched orders are averaged to obtain the modification amplitude value and marked as Mi; The modification amplitude value Mi of each elastic requirement item TXi is used as the feature input of clustering. The modification amplitude values ​​of all elastic requirement items are clustered using the K-means clustering algorithm. The optimal number of clusters K is determined using the elbow method. The elastic requirement items in each cluster form a level. Each elastic requirement item is matched with its classified level to obtain the user portrait of the target order-issuing user. Evaluating the matching failure rate based on the results of the profiling analysis refers to: Get the user profile of the target order-issuing user, and then set the weight coefficient for the elastic requirement items in this elastic requirement subset to meet the following requirements: ; The level of the elasticity requirement item is expressed as , Indicates the level of flexibility required for the project The corresponding weight coefficient, K represents the total number of levels; Then use the Hamming distance to represent the matching results of each elastic requirement item: ; It represents the Hamming distance corresponding to the elastic requirement item. The difference amplitude value is obtained by first subtracting the data corresponding to the elastic requirement item in the order service set of the order-taking user from the corresponding requirement data in the elastic requirement subset, and then dividing the obtained difference by the requirement data in the elastic requirement subset. The obtained ratio is then processed into the absolute value to obtain the difference amplitude value; Calculate the overall difference value D: ; n represents the total number of elasticity requirement items; Then substitute it into the pre-trained comprehensive matching model formula: ; To adjust the sensitivity parameter, is the offset used to adjust the starting point of the matching degree. is the comprehensive matching degree; The matching failure rate is calculated by subtracting the comprehensive matching degree from the value one. get.

2. A multi-regional transportation service mode matching method according to claim 1, characterized in that: The order requirement set includes a hard requirement subset and a flexible requirement subset. The hard requirement subset refers to the set of all hard requirements of the target order user that cannot be changed, and the flexible requirement subset refers to the set of all flexible requirements of the target order user that can be changed. The order-taking service set includes a hard service subset and a flexible service subset. The basic service subset refers to the set of all hard services of the order-taking user that cannot be changed, and the flexible service subset refers to the set of all flexible services of the order-taking user that can be changed.

3. A multi-regional transportation service mode matching method according to claim 2, characterized in that: Hard condition screening means that the order-accepting service set satisfies all the hard requirements in the hard requirement subset that cannot be changed.

4. A multi-regional transportation service mode matching method according to claim 2, characterized in that: Flexible condition screening means: Obtain a subset of elastic requirements in the order requirement set, and compare each elastic requirement in the elastic requirement subset with the order service set of the retained order-taking user one by one. If all elastic requirements are met, it is marked as an accepted matching object; if all elastic requirements are not met, it is marked as a rejected matching object.

5. A multi-regional transportation service mode matching method according to claim 4, characterized in that: The target order-taking users obtained according to the preset satisfaction strategy are: The historical service scores of all order-taking users marked as accepting matching objects are obtained, and then sorted in descending order. The order-taking user ranked first is the target order-taking user.

6. A multi-regional transportation service mode matching system, based on a multi-regional transportation service mode matching method according to any one of claims 1 to 5, characterized in that: include: The identity information collection module divides all user identities and sets two roles: order-issuing users and order-receiving users. It then obtains the order-issuing requirement set of the target order-issuing users and the order-receiving service set of all order-receiving users who are not in the order-receiving state. The screening module performs hard condition screening based on the order placing requirement set and order receiving service set, retains the order receiving users who meet the hard conditions, and then performs flexible condition screening; The negotiation module, if there is no order-taking user that meets all elastic conditions, obtains the historical order-issuing records of the target order-issuing user and uses the K-means clustering algorithm to perform a portrait analysis, and evaluates the matching failure rate based on the results of the portrait analysis. Based on the matching failure rate evaluation results, negotiation communication channels are successively established for the target order-issuing users; The matching module obtains the target order-taking user and matches it with the target order-issuing user according to the preset satisfaction strategy if there is an order-taking user who meets all the elastic conditions, and when the negotiation is in agreement, the order-taking user corresponding to the negotiation communication channel is matched with the target order-issuing user; Service feedback module: after the service is completed, the target order-issuing user will give a service rating to the order-receiving user who has been matched with the service.

Images