Quotation prediction method and device, electronic equipment and storage medium

Abstract

The invention provides a quotation prediction method and device, electronic equipment and a storage medium. The method comprises the steps that when a to-be-transported goods source does not make a transaction after a first time period after being put on a shelf, real-time statistical information is acquired, the real-time statistical information comprises exposure statistical data, route statistical data and quotation information, the exposure statistical data is used for indicating the exposure condition of the to-be-transported goods source at the current moment, and the route statistical data is used for indicating the quotation information of the to-be-transported goods source; the route statistical data is used for indicating a quantity statistical value of a first goods source, the first goods source is a goods source which is transported through a demand transportation route of a to-be-transported goods source at the current moment, and the quotation information comprises a driver quotation and an owner quotation of the to-be-transported goods source at the current moment; according to the real-time statistical information, dynamic features are determined, and the dynamic features are used for representing features influencing the transaction probability of the to-be-transported goods source at the current moment; and according to the dynamic characteristics, a predicted quotation is determined, and the predicted quotation is used for guiding the owner to modify the owner quotation, so that the transaction probability and the transaction speed of the to-be-transported goods source are improved.

Description

Technical Field

[0001] The present application relates to the field of logistics technology, and in particular to a quotation prediction method, device, electronic device and storage medium. Background Art

[0002] With the development of online freight, various transportation service platforms have emerged. When cargo owners publish cargo sources through transportation service platforms, drivers choose whether to accept orders on the transportation service platforms and then transport the cargo sources.

[0003] In related technologies, for a batch of cargo, the transportation service platform can obtain a forecast quotation based on cargo attributes such as cargo weight and cargo type, as well as the route between the cargo's origin and destination, and recommend the forecast quotation to the cargo owner. The cargo owner uses the forecast quotation as a reference to set the actual quotation. The driver determines whether to accept the order based on the actual quotation.

[0004] However, the predicted quotes obtained by the transportation service platform may be inaccurate, resulting in unreasonable actual quotes set by cargo owners based on the predicted quotes. When the actual quotes are low, drivers may not accept the orders, resulting in a lower transaction probability and transaction speed for the batch of goods. Summary of the invention

[0005] The present application provides a quotation prediction method, device, electronic device and storage medium to solve the problem that the predicted quotation is inaccurate, resulting in a low transaction speed and transaction probability of the source of goods, thereby achieving the purpose of improving the transaction speed and transaction probability of the source of goods.

[0006] In a first aspect, the present application provides a quotation prediction method, comprising:

[0007] When a transaction has not been completed after a first period of time has passed since the source of goods to be transported was put on the shelf, real-time statistical information is obtained, the real-time statistical information including exposure statistical data, route statistical data and quotation information, the exposure statistical data is used to indicate the exposure status of the source of goods to be transported at the current moment, the route statistical data is used to indicate the quantity statistical value of the first source of goods, the first source of goods is the source of goods transported through the demand transportation route of the source of goods to be transported at the current moment, and the quotation information includes the driver quotation and the cargo owner quotation of the source of goods to be transported at the current moment;

[0008] Determine dynamic features according to the real-time statistical information, wherein the dynamic features are used to characterize features that affect the probability of a transaction of the source of goods to be transported at the current moment;

[0009] A predicted quotation is determined according to the dynamic characteristics, and the predicted quotation is used to guide the consignor to modify the consignor's quotation.

[0010] Through the quotation prediction method provided in the first aspect, when the goods to be transported are put on the shelves and the transaction has not been completed after the first period of time, real-time statistical information is obtained, and the real-time statistical information includes exposure statistical data, route statistical data and quotation information, so that the quotation prediction can be started in time, and reasonable predicted quotations can be provided to the cargo owner, which is helpful to promote the transaction of the goods to be transported as soon as possible, and the information that directly or indirectly affects the transaction probability and transaction speed of the goods to be transported can be obtained from the real-time statistical information. According to the exposure statistical data, route statistical data and quotation information in the real-time statistical information, the dynamic characteristics are determined, and the latest dynamic characteristics at the current moment can be obtained, so that the latest factors related to the goods to be transported and affecting the transaction probability of the goods to be transported can be considered when predicting the quotation, which helps to improve the accuracy and real-time nature of the quotation prediction. The transportation service platform determines the predicted quotation based on the dynamic characteristics. When the transportation service platform predicts the quotation, it introduces dynamic characteristics that change over time, and can consider the degree of influence of each dynamic characteristic on the transaction probability of the goods to be transported, so as to obtain a quotation with higher real-time and accuracy, thereby improving the transaction probability and transaction speed of the goods to be transported.

[0011] In one possible design, the method further includes:

[0012] Sending forecast information to the consignor, the forecast information including the forecast quotation, or the forecast information including the difference between the forecast quotation and the consignor's quotation;

[0013] After the second period, continue to obtain real-time statistical information and determine new forecast quotes.

[0014] In one possible design, the method further includes:

[0015] Sending forecast information to the consignor, the forecast information including the forecast quotation, or the forecast information including the difference between the forecast quotation and the consignor's quotation;

[0016] After the cargo owner modifies the cargo owner's quotation according to the forecast information, after a third period of time, it is determined whether the cargo to be transported is sold;

[0017] When the source of goods to be transported has not been sold, real-time statistical information continues to be obtained and a new forecast quotation is determined.

[0018] In a possible design, determining the dynamic feature according to the real-time statistical information includes:

[0019] Determine a real-time transaction rate and a real-time supply-demand ratio according to the route statistical data, wherein the real-time transaction rate and the real-time supply-demand ratio are used to characterize the transaction status of the first source of goods;

[0020] Determine a real-time click rate and a real-time feedback rate according to the exposure statistics, wherein the real-time click rate and the real-time feedback rate are used to characterize the degree of interest of the driver in the cargo to be transported;

[0021] Determining the dynamic features includes the real-time transaction rate, the real-time supply-demand ratio, the first duration, the cargo owner's quotation, the driver's quotation, the real-time click rate and the real-time feedback rate.

[0022] In a possible design, the route statistical data include the total quantity of the first source of goods, the number of transactions and the number of feedbacks, the real-time transaction rate is the ratio of the number of transactions to the total quantity, the real-time supply-demand ratio is the ratio of the number of feedbacks to the total quantity, the exposure statistical data include the first duration, the exposure of the source of goods to be transported, the number of driver clicks and the number of driver feedbacks, the real-time click rate is the ratio of the driver clicks to the exposure, and the real-time feedback rate is the ratio of the driver feedback to the driver clicks.

[0023] In a possible design, determining the predicted quotation according to the dynamic feature includes:

[0024] Obtain static features, including the weight, volume, category, length of the required vehicle, model of the required vehicle, departure information, destination information, historical transaction rate of the required transportation route, historical supply-demand ratio, historical transaction price statistics, historical kilometer price statistics, and historical deposit ratio statistics of the required transportation route;

[0025] The predicted quotation is determined according to the static feature and the dynamic feature.

[0026] In a possible design, determining the predicted price quotation according to the static feature and the dynamic feature includes:

[0027] The static features and the dynamic features are input into a quotation prediction model to obtain the predicted quotation. The quotation prediction model is obtained by training a basic model based on the transaction price of a sample source of goods, static sample features, and dynamic sample features.

[0028] In a possible design, the basic model is a LightGBM model, the loss function of the basic model is a mean absolute error function, and the evaluation index of the basic model is Formula 1;

[0029] Among them, formula 1 is:

[0030] Among them, P1 is the predicted price of the sample source of goods, P2 is the transaction price of the sample source of goods, k is a preset threshold, and k is greater than or equal to 0.

[0031] In a second aspect, the present application provides a quotation prediction device, comprising: a module for executing the method in the first aspect and any possible involvement in the first aspect.

[0032] In a third aspect, the present application provides an electronic device, including: a processor;

[0033] The processor is used to execute the computer executable program or instructions in the memory, so that the electronic device executes the method in the first aspect and any possible design of the first aspect.

[0034] In a fourth aspect, the present application provides an electronic device comprising: a memory and a processor; the memory is used to store program instructions; the processor is used to call the program instructions in the memory so that the electronic device executes the method in the first aspect and any possible design of the first aspect.

[0035] In a fifth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor in an electronic device, the computer program implements the method in the first aspect and any possible design of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A flowchart of a quotation prediction method provided in one embodiment of the present application.

[0037] Figure 2 A flowchart of a quotation prediction method provided in one embodiment of the present application.

[0038] Figure 3 A flowchart of a quotation prediction method provided in one embodiment of the present application.

[0039] Figure 4 A flow chart of a method for determining dynamic features provided in one embodiment of the present application.

[0040] Figure 5 A flow chart of a method for determining a predicted quotation provided in one embodiment of the present application.

[0041] Figure 6 A flow chart of a method for determining a predicted quotation provided in one embodiment of the present application.

[0042] Figure 7 A schematic diagram of the structure of a quotation prediction device provided in one embodiment of the present application.

[0043] Figure 8 A schematic diagram of the structure of an electronic device provided in one embodiment of the present application.

[0044] Fig. 9 A schematic diagram of the structure of an electronic device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0045] In this application, "at least one" means one or more, and "plurality" means two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c alone can represent: a, b, c, a and b, a and c, or a, b, and c, where a, b, and c can be single or multiple. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance.

[0046] The directions or positional relationships indicated by terms such as "center", "longitudinal", "lateral", "up", "down", "left", "right", "front", and "back" are based on the directions or positional relationships shown in the accompanying drawings and are only for the convenience of describing the present application and simplifying the description. They do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as a limitation on the present application.

[0047] The terms "connected" and "connected" should be understood in a broad sense. For example, the "connected" or "connected" of a circuit structure can refer to not only physical connection, but also electrical connection or signal connection. For example, it can be directly connected, that is, physically connected, or indirectly connected through at least one intermediate element, as long as the circuit is connected, or it can be the internal connection of two elements; signal connection can refer to signal connection through a circuit or through a media medium, such as radio waves. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0048] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0049] Exemplarily, the present application provides a quotation prediction method, device, electronic device, computer-readable storage medium and computer program product, which predicts the source of cargo to be transported through dynamic characteristics such as the shelf time of the current source of cargo to be transported on the transportation service platform, real-time click-through rate, real-time feedback rate, driver quotation, consignor quotation, real-time transaction rate on the transportation route and real-time supply-demand ratio, combined with static characteristics such as source attributes and historical transaction status, to obtain a predicted quotation with a higher accuracy. When the dynamic characteristics change over time, the predicted quotation is updated at any time, so that a more accurate predicted quotation can be recommended to the consignor in real time, which helps to improve the transaction probability and transaction speed of the source of cargo to be transported.

[0050] Among them, the quotation prediction method provided in the present application can be executed by an electronic device, or by a quotation prediction device in an electronic device.

[0051] Among them, the electronic device can be a server, a desktop computer, a mobile phone, a tablet computer, a laptop computer, a wearable device, a vehicle-mounted device, an augmented reality (AR) / virtual reality (VR) device, etc.

[0052] Among them, the quotation prediction device can be implemented by a combination of software and / or hardware. For example, the quotation prediction device can be an application (APP), a web page or a public account, etc. For example, the quotation prediction device can be a transportation service platform. In order to simplify the description, the embodiment of the present application is described by taking the execution of the transportation service platform as an example.

[0053] Next, combine Figure 1 , the quotation prediction method provided in the embodiment of the present application is described.

[0054] See also Figure 1 , Figure 1 This is a flowchart of a quotation prediction method provided in one embodiment of the present application. Figure 1 As shown, the method includes:

[0055] S101. When a first period of time has passed since the goods to be transported are put on the shelves but no transaction has been completed, real-time statistical information is obtained.

[0056] When a cargo owner has a demand for transportation of goods, the cargo owner can upload the cargo owner's quotation and attribute information of the goods to be transported through the transportation service platform. The transportation service platform generates a transportation order based on the cargo owner's quotation and attribute information, puts the goods to be transported on the shelves, and pushes the transportation order corresponding to the goods to be transported to the driver.

[0057] Drivers can see the transportation orders corresponding to the goods to be transported on the transportation service platform and decide whether to accept the order.

[0058] After the cargo to be transported is put on the shelf, the cargo owner can modify the cargo owner's quotation at any time. The driver can also make a driver quotation for the cargo to be transported and upload it to the transportation service platform. The cargo owner can view the driver quotation through the transportation service platform.

[0059] Among them, the attribute information of the cargo to be transported may include the weight, volume, category, length of the required vehicle, model of the required vehicle, departure information and destination information, etc.

[0060] Among them, the categories of goods to be transported may include agricultural products, industrial products, and daily chemical products.

[0061] The departure information is used to indicate the location of the departure of the cargo to be transported. For example, the departure information may be the location coordinates of the departure.

[0062] The destination information is used to indicate the destination location of the cargo to be transported. For example, the destination information may be the location coordinates of the destination.

[0063] The required vehicle is the vehicle that the cargo owner wants to transport the cargo to be transported. The vehicle model of the required vehicle can be a van, a box truck or a truck. Depending on the model, the required vehicle can have different lengths, such as a 5-meter truck, a 7-meter truck, etc. The cargo owner can provide the transportation service platform with multiple types and lengths of required vehicles at the same time to expand the range of drivers who can accept orders, thereby increasing the transaction probability and transaction speed of the cargo to be transported.

[0064] Among them, the cargo owner's quotation is the price filled in by the cargo owner himself.

[0065] The cargo owner can determine the cargo owner's quotation according to his budget.

[0066] When the transportation service platform receives the attribute information of the goods to be transported, it can also estimate a reference price or reference price range based on the attribute information. The cargo owner can determine the cargo owner's quotation based on the reference price or reference price range and provide it to the transportation service platform. Based on this, the cargo owner's quotation is more accurate, which can increase the transaction probability and transaction speed of the goods to be transported.

[0067] After receiving the attribute information of the cargo to be transported and the cargo owner's quotation, the transportation service platform can store the attribute information and the cargo owner's quotation.

[0068] Based on this, the transportation service platform generates a transportation order corresponding to the cargo to be transported according to the attribute information of the cargo to be transported and the cargo owner's quotation, puts the cargo to be transported on the shelf, and pushes the transportation order to the driver, waiting for the driver to accept the order.

[0069] The transportation service platform will monitor the shelf time of the goods to be transported, that is, the time it takes for the transportation order to be pushed to the driver.

[0070] After the first period of time, if no driver has accepted the order, the transportation service platform can determine that the cargo to be transported has not been sold. At this time, the cargo owner needs to adjust the quotation to encourage the driver to accept the order as soon as possible.

[0071] The first duration may be set to 1 minute, 5 minutes, or 10 minutes, etc. The transportation platform may determine the first duration according to the needs of the cargo owner.

[0072] Based on this, the transportation service platform can start to make quotation forecasts and provide shippers with reasonable forecast quotations to facilitate the transaction of goods to be transported as soon as possible.

[0073] After the goods to be transported are put on the shelves, in addition to the attribute information of the goods to be transported and the quotations of the cargo owners, there are also many factors that change with time, which can directly or indirectly affect the transaction probability and transaction speed of the goods to be transported.

[0074] Considering that when making quotation forecasts, the transportation service platform needs to obtain a quotation that makes the transaction probability and transaction speed of the goods to be transported as high as possible. Based on this, the transportation service platform obtains real-time statistical information to determine the above-mentioned multiple factors that change over time from the real-time statistical information.

[0075] Among them, the real-time statistical information includes exposure statistics, route statistics and quotation information.

[0076] Among them, the exposure statistics are used to indicate the exposure status of the goods to be transported at the current moment.

[0077] After the goods to be transported are put on the shelves, the transportation service platform can record the shelf time of the goods to be transported in real time.

[0078] The transportation service platform can push the transportation orders corresponding to the goods to be transported to the drivers, and the number of drivers pushed is the initial exposure.

[0079] When drivers browse the transport orders corresponding to the cargo sources to be transported through the transport service platform, the transport service platform can record the number of drivers browsing the transport orders in real time to obtain the driver browsing volume.

[0080] The sum of the current driver views and the initial exposure is the current exposure.

[0081] When the driver clicks on the transport order corresponding to the cargo to be transported, the transport service platform can record the number of clicks by the driver, thereby obtaining the number of clicks by the driver on the cargo to be transported.

[0082] When the driver contacts the cargo owner through the transportation service platform, the transportation service platform can record the number of contacts made by the driver, thereby obtaining feedback from the driver regarding the cargo to be transported.

[0083] The above-mentioned shelf time, exposure, driver clicks and driver feedback can all reflect the exposure status of the goods to be transported at the current moment.

[0084] Based on this, the transportation service platform can determine exposure statistics based on shelf time, exposure, driver clicks, and driver feedback.

[0085] Exposure statistics will accumulate as the goods to be transported are kept on the shelves for a certain length of time, and the transportation service platform can update the exposure statistics in real time.

[0086] In some examples, exposure statistics include first duration, exposure amount, driver click amount, and driver feedback amount.

[0087] Among them, the first duration is the shelf time of the goods to be transported at the current moment.

[0088] Based on this, the transportation service platform obtains the exposure statistics of the goods to be transported at the current moment.

[0089] The route statistical data is used to indicate the quantity statistical value of the first source of goods.

[0090] The transportation service platform can monitor all sources of cargo on the platform in real time, count and update in real time the total quantity and transaction volume of cargo transported through different transportation routes, and the total quantity and transaction volume of cargo transported by different vehicle models and lengths.

[0091] The total quantity and transaction quantity of other sources of goods transported through the demand transportation routes of the goods to be transported will affect the transaction probability of the goods to be transported.

[0092] For example, if the total number of other cargo sources transported through the required transportation route is large, there may not be enough drivers available to transport the cargo sources to be transported, resulting in a lower probability of transaction for the cargo sources to be transported.

[0093] For example, if the ratio of the number of transactions to the total number of other sources of goods transported through the demand transportation route is low, then the driver may not be interested in the demand transportation route, and the probability of accepting orders for the goods to be transported is also low, resulting in a low probability of transactions for the goods to be transported.

[0094] In some examples, the route statistics include the total quantity of the first source, the number of transactions, and the number of feedbacks.

[0095] Among them, the transaction quantity of the first source of goods is the number of sources of goods sold in the first source of goods.

[0096] Among them, the feedback quantity of the first source of cargo is the number of sources of cargo that have been fed back by drivers in the first source of cargo.

[0097] Among them, the first source of goods is the source of goods that is currently being transported through the demand transportation route of the source of goods to be transported.

[0098] The required transportation route may include one or more routes.

[0099] When the required transportation route includes one route, the first source of goods includes the source of goods that is currently being transported through the route.

[0100] When the required transportation route includes multiple routes, the first source of goods includes all sources of goods that are currently being transported through the multiple routes.

[0101] Among them, there are various situations in which transportation routes are required.

[0102] In some examples, the required transportation route includes one or more candidate routes generated by the transportation service platform according to the departure information and the destination information.

[0103] For example, if the departure point is A and the destination is B, the transportation service platform can select a transportation route that includes A and B in the route library as a candidate route. The route library is a library pre-stored in the transportation service platform, and the route library includes multiple transportation routes.

[0104] In other examples, the required transportation route is a route selected by the shipper from among the candidate routes.

[0105] The transportation service platform can recommend alternative routes to the shipper, and the shipper can select the required route from the alternative routes as the required transportation route.

[0106] Based on this, the transportation service platform can obtain route statistics of the goods to be transported at the current moment.

[0107] The quotation information includes the driver's quotation and the cargo owner's quotation of the cargo to be transported at the current moment.

[0108] The transportation service platform can record in real time the driver's quotes and the cargo owner's quotes for the cargo to be transported.

[0109] Based on this, the transportation service platform can obtain quotation information for the goods to be transported at the current moment.

[0110] In summary, the transportation service platform obtains real-time statistical information, which includes exposure statistics, route statistics and quotation information.

[0111] S102: Determine dynamic features based on real-time statistical information.

[0112] Among them, the dynamic features are used to characterize the features that affect the transaction probability of the goods to be transported at the current moment.

[0113] Specifically, after obtaining the real-time statistical information, the transportation service platform can determine various features in the real-time statistical information that directly or indirectly affect the transaction probability of the goods to be transported as dynamic features.

[0114] Among them, exposure statistics can reflect the degree of interest of drivers in transporting cargo, and the degree of interest of drivers in transporting cargo affects the transaction probability of the cargo to be transported.

[0115] For example, the first duration in the exposure statistics, the longer the first duration, the lower the driver's interest in transporting goods. The lower the driver's interest in transporting goods, the lower the probability of transaction of the transported goods.

[0116] Among them, route statistics can reflect the situation of the cargo transported through the required transportation route. The situation of the cargo transported through the required transportation route can reflect the driver's interest in the required transportation route. The driver's interest in the required transportation route affects the transaction probability of the cargo to be transported.

[0117] For example, the number of transactions for the first source of goods in the route statistics. The more transactions, the higher the driver's interest in the required transportation route. The higher the driver's interest in the required transportation route, the higher the probability of a transaction for the source of goods to be transported.

[0118] Among them, the quotation information can directly reflect the transaction probability of the goods to be transported.

[0119] For example, the driver's quotation and the cargo owner's quotation in the quotation information. The closer the driver's quotation is to the cargo owner's quotation, the higher the probability of the transaction of the cargo to be transported.

[0120] Based on this, the transportation service platform can determine the dynamic features according to the exposure statistics, route statistics and quotation information in the real-time statistical information, and can obtain the latest dynamic features at the current moment, so that it can take into account the latest factors related to the goods to be transported and affecting the transaction probability of the goods to be transported when predicting quotations, which helps to improve the accuracy and real-time nature of quotation predictions.

[0121] Based on this, the transportation service platform determines the dynamic characteristics so as to make quotation predictions based on the dynamic characteristics.

[0122] S103: Determine a predicted quotation based on the dynamic characteristics.

[0123] Since the dynamic characteristics jointly affect the transaction probability of the goods to be transported, when determining the predicted quotation, for each dynamic characteristic, the transportation service platform needs to consider the impact of the dynamic characteristic on the transaction probability of the goods to be transported when making a quotation prediction, so as to obtain a more accurate quotation when predicting the quotation.

[0124] Among them, the transportation service platform can determine the predicted quotation through a variety of methods.

[0125] In some examples, the transportation service platform inputs the dynamic features into a preset model to obtain a predicted quote.

[0126] Among them, the preset model is obtained by training the basic model based on the dynamic characteristics of historical transaction sources and transaction prices.

[0127] The basic model may be a decision tree model, such as a random forest model, a gradient boosting decision tree (GBDT) model, an XGBoost model, or a LightGBM model.

[0128] The decision tree model does not require data assumptions, that is, it does not need to make any assumptions about the structure and distribution of dynamic features. It can handle various types of data and does not require excessive preprocessing or conversion of dynamic features when making quotation predictions. The decision tree model can capture the interaction between dynamic features. Since the quotation of the source of goods to be transported will be affected by the dynamic features, the decision tree model can effectively identify the interaction between dynamic features, thereby obtaining more accurate prediction results.

[0129] In other examples, the transportation service platform sets a weight factor for each dynamic feature in advance according to the degree of influence of the dynamic feature on the transaction probability of the transported goods. When predicting, the transportation service platform calculates the predicted quotation based on the feature value and weight factor of each dynamic feature and the current price of the cargo owner.

[0130] In an embodiment of the present application, the transportation service platform obtains real-time statistical information when the first period of time has passed since the goods to be transported were put on the shelves and the transaction has not been completed. The real-time statistical information includes exposure statistical data, route statistical data and quotation information. Thus, the transportation service platform can start to make quotation predictions in time, provide reasonable predicted quotations for the cargo owners, help to promote the transaction of the goods to be transported as soon as possible, and enable the transportation service platform to obtain information that directly or indirectly affects the transaction probability and transaction speed of the goods to be transported from the real-time statistical information. The transportation service platform determines the dynamic features based on the exposure statistical data, route statistical data and quotation information in the real-time statistical information, and can obtain the latest dynamic features at the current moment, so that the latest factors related to the goods to be transported and affecting the transaction probability of the goods to be transported can be considered when predicting the quotation, which helps to improve the accuracy and real-time nature of the quotation prediction. The transportation service platform determines the predicted quotation based on the dynamic features. When the transportation service platform makes quotation predictions, it introduces dynamic features that change over time, and can consider the degree of influence of each dynamic feature on the transaction probability of the goods to be transported, so as to obtain a quotation with higher real-time and accuracy, thereby improving the transaction probability and transaction speed of the goods to be transported.

[0131] Based on the above exemplary description, after determining the forecast quotation, the transportation service platform can update the forecast quotation in a variety of ways.

[0132] Next, combine Figure 2 and Figure 3 , the method for the transportation service platform to update the forecast quotation is explained.

[0133] See also Figure 2 , Figure 2 This is a flowchart of a quotation prediction method provided in one embodiment of the present application. Figure 2 As shown, after S103, the quotation prediction method further includes:

[0134] S104A: Send forecast information to the cargo owner.

[0135] Among them, the transportation service platform can send forecast information to shippers in a variety of ways.

[0136] In some examples, the forecast information includes forecast quotes.

[0137] The transportation service platform can directly push the predicted quotation to the consignee, so that the consignee can intuitively know how much the consignee's quotation needs to be modified.

[0138] For example, if the predicted price determined by the transportation service platform is 10,000 yuan, the forecast information "the predicted price is 10,000 yuan" can be pushed to the cargo owner.

[0139] In other examples, the forecast information includes the difference between the forecast price and the shipper's price.

[0140] The transportation service platform can first calculate the difference between the predicted quotation and the consignor's quotation, and push the difference to the consignor, so that the consignor can know the changes between the predicted quotation and the consignor's quotation, which helps the consignor to weigh the budget for the quotation and thus improve the user experience.

[0141] For example, the cargo owner's quoted price is 9,000 yuan, and the predicted price determined by the transportation service platform is 10,000 yuan. The transportation service platform determines that the difference between the predicted price and the cargo owner's quoted price is +1,000 yuan. Then the predicted information "additional price of +1,000 yuan" can be pushed to the cargo owner.

[0142] Of course, the transportation service platform can also send the predicted quotation and the difference between the predicted quotation and the consignor's quotation to the consignor at the same time, and this application does not impose any restrictions on this.

[0143] S105A: After the second period of time, continue to obtain real-time statistical information and determine a new forecast quotation until the transaction of the cargo to be transported is completed.

[0144] Taking into account that dynamic characteristics will change over time, the transportation service platform can monitor the transaction status of the goods to be transported in real time, and continue to execute S101 after the second period of time, re-acquire the real-time statistical information at the next moment, and determine a new predicted quotation, so that it can periodically perform quotation predictions before the transaction, and recommend the latest predicted quotation to the cargo owner, providing convenience for the cargo owner to modify the cargo owner's quotation, thereby further improving the transaction probability and transaction speed of the goods to be transported.

[0145] remove Figure 2 In addition to the methods shown, see Figure 3 , Figure 3 This is a flowchart of a quotation prediction method provided in one embodiment of the present application. Figure 3 As shown, after S103, the quotation prediction method further includes:

[0146] S104B. Send forecast information to the cargo owner.

[0147] For the description of S104B, please refer to S104A and will not be repeated here.

[0148] S105B. After the cargo owner modifies the cargo owner's quotation according to the forecast information, after a third period of time, it is determined whether the cargo to be transported has been sold.

[0149] When the supply of goods to be transported has not been concluded, the transport service platform executes S106B. When the supply of goods to be transported has been concluded, the transport service platform executes S107B.

[0150] After the cargo owner modifies the quotation based on the forecast information, the transportation service platform can wait for a third period of time to reserve transaction time between the driver and the cargo owner. After the third period of time, it can determine whether the transaction of the cargo to be transported has been completed, and thus decide whether to make the next forecast based on the status of the cargo to be transported.

[0151] When the cargo to be transported has not been sold, the transportation service platform determines that the next forecast needs to be made. When the cargo to be transported has been sold, the transportation service platform does not need to make the next forecast, thereby saving computing resources.

[0152] S106B, continue to obtain real-time statistical information, and determine a new forecast quotation.

[0153] Based on this, the transportation service platform can make quotation forecasts in a timely manner when the cargo to be transported has not been traded, and recommend the latest predicted quotation to the cargo owner, providing convenience for the cargo owner to modify the cargo owner's quotation, improving the user experience, and further increasing the transaction probability and transaction speed of the cargo to be transported.

[0154] S107B, end prediction.

[0155] Based on the above exemplary description, in S102, the transportation service platform can Figure 4 The method shown determines dynamic characteristics.

[0156] See also Figure 4 , Figure 4 A flow chart of a method for determining dynamic features provided in one embodiment of the present application. Figure 4 As shown, the method includes:

[0157] S201. Determine the real-time transaction rate and the real-time supply-demand ratio according to route statistical data.

[0158] Among them, the real-time transaction rate and real-time supply-demand ratio are used to characterize the transaction status of the first source of goods.

[0159] Among them, route statistics include the total quantity of the first source of goods, the number of transactions and the number of feedbacks.

[0160] Among them, the transportation service platform can determine the ratio of the transaction quantity of the first source of goods to the total quantity of the first source of goods as the real-time transaction rate.

[0161] For example, the total quantity of the first source of goods is 1000, and the transaction quantity of the first source of goods is 9000, then the real-time transaction rate is 90%.

[0162] Among them, the real-time transaction rate can reflect the proportion of the number of cargo sources sold on the demand transportation route to the total number of cargo sources, thereby reflecting the driver's willingness to accept orders for cargo sources on the demand transportation route. Then, when making quotation forecasts, the transportation service platform can adjust the forecast quotation based on the driver's willingness to accept orders for cargo sources on the demand transportation route.

[0163] For example, when the real-time transaction rate reflects that the driver has a low interest in the cargo on the required transportation route, the weight of the real-time transaction rate can be increased when determining the forecast quotation, so that the transportation service platform can obtain a more accurate forecast quotation.

[0164] Among them, the transportation service platform can determine the ratio of the feedback quantity of the first source of goods to the total quantity of the first source of goods as the real-time supply and demand ratio.

[0165] For example, the total quantity of the first source of goods is 1000, and the feedback quantity of the first source of goods is 9800, then the real-time supply-demand ratio is 98%.

[0166] The real-time supply-demand ratio can reflect the proportion of the number of sources of goods that have received feedback on the demand transportation route to the total number of sources of goods, thereby reflecting the degree of interest of drivers in the sources of goods on the demand transportation route. When making quotation forecasts, the transportation service platform can adjust the forecasted quotation based on the degree of interest of drivers in the sources of goods on the demand transportation route.

[0167] For example, when the real-time supply-demand ratio reflects that drivers have low interest in cargo on the demand transportation route, the weight of the real-time supply-demand ratio can be increased when determining the forecast quotation, so that the transportation service platform can obtain a more accurate forecast quotation.

[0168] S202: Determine the real-time click rate and the real-time feedback rate according to the exposure statistics.

[0169] Among them, the real-time click rate and real-time feedback rate are used to represent the driver's interest in the transportation source.

[0170] Among them, exposure statistics include the first duration, exposure of the cargo to be transported, driver clicks and driver feedback.

[0171] Among them, the transportation service platform can determine the real-time click-through rate as the ratio of the number of clicks by drivers of the cargo to be transported and the amount of exposure of the cargo to be transported.

[0172] For example, if the number of clicks by drivers of the cargo to be transported is 100 and the exposure of the cargo to be transported is 1000, the real-time click rate is 10%.

[0173] The real-time click rate can reflect the ratio of the number of drivers who click on the transport order corresponding to the source of goods to be transported to the number of drivers who receive the transport order push, thereby reflecting the driver's interest in the source of goods to be transported. When the transport service platform makes a quotation forecast, it can adjust the forecast quotation according to the driver's interest in the source of goods to be transported.

[0174] For example, when the real-time click rate reflects that the driver has a low interest in transporting cargo, the weight of the real-time click rate when determining the forecast quotation can be increased, so that the transportation service platform can obtain a more accurate forecast quotation.

[0175] Among them, the transportation service platform can determine the real-time feedback rate as the ratio of the feedback amount of drivers of the cargo to be transported to the number of clicks of drivers of the cargo to be transported.

[0176] For example, if the number of feedbacks from drivers of cargo to be transported is 100 and the number of clicks from drivers of cargo to be transported is 1000, then the real-time feedback rate is 10%.

[0177] The real-time feedback rate can reflect the ratio of the number of drivers who have provided feedback on the transport source to the number of drivers who have clicked on the transport order, thus reflecting the driver's interest in the transport source. When making quotation forecasts, the transport service platform can adjust the forecasted quotation based on the driver's interest in the transport source.

[0178] For example, when the real-time feedback rate reflects that the driver has a low level of interest in transporting cargo, the weight of the real-time feedback rate when determining the forecast quotation can be increased, so that the transportation service platform can obtain a more accurate forecast quotation.

[0179] Among them, when the transportation service platform executes S201 and S202, there is no order. The transportation service platform can execute S201 first, that is, first determine the real-time transaction rate and the real-time supply-demand ratio, and then execute S202, that is, then determine the real-time click rate and the real-time feedback rate. It can also execute S202 first, that is, first determine the real-time click rate and the real-time feedback rate, and then execute S202, that is, then determine the real-time transaction rate and the real-time supply-demand ratio. Of course, the transportation service platform can also execute S201 and S202 at the same time, and this application does not limit this.

[0180] S203, determining dynamic features including real-time transaction rate, real-time supply-demand ratio, first duration, cargo owner's quotation, driver's quotation, real-time click rate and real-time feedback rate.

[0181] The first duration is the shelf time of the cargo to be transported, which reflects the driver's interest in the cargo. The longer the shelf time, the lower the driver's interest in the cargo.

[0182] Among them, the shipper's quotation and the driver's quotation can provide a reference for the transportation service platform to determine the predicted quotation.

[0183] Based on this, the transportation service platform can determine dynamic features including real-time transaction rate, real-time supply-demand ratio, first duration, shipper quotation, driver quotation, real-time click rate and real-time feedback rate.

[0184] Based on the above exemplary description, in S103, the transportation service platform can Figure 5 The method shown determines the forecast quote.

[0185] See also Figure 5 , Figure 5 A flow chart of a method for determining a forecast quotation provided in an embodiment of the present application. Figure 5 As shown, the method includes:

[0186] S301, obtaining static features.

[0187] Among them, the static characteristics include the weight, volume, category of the goods to be transported, the length of the required vehicle, the model of the required vehicle, the departure information, the destination information, the historical transaction rate of the required transportation route, the historical supply-demand ratio, the historical transaction price statistics, the historical kilometer price statistics and the historical deposit ratio statistics.

[0188] Among them, the attribute information of the goods to be transported will affect the transaction probability of the goods to be transported. Based on this, the transportation service platform can determine the attribute information of the goods to be transported as a static feature, so that the impact of the attribute information of the goods to be transported can be considered when making quotation predictions.

[0189] Among them, the transportation service platform will store the transaction status of the goods sold on the platform in the historical period. When making quotation forecasts, the transportation service platform can also refer to the transaction status of the goods in the historical period, so as to improve the accuracy of the quotation forecast.

[0190] Among them, the historical transaction rate of the demand transportation route is the ratio of the total number of cargoes transported through the demand transportation route to the number of cargoes sold during the historical period.

[0191] The historical period may be the past year or the past three months, etc.

[0192] Among them, the historical supply-demand ratio of the demand transportation route is the ratio of the number of cargo sources reported by drivers and transported through the demand transportation route during the historical period to the total number of cargo sources.

[0193] Among them, the historical transaction price statistics of the demand transportation routes include multiple types.

[0194] In some examples, the historical transaction price statistics of the demand transportation route include statistical values ​​of transaction prices of delivery sources transported through the demand transportation route during a historical period.

[0195] In other examples, the historical transaction price statistics of the demand transportation route include the transaction price statistics of the cargo sources transported through the demand transportation route within the historical period and having the same vehicle length or vehicle model as the demand vehicle to be transported.

[0196] Among them, the historical kilometer price statistics of the demand transportation routes include multiple types.

[0197] In some examples, the historical kilometer price statistics of the demand transportation route may include statistical values ​​of the price per kilometer of the delivery source for the goods transported through the demand transportation route during a historical period.

[0198] In other examples, the historical kilometer price statistics of the demand transportation route may include the statistical values ​​of the per-kilometer price of the cargo source that is transported through the demand transportation route during a historical period and has the same vehicle length or model as the demand vehicle to be transported.

[0199] Among them, the historical deposit ratio statistics of the demand transportation routes include multiple types.

[0200] In some examples, the historical deposit ratio statistics of the demand transportation route may include the statistical values ​​of the ratio of the deposit to the transaction price of the delivery source transported through the demand transportation route during the historical period.

[0201] In other examples, the historical kilometer price statistics of the demand transportation route may include the statistical values ​​of the ratio of the deposit to the transaction price of the completed goods that were transported through the demand transportation route during the historical period and had the same vehicle length or model as the demand vehicle to be transported.

[0202] The above statistical value may be a median or an average, etc.

[0203] Based on this, the transportation service platform determines static features, which helps to take into account various statistical values ​​in the historical period and the attribute information of the goods to be transported when making quotation forecasts, thereby helping to obtain more accurate forecast quotations.

[0204] S302: Determine a predicted quotation based on static features and dynamic features.

[0205] The transportation service platform can combine static and dynamic features to obtain more accurate prediction quotations.

[0206] Based on the above exemplary description, in S302, the transportation service platform can be implemented through the following S302A.

[0207] See also Figure 6 , Figure 6 A flow chart of a method for determining a forecast quotation provided in an embodiment of the present application. Figure 6 As shown, S302 can be implemented by S302A:

[0208] S302A, input the static features and the dynamic features into the quotation prediction model to obtain the predicted quotation. The quotation prediction model is obtained by training the basic model according to the transaction price of the sample supply, the static sample features and the dynamic sample features.

[0209] Among them, the basic model is the LightGBM model.

[0210] Among them, the loss function of the basic model is the mean absolute error function (MAE). The mean absolute error function can avoid the problem of errors offsetting each other, so that it can accurately reflect the size of the error between the predicted price and the transaction price during the training process and improve the accuracy of the model.

[0211] Among them, the evaluation index of the basic model is as follows:

[0212]

[0213] Among them, P1 is the predicted price of the sample source, P2 is the transaction price of the sample source, k is the preset threshold, and k is greater than or equal to 0.

[0214] Here, k may be 10%.

[0215] Specifically, the transportation service platform can obtain the goods traded in the platform during the historical period as sample goods, extract the static sample features and dynamic sample features of the sample goods, and thus form a sample set.

[0216] The sample set is divided into a training set and a validation set according to a preset ratio. The basic model is trained according to the training set. During the training process, the model parameters are adjusted according to the loss function to obtain the trained basic model. The trained basic model is verified by the validation set. When the result of the trained basic model meets the requirements of the evaluation indicators, the training is completed and the quotation prediction model is obtained.

[0217] Figure 7 This is a schematic diagram of the structure of a quotation prediction device provided in one embodiment of the present application. Figure 7 As shown, the device includes: an acquisition module 101, a determination module 102 and a prediction module 103.

[0218] The acquisition module 101 is used to acquire real-time statistical information when a transaction has not been completed after a first period of time has passed since the source of goods to be transported was put on the shelf. The real-time statistical information includes exposure statistical data, route statistical data and quotation information. The exposure statistical data is used to indicate the exposure of the source of goods to be transported at the current moment. The route statistical data is used to indicate the quantity statistical value of the first source of goods. The first source of goods is the source of goods transported through the demand transportation route of the source of goods to be transported at the current moment. The quotation information includes the driver quotation and the cargo owner quotation of the source of goods to be transported at the current moment.

[0219] A determination module 102 is used to determine dynamic features based on real-time statistical information, where the dynamic features are used to characterize features that affect the probability of a transaction of the goods to be transported at the current moment;

[0220] The prediction module 103 is used to determine the predicted quotation according to the dynamic characteristics, and the predicted quotation is used to guide the consignor to modify the consignor quotation.

[0221] It should be noted that the quotation prediction device of the embodiment of the present application can be used to execute the technical solution of the above-mentioned method embodiment, and its implementation principle and technical effect are similar, which will not be repeated here.

[0222] In some examples, the quotation prediction device further includes a transceiver module:

[0223] A transceiver module, used for sending forecast information to the consignor, the forecast information including the forecast quotation, or the forecast information including the difference between the forecast quotation and the consignor's quotation;

[0224] After the second period, continue to obtain real-time statistical information and determine new forecast quotes.

[0225] In some examples, the quotation prediction device further includes a transceiver module and a judgment module:

[0226] A transceiver module, used for sending forecast information to the consignor, the forecast information including the forecast quotation, or the forecast information including the difference between the forecast quotation and the consignor's quotation;

[0227] A judgment module is used to judge whether the cargo to be transported has been sold after the third period of time has passed after the cargo owner has modified the cargo owner's quotation according to the forecast information;

[0228] When the cargo to be transported has not been sold, continue to obtain real-time statistical information and determine new forecast quotations.

[0229] In some examples, the determination module 102 is specifically used to determine the real-time transaction rate and the real-time supply-demand ratio according to the route statistical data, and the real-time transaction rate and the real-time supply-demand ratio are used to characterize the transaction status of the first source of goods;

[0230] According to the exposure statistics, the real-time click rate and real-time feedback rate are determined. The real-time click rate and real-time feedback rate are used to represent the degree of interest of the driver in the transport source;

[0231] The dynamic features determined include real-time transaction rate, real-time supply-demand ratio, first duration, shipper quotation, driver quotation, real-time click rate and real-time feedback rate.

[0232] In some examples, route statistics include the total quantity of the first source of cargo, the number of transactions, and the number of feedbacks. The real-time transaction rate is the ratio of the number of transactions to the total number. The real-time supply-demand ratio is the ratio of the number of feedbacks to the total number. Exposure statistics include the first duration, the exposure of the source of cargo to be transported, the number of driver clicks, and the number of driver feedbacks. The real-time click rate is the ratio of the driver clicks to the exposure, and the real-time feedback rate is the ratio of the driver feedback to the driver clicks.

[0233] In some examples, the prediction module 103 is specifically used to obtain static features, which include the weight, volume, category, length of the required vehicle, model of the required vehicle, departure information, destination information, historical transaction rate of the required transportation route, historical supply-demand ratio, historical transaction price statistics, historical kilometer price statistics, and historical deposit ratio statistics of the cargo source to be transported;

[0234] Determine the predicted quotation based on static and dynamic features.

[0235] In some examples, the prediction module 103 is specifically used to input static features and dynamic features into a quotation prediction model to obtain a predicted quotation. The quotation prediction model is obtained by training a basic model based on the transaction price of the sample source, static sample features, and dynamic sample features.

[0236] In some examples, the base model is a LightGBM model, the loss function of the base model is a mean absolute error function, and the evaluation index of the base model is Formula 1;

[0237] Among them, formula 1 is:

[0238] Among them, P1 is the predicted price of the sample source, P2 is the transaction price of the sample source, k is the preset threshold, and k is greater than or equal to 0.

[0239] Figure 8 This is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application. Figure 8 As shown, the electronic device may include: a processor 201 and a memory 202, wherein the memory 202 stores a computer program, and when the processor 201 executes the computer program, the embodiment of the present application is implemented. Figures 1 to 6 The quote prediction method shown.

[0240] Fig. 9 This is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application. Fig. 9 As shown, the electronic device may include: a processor 301, and when the processor 301 executes a computer executable program or an instruction in a memory to execute a computer program, the embodiment of the present application is implemented. Figures 1 to 6 The quote prediction method shown.

[0241] Another embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the present application can be implemented. Figures 1 to 6 The quote prediction method shown.

[0242] Another embodiment of the present application further provides a computer program product, including: an execution instruction, the execution instruction is stored in a readable storage medium, at least one processor of the electronic device can read the execution instruction from the readable storage medium, and at least one processor executes the execution instruction so that the electronic device implements the embodiment of the present application Figures 1 to 6 The quote prediction method shown.

[0243] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0244] Those skilled in the art will appreciate that, although some embodiments herein include certain features included in other embodiments, the combination of features of different embodiments is meant to be within the scope of the present application and form different embodiments. For example, in the claims, any one of the claimed embodiments may be used in any combination.

[0245] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A quotation prediction method, characterized in that: The method comprises: When a transaction has not been completed after a first period of time has passed since the source of goods to be transported was put on the shelf, real-time statistical information is obtained, the real-time statistical information including exposure statistical data, route statistical data and quotation information, the exposure statistical data is used to indicate the exposure status of the source of goods to be transported at the current moment, the route statistical data is used to indicate the quantity statistical value of the first source of goods, the first source of goods is the source of goods transported through the demand transportation route of the source of goods to be transported at the current moment, and the quotation information includes the driver quotation and the cargo owner quotation of the source of goods to be transported at the current moment; Determine dynamic features according to the real-time statistical information, wherein the dynamic features are used to characterize features that affect the probability of a transaction of the source of goods to be transported at the current moment; A predicted quotation is determined according to the dynamic characteristics, and the predicted quotation is used to guide the consignor to modify the consignor's quotation.

2. The method according to claim 1, characterized in that The method further comprises: Sending forecast information to the consignor, the forecast information including the forecast quotation, or the forecast information including the difference between the forecast quotation and the consignor's quotation; After the second period, continue to obtain real-time statistical information and determine new forecast quotes.

3. The method according to claim 1, characterized in that: The method further comprises: Sending forecast information to the consignor, the forecast information including the forecast quotation, or the forecast information including the difference between the forecast quotation and the consignor's quotation; After the cargo owner modifies the cargo owner's quotation according to the forecast information, after a third period of time, it is determined whether the cargo to be transported is sold; When the source of goods to be transported has not been sold, real-time statistical information continues to be obtained and a new forecast quotation is determined.

4. The method according to claim 1, characterized in that: Determining the dynamic features according to the real-time statistical information includes: Determine a real-time transaction rate and a real-time supply-demand ratio according to the route statistical data, wherein the real-time transaction rate and the real-time supply-demand ratio are used to characterize the transaction status of the first source of goods; Determine a real-time click rate and a real-time feedback rate according to the exposure statistics, wherein the real-time click rate and the real-time feedback rate are used to characterize the degree of interest of the driver in the cargo to be transported; Determining the dynamic features includes the real-time transaction rate, the real-time supply-demand ratio, the first duration, the cargo owner's quotation, the driver's quotation, the real-time click rate and the real-time feedback rate.

5. The method according to claim 4, characterized in that The route statistical data include the total quantity of the first source of goods, the number of transactions and the number of feedbacks, the real-time transaction rate is the ratio of the number of transactions to the total quantity, the real-time supply-demand ratio is the ratio of the number of feedbacks to the total quantity, the exposure statistical data include the first duration, the exposure of the source of goods to be transported, the number of driver clicks and the number of driver feedbacks, the real-time click rate is the ratio of the number of driver clicks to the exposure, and the real-time feedback rate is the ratio of the number of driver feedbacks to the number of driver clicks.

6. The method according to any one of claims 1 to 5, characterized in that: The step of determining the predicted quotation according to the dynamic feature includes: Obtain static features, including the weight, volume, category, length of the required vehicle, model of the required vehicle, departure information, destination information, historical transaction rate of the required transportation route, historical supply-demand ratio, historical transaction price statistics, historical kilometer price statistics, and historical deposit ratio statistics of the required transportation route; The predicted quotation is determined according to the static feature and the dynamic feature.

7. The method according to claim 6, characterized in that The step of determining the predicted quotation according to the static feature and the dynamic feature includes: The static features and the dynamic features are input into a quotation prediction model to obtain the predicted quotation. The quotation prediction model is obtained by training a basic model based on the transaction price of a sample source of goods, static sample features, and dynamic sample features.

8. The method according to claim 7, characterized in that The basic model is the LightGBM model, the loss function of the basic model is the mean absolute error function, and the evaluation index of the basic model is Formula 1; Among them, formula 1 is: Among them, P1 is the predicted price of the sample source of goods, P2 is the transaction price of the sample source of goods, k is a preset threshold, and k is greater than or equal to 0.

9. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the program, the method according to any one of claims 1 to 8 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images