Method and device for recommending bottom guarantee ticket of travel transfer scheme, and electronic equipment

Abstract

The invention provides a travel transfer scheme guarantee ticket recommendation method and device and electronic equipment, and relates to the technical field of computers. The method comprises the following steps: acquiring sorted recommendation schemes and candidate schemes to be inserted; assembling the sorted recommendation schemes and the candidate schemes to be inserted to obtain an assembly list; evaluating a first assembly scheme in the assembly list to obtain a first evaluation score, and evaluating each second assembly scheme in the assembly list to obtain a second evaluation score corresponding to each second assembly scheme; selecting a maximum score from the second evaluation scores, and calculating the maximum score and the first evaluation score to obtain a calculation result; and under the condition that the calculation result is greater than a preset insertion threshold value, inserting the candidate scheme to be inserted corresponding to the maximum score in the first place of the sorted recommendation schemes to obtain a recommended guaranteed bottom ticket. The preference of the user can be met, and the influence on the existing recommendation scheme can be considered.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, device and electronic device for recommending a guaranteed ticket for a travel transfer plan. Background Art

[0002] When some seats on the direct train that the user intends to buy are out of stock, a recommended list will appear below, including transfers, joint rides, flights, etc. The guaranteed ticket option is usually an alternative option that meets the user's preferences after comprehensively considering multiple factors such as time, price, and comfort. Although it may not be the user's initial first choice, it provides a relatively reasonable alternative option when there are no tickets for the direct train, which helps to facilitate the successful completion of the transaction and meet the user's travel needs.

[0003] In the related art, the ctcvr estimate is directly performed on the candidate resources to be inserted. When the ctcvr estimate value is greater than a certain threshold, the candidate resource with the largest ctcvr estimate value is selected for first insertion. This method can ensure that the daily purchase volume of the inserted candidate resource is greater than a certain value, but it will affect the existing list. The daily purchase volume of the existing list will decrease, resulting in a decrease in the overall purchase volume. The root cause of this problem is that only the user's preference for the inserted candidate resource is considered, and the impact of the insertion on the existing recommendation plan is not considered. Summary of the invention

[0004] The present application provides a method, device and electronic device for recommending a guaranteed ticket for a travel transfer plan, which can meet the user's preferences and take into account the impact on existing recommended plans.

[0005] The technical solution of the embodiment of the present application is as follows: In a first aspect, an embodiment of the present application provides a method for recommending a guaranteed ticket for a travel transfer plan, the method comprising: Get the sorted recommended solutions and candidate solutions to be inserted; Assembling the sorted recommended solutions and the candidate solutions to be inserted to obtain an assembly list, wherein the assembly list includes a first assembly solution composed of the sorted recommended solutions, and a plurality of second assembly solutions composed of one candidate solution to be inserted and all the sorted recommended solutions; Evaluate the first assembly scheme in the assembly list to obtain a first evaluation score, and evaluate each of the second assembly schemes in the assembly list to obtain a second evaluation score corresponding to each of the second assembly schemes; Select a maximum score from each of the second evaluation scores, and calculate the maximum score and the first evaluation score to obtain a calculation result; When the calculation result is greater than a preset insertion threshold, the candidate solution to be inserted corresponding to the maximum score is inserted at the first position of the sorted recommended solutions to obtain a recommended minimum ticket.

[0006] In the above technical solution, firstly, the sorted recommended solutions and the candidate solutions to be inserted are obtained, and the obtained solutions provide data support for the subsequent insertion recommendation; the sorted recommended solutions and the candidate solutions to be inserted are assembled to obtain an assembly list, wherein the assembly list includes a first assembly plan composed of the sorted recommended solutions, and a plurality of second assembly plans composed of a candidate solution to be inserted and all the sorted recommended solutions, and different plans are assembled so that in the subsequent insertion, not only the user's preferences can be considered, but also the impact on the sorted recommended solutions can be taken into account; the first assembly plan in the assembly list is evaluated to obtain a first evaluation score, and each second assembly plan in the assembly list is evaluated to obtain a second evaluation score corresponding to each second assembly plan, and by scoring, it is determined whether to insert the first place in the subsequent process to better meet the user's preferences; the maximum score is selected from each second evaluation score, and the maximum score is calculated with the first evaluation score to obtain a calculation result; when the calculation result is greater than the preset insertion threshold, the candidate solution to be inserted corresponding to the maximum score is inserted at the first place of the sorted recommended solutions to obtain a recommendation guarantee ticket, which can meet the user's preferences according to the above judgment and also take into account the impact on the existing recommended solutions.

[0007] In some embodiments of the present application, evaluating the first assembly scheme in the assembly list to obtain a first evaluation score, and evaluating each of the second assembly schemes in the assembly list to obtain a second evaluation score corresponding to each of the second assembly schemes, include: The first assembly scheme in the assembly list is evaluated using a preset sequence scoring model to obtain a first evaluation score, and each of the second assembly schemes in the assembly list is evaluated using the sequence scoring model to obtain a second evaluation score corresponding to each of the second assembly schemes.

[0008] In the above technical scheme, by evaluating the first assembly scheme and the second assembly scheme respectively, a first evaluation score corresponding to the first assembly scheme and a second evaluation score corresponding to each second assembly scheme are obtained, which is conducive to the subsequent determination of the candidate scheme to be inserted according to the first evaluation score and the second evaluation score, so as to meet the user's preferences and take into account the impact on the existing recommended schemes.

[0009] In some embodiments of the present application, the training process of the sequence scoring model includes: Acquire a sample data set, the sample data set comprising a plurality of sequence sample data and an exposure-purchase rate score corresponding to each of the sequence sample data; Scoring the sequence sample data using a preset initial scoring model to obtain a scoring result; Performing a loss function calculation on the scoring result and the exposure-purchase rate score to obtain a value of the loss function; The value of the loss function is used to iteratively adjust the parameters of the initial scoring model, and the sequence scoring model is obtained when the preset training conditions are met.

[0010] In the above technical solution, before scoring, the initial scoring model is first trained with a large amount of data in the sample data set to obtain a sequence scoring model, so as to increase the stability and generalization of the sequence scoring model, thereby improving the accuracy of scoring and improving the accuracy of recommendations.

[0011] In some embodiments of the present application, the sequence sample data is list-assembled sample sequence data, and the list-assembled sample sequence data includes a sequence data list and tag data; The sequence data list represents a list of each sample data sorted in the order of exposure; The flag data indicates whether each of the sample data is purchased.

[0012] In the above technical solution, the sequence data list and the labeled data are data obtained after preprocessing, and the accuracy of the model can be improved by processing the sequence sample data.

[0013] In some embodiments of the present application, before obtaining the sorted recommendation solutions, the method further includes: Obtain L solutions to be sorted; The L to-be-sorted solutions are sorted using a preset sorting algorithm to obtain the sorted recommended solutions, wherein the sorting algorithm is a refined sorting model or a sorting strategy.

[0014] In the above technical solution, L solutions to be sorted are obtained to provide data support. Then, the L solutions to be sorted are sorted using a refined sorting model or one of the sorting strategies, and sorted according to the exposure-click rate index to obtain sorted recommended solutions. Obtaining the sorted recommended solutions is conducive to subsequent solution reorganization, thereby digging out the impact of the inserted solution on the recommended solution.

[0015] In some embodiments of the present application, before obtaining the candidate solution to be inserted, the method further includes: Obtain N candidate solutions; The N candidate solutions are scored and ranked using a preset estimation model, and x candidate solutions with the highest scores are selected as the candidate solutions to be inserted, wherein the estimation model is a purchase rate estimation model.

[0016] In the above technical solution, N candidate solutions are obtained to provide data support. The N candidate solutions are evaluated using a preset estimation model to obtain the exposure quality of the candidate solutions and obtain evaluation scores. The evaluation scores are then sorted, and the x candidate solutions with the highest scores are selected as the candidate solutions to be inserted. The candidate solutions to be inserted can meet the user's preferences and can guarantee travel needs.

[0017] In some embodiments of the present application, after calculating the maximum score and the first evaluation score to obtain a calculation result, the method further includes: When the calculation result is less than or equal to the preset insertion threshold, the first solution of the sorted recommended solutions is recommended.

[0018] In the above technical solution, when the calculation result is less than or equal to the preset insertion threshold, it indicates that the candidate solution to be inserted is more in line with the user's travel needs than the sorted recommended solution. Therefore, the first solution of the sorted recommended solution is recommended, which can better meet the user's preferences and facilitate the user's travel.

[0019] In a second aspect, an embodiment of the present application provides a device for recommending a guaranteed ticket for a travel transfer plan, the device comprising: A data acquisition module is used to obtain sorted recommended solutions and candidate solutions to be inserted; A data assembly module, used for assembling the sorted recommended solutions and the candidate solutions to be inserted to obtain an assembly list, wherein the assembly list includes a first assembly plan composed of the sorted recommended solutions, and a plurality of second assembly plans composed of one candidate solution to be inserted and all the sorted recommended solutions; A scheme evaluation module, used to evaluate the first assembly scheme in the assembly list to obtain a first evaluation score, and to evaluate each of the second assembly schemes in the assembly list to obtain a second evaluation score corresponding to each of the second assembly schemes; A score calculation module, used for selecting a maximum score from each of the second evaluation scores, and calculating the maximum score and the first evaluation score to obtain a calculation result; The solution insertion module is used to insert the candidate solution to be inserted corresponding to the maximum score at the first place of the sorted recommended solutions when the calculation result is greater than a preset insertion threshold, so as to obtain a recommended minimum ticket.

[0020] In a third aspect, an embodiment of the present application provides an electronic device, comprising a processor, a memory, a user interface, a communication bus and a network interface, wherein the processor, the memory, the user interface and the network interface are respectively connected to the communication bus, the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes any one of the methods provided in the first aspect.

[0021] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, and when the instructions are executed, any one of the methods provided in the first aspect is executed.

[0022] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. Since the sorted recommended solutions and candidate solutions to be inserted are obtained, data support is provided for subsequent insertion recommendations by obtaining solutions; the sorted recommended solutions and candidate solutions to be inserted are assembled to obtain an assembly list, wherein the assembly list includes a first assembly solution composed of sorted recommended solutions, and a plurality of second assembly solutions composed of a candidate solution to be inserted and all sorted recommended solutions, and different solutions are assembled so that in the subsequent insertion, not only user preferences can be considered, but also the impact on the sorted recommended solutions can be taken into account; the first assembly solution in the assembly list is evaluated to obtain a first evaluation score, and each second assembly solution in the assembly list is evaluated to obtain a second evaluation score corresponding to each second assembly solution, and by scoring, it is determined whether to insert the first one in the subsequent process to better meet the user's preferences; the maximum score is selected from each second evaluation score, and the maximum score is calculated with the first evaluation score to obtain a calculation result; when the calculation result is greater than the preset insertion threshold, the candidate solution to be inserted corresponding to the maximum score is inserted at the first position of the sorted recommended solutions to obtain a recommendation guarantee ticket, and according to the above judgment, the user's preferences can be met and the impact on the existing recommended solutions can be taken into account. Therefore, the problem that the related art only considers the user's preference for inserting candidate resources, but does not consider the impact of the insertion on the existing recommendation solutions is effectively solved.

[0023] 2. Scoring through the sequential scoring model will help improve the estimated exposure-purchase rate in the future. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is one of the flow charts of a method for recommending a guaranteed ticket for a travel transfer solution provided by an embodiment of the present application; Figure 2This is the second flow chart of a method for recommending a guaranteed ticket for a travel transfer solution provided by an embodiment of the present application; Figure 3 Flow chart 3 of a method for recommending a guaranteed ticket for a travel transfer solution provided by an embodiment of the present application; Figure 4 Flow chart 4 of a method for recommending a guaranteed ticket for a travel transfer solution provided by an embodiment of the present application; Figure 5 This is a schematic diagram of the overall process of a method for recommending a guaranteed ticket for a travel transfer solution provided by an embodiment of the present application; Figure 6 It is a structural schematic diagram of a device for recommending guaranteed tickets for a travel transfer solution provided by an embodiment of the present application; Figure 7 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0026] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0027] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0028] The embodiment of the present application provides a method, device, electronic device and readable storage medium for recommending a guaranteed ticket for a travel transfer scheme. The method for recommending a guaranteed ticket for a travel transfer scheme first obtains sorted recommended schemes and candidate schemes to be inserted, and provides data support for subsequent insertion recommendations by obtaining the schemes; assembles the sorted recommended schemes and the candidate schemes to be inserted to obtain an assembly list, wherein the assembly list includes a first assembly scheme composed of sorted recommended schemes, and a plurality of second assembly schemes composed of a candidate scheme to be inserted and all sorted recommended schemes, and assembles different schemes so that when inserting them later, not only user preferences can be taken into account, but also the impact on the sorted recommended schemes can be taken into account; The first assembly scheme in the assembly list is evaluated to obtain a first evaluation score, and each second assembly scheme in the assembly list is evaluated to obtain a second evaluation score corresponding to each second assembly scheme. The scoring is performed to determine whether to insert the first scheme in the future to better meet the user's preferences; the maximum score is selected from each second evaluation score, and the maximum score is calculated with the first evaluation score to obtain a calculation result; when the calculation result is greater than a preset insertion threshold, the candidate scheme to be inserted corresponding to the maximum score is inserted at the first position of the sorted recommended scheme to obtain a recommended minimum ticket. According to the above judgment, the user's preferences can be met and the impact on the existing recommended schemes can be taken into account.

[0029] It should be noted that the method for recommending guaranteed tickets for travel transfer plans is applied to the recommendation of transfer and combined travel plans of trains, buses, airplanes, etc., which can meet the travel needs of users, take into account the influence of user preferences and existing recommended plans, and improve the exposure-purchase rate.

[0030] The technical solution provided in the embodiments of the present application is further described below in conjunction with the accompanying drawings.

[0031] Reference Figure 1 , Figure 1 1 is a flow chart of a method for recommending a guaranteed ticket for a travel transfer scheme provided in an embodiment of the present application. The method for recommending a guaranteed ticket for a travel transfer scheme is applied to a device for recommending a guaranteed ticket for a travel transfer scheme, and the method for recommending a guaranteed ticket for a travel transfer scheme is executed by a processor in an electronic device or a readable storage medium, and the method for recommending a guaranteed ticket for a travel transfer scheme includes steps S110, S120, S130, S140, and S150.

[0032] Step S110, obtaining sorted recommended solutions and candidate solutions to be inserted.

[0033] In one embodiment, the sorted recommended scheme is a scheme that has been sorted by a sorting algorithm. The sorted recommended schemes are numbered in sequence according to the sorted sequence, and a single recommended scheme data is represented as [total duration of the guaranteed ticket, departure time of the guaranteed ticket, ticket price, ..., total price of the guaranteed ticket] = [180, 12.5, 168, ..., 268]. The candidate scheme to be inserted is an inserted scheme that caters to the user's preferences and has a small impact on the recommended scheme to ensure the user's travel needs. The representation format of the candidate scheme to be inserted is [total duration of the guaranteed ticket, departure time of the guaranteed ticket, ticket price, ..., total price of the guaranteed ticket] = [140, 12.5, 148, ..., 240]. The preset data reading function can be used to read the sorted recommended schemes and the candidate schemes to be inserted, and data support is provided for subsequent insertion recommendations by obtaining the sorted recommended schemes and the candidate schemes to be inserted. Among them, the preset data reading function is the read() function.

[0034] like Figure 2 As shown, before obtaining the sorted recommended solutions, the method for recommending guaranteed tickets for travel transfer solutions also includes but is not limited to the following steps: Step S210, obtaining L solutions to be sorted.

[0035] Step S220, using a preset sorting algorithm to sort the L to-be-sorted solutions to obtain sorted recommended solutions, wherein the sorting algorithm is a refined sorting model or a sorting strategy.

[0036] In some possible embodiments of the present application, L to-be-sorted schemes are obtained through a data reading function, and the to-be-sorted schemes can also be expressed as [total duration of guaranteed ticket, departure time of guaranteed ticket, ticket price, ..., total price of guaranteed ticket] = [180, 12.5, 168, ..., 268]. The L to-be-sorted schemes are sorted using a refined sorting model or one of the sorting strategies, and sorted according to the exposure-click rate index to obtain sorted recommended schemes. Obtaining the sorted recommended schemes is conducive to subsequent scheme reorganization, thereby digging out the impact of the inserted scheme on the recommended scheme. Among them, the refined sorting model can be a machine learning algorithm, and the training of the machine learning algorithm will not be repeated here.

[0037] like Figure 3 As shown, before obtaining the candidate plan to be inserted, the method for recommending a guaranteed ticket for a travel transfer plan also includes but is not limited to the following steps: Step S310, obtaining N candidate solutions.

[0038] Step S320, using a preset estimation model to score and sort the N candidate solutions, and selecting the x candidate solutions with the highest scores as the candidate solutions to be inserted, wherein the estimation model is a purchase rate estimation model.

[0039] In some possible embodiments of the present application, N candidate plans are obtained through a data reading function, and the candidate plans can also be expressed as [total duration of the guaranteed ticket, departure time of the guaranteed ticket, ticket grabbing price, ..., total price of the guaranteed ticket] = [140, 12.5, 148, ..., 240]. The N candidate plans are evaluated using the purchase rate estimation model to derive the exposure quality of the candidate plans and obtain the evaluation scores. The evaluation scores are then sorted, exemplarily in descending order, and the x candidate plans with the highest scores are selected as the candidate plans to be inserted. The candidate plans to be inserted can be expressed as t1, t2...tx. The candidate plan to be inserted can meet user preferences and guarantee travel needs.

[0040] Step S120, assembling the sorted recommended solutions and the candidate solutions to be inserted to obtain an assembly list, wherein the assembly list includes a first assembly solution composed of the sorted recommended solutions, and multiple second assembly solutions composed of a candidate solution to be inserted and all the sorted recommended solutions.

[0041] In one embodiment, the sorted recommended solutions and the candidate solutions to be inserted are assembled by splicing each candidate solution to be inserted with the sorted recommended solutions to obtain an assembly list. The assembly list includes a first assembly solution composed of the sorted recommended solutions, and the first assembly solution does not include the candidate solution to be inserted, but only the sorted recommended solution, which is labeled 0. The assembly list also includes a plurality of second assembly solutions composed of a candidate solution to be inserted and all the sorted recommended solutions, that is, each candidate solution to be inserted is inserted before the sorted dispatch solution to form a second assembly solution, and the number of solutions in the second assembly solution is the same as the number of candidate solutions to be inserted, and the arrangement sequence is from 1 to x. Therefore, the assembly list includes x+1 assembly solutions.

[0042] In another embodiment, the assembly list is further numbered according to the number, the first assembly scheme consisting of the sorted recommended schemes is numbered 0, and the multiple second assembly schemes are numbered 1, 2...x in sequence according to the sequence numbers of the candidate schemes to be inserted, to obtain an assembly list. By assembling different schemes, not only user preferences can be considered during subsequent insertion, but also the impact on the sorted recommended schemes can be taken into account. Exemplarily, the assembly list can be expressed as follows: 0: [L sorted recommended schemes], 1: [t1, L sorted recommended schemes], 2: [t2, L sorted recommended schemes], x: [tx, L sorted recommended schemes].

[0043] Step S130, evaluating the first assembly scheme in the assembly list to obtain a first evaluation score, and evaluating each second assembly scheme in the assembly list to obtain a second evaluation score corresponding to each second assembly scheme.

[0044] In one embodiment, the first assembly scheme in the assembly list is evaluated to obtain a first evaluation score, and each second assembly scheme in the assembly list is evaluated to obtain a second evaluation score corresponding to each second assembly scheme, including but not limited to: using a preset sequence scoring model to evaluate the first assembly scheme in the assembly list to obtain a first evaluation score, and using a sequence scoring model to evaluate each second assembly scheme in the assembly list to obtain a second evaluation score corresponding to each second assembly scheme.

[0045] In some possible embodiments of the present application, the preset sequence scoring model may be a multi-head attention mechanism model or a recurrent neural network model for context perception of the sequence. Exemplarily, the first assembly scheme in the assembly list is evaluated using the head attention mechanism model to obtain a first evaluation score, and the second assembly schemes in the assembly list are evaluated using the head attention mechanism model to obtain the second evaluation scores corresponding to the second assembly schemes, so as to determine whether to insert the first place of the sorted recommended scheme according to the first evaluation score and the second evaluation score, thereby recommending the scheme that is most conducive to the user's travel.

[0046] like Figure 4 As shown, the training process of the sequence scoring model includes but is not limited to the following steps: Step S410, obtaining a sample data set, where the sample data set includes a plurality of sequence sample data and an exposure-purchase rate score corresponding to each sequence sample data.

[0047] In one embodiment, the sequence sample data is list-assembled sample sequence data, which includes a sequence data list and tag data; the data in the sequence data list is ordered, and is a list of each sample data sorted in the order of exposure; the tag data indicates whether each sample data is purchased. The list-assembled sample sequence data can be obtained from the Internet through crawler technology, and then through data preprocessing and manual labeling to obtain a sample data set. The sample data set is stored so that the data can be read later for model training. Among them, the data preprocessing can be missing value processing, normalization processing, discrete feature encoding processing, etc.

[0048] In another embodiment, a sample data set may be obtained through a data reading function, the sample data set including a plurality of sequence sample data and exposure-purchase rate scores corresponding to each sequence sample data, and the sample data set may be obtained so that the sequence sample data and the exposure-purchase rate scores may be used for model training to obtain a sequence scoring model. The sample data set is in the form of an assembled list, including sequence feature data of a plurality of sequence sample data and their label data, which is a list of schemes sorted in order of exposure, and may be expressed as (m is the length of the list of schemes in the sequence): [the first ranked scheme, the second ranked scheme, ..., the mth ranked scheme]. Exemplarily, the single scheme feature data structure is as follows (n is the length of the single scheme feature list): [X1, X2, X3, ..., Xn], such as [total duration, departure time, destination, ..., total price] = [120, 12.5, 268, ..., 168]. Therefore, the feature data structure of a single ordered sequence is as follows: [X1, X2, X3, ..., Xn] of the 1st-ranked solution, [X1, X2, X3, ..., Xn] of the 2nd-ranked solution, ..., [X1, X2, X3, ..., Xn] of the mth-ranked solution. The tag data structure of a single ordered sequence is as follows: [Y of the 1st-ranked solution, Y of the 2nd-ranked solution, ..., Y of the mth-ranked solution], such as [1, 0, 0, ..., 0], where 1 can be defined as the solution being purchased, and 0 can be defined as the solution being exposed but not purchased.

[0049] Step S420: score the sequence sample data using a preset initial scoring model to obtain a scoring result.

[0050] In one embodiment, the preset initial scoring model is a multi-head attention mechanism model; it can also be a recurrent neural network model, and the recurrent neural network model can be a unidirectional long short-term memory network or a bidirectional long short-term memory network. The multi-head attention mechanism model is used to extract context-aware information from the sequence sample data to perform exposure-purchase rate scoring and obtain a scoring result. The obtained scoring result is conducive to the subsequent loss function calculation to train the model.

[0051] Step S430, performing loss function calculation on the scoring result and the exposure-purchase rate score to obtain the value of the loss function.

[0052] In one embodiment, the Logloss() function is used as the loss function, and the loss function is calculated by using Logloss() on the scoring results and the exposure-purchase rate scores to obtain the value of the loss function. The loss function calculation is performed to facilitate subsequent back propagation of the model.

[0053] Step S440, using the value of the loss function to iteratively adjust the parameters of the initial scoring model, and obtaining a sequence scoring model when the preset training conditions are met.

[0054] In one embodiment, the preset training condition may be that the loss function tends to be stable, or may be the number of training iterations (the number of iterations is set based on experience). Exemplarily, the number of training iterations is used as the training condition, and the value of the loss function obtained in step S430 is used to iteratively adjust the parameters of the initial scoring model, specifically: for each round of iteration, all sequence sample data are input into the initial scoring model, context-aware information is extracted, the value of the loss function is calculated, and the value of the loss function is used to adjust the parameters of the initial scoring model, and after the set number of iterations, a sequence scoring model is obtained. According to the obtained sequence scoring model, the assembled data is scored so as to determine whether to insert the first position later to better meet the user's preferences.

[0055] Step S140: Select a maximum score from each second evaluation score, and calculate the maximum score and the first evaluation score to obtain a calculation result.

[0056] In one embodiment, according to the first evaluation score and the second evaluation score obtained in step S130, the maximum score is selected from each second evaluation score, and the maximum score indicates that the assembly plan not only meets the user's travel needs and caters to the user's preferences, but also can reduce the impact on the sorted recommended plans. Then the maximum score is calculated with the first evaluation score, specifically, the maximum score is subtracted from the first evaluation score to obtain a calculation result. According to the calculation result, it is determined later whether to insert the candidate plan to be inserted at the first position of the sorted recommended plans.

[0057] Step S150, when the calculated result is greater than the preset insertion threshold, the candidate solution to be inserted corresponding to the maximum score is inserted at the first position of the sorted recommended solutions to obtain a recommended minimum ticket.

[0058] In one embodiment, when the calculation result is greater than the preset insertion threshold, it indicates that the candidate plan to be inserted is more in line with the user's travel needs and has less impact on the sorted recommended plans. Therefore, the candidate plan to be inserted corresponding to the maximum score is inserted at the first place of the sorted recommended plans to recommend the candidate plan to be inserted to the user and obtain a recommended minimum ticket. The recommended minimum ticket can meet the user's preferences and take into account the impact on the existing recommended plans.

[0059] In one embodiment, after calculating the maximum score and the first evaluation score to obtain the calculation result, the method for recommending a guaranteed ticket for a travel transfer solution further includes step S160: Step S160: When the calculation result is less than or equal to the preset insertion threshold, recommend the first solution of the sorted recommended solutions.

[0060] In one embodiment, when the calculation result is less than or equal to the preset insertion threshold, it indicates that the candidate solution to be inserted is more in line with the user's travel needs than the sorted recommended solutions. Therefore, the first solution of the sorted recommended solutions is recommended, which can better meet the user's preferences and facilitate the user's travel.

[0061] like Figure 5 As shown, an embodiment of the present application provides a schematic diagram of the overall process of a method for recommending a guaranteed ticket for a travel transfer plan. Before obtaining the sorted recommended plan, obtain L plans to be sorted, and use a preset sorting algorithm to sort the L plans to be sorted to obtain a sorted recommended plan. Before obtaining the candidate plan to be inserted, obtain N candidate plans to be inserted, use a preset estimation model to score and sort the N candidate plans to be inserted, and select the x candidate plans with the highest scores as the candidate plans to be inserted according to the evaluation scores. The candidate plans to be inserted can be represented as t1, t2...tx. Then obtain the sorted recommended plan and the candidate plan to be inserted, and provide data support for subsequent insertion recommendations by obtaining the plan; assemble the sorted recommended plan and the candidate plan to be inserted to obtain an assembly list, wherein the assembly list includes a first assembly plan composed of the sorted recommended plan, and a plurality of second assembly plans composed of a candidate plan to be inserted and all the sorted recommended plans, and assemble different plans so that when inserting later, not only the user's preferences can be considered, but also the impact on the sorted recommended plan can be taken into account. The assembly candidate sequence can be expressed as 0: [L sorted recommended solutions], 1: [t1, L sorted recommended solutions], 2: [t2, L sorted recommended solutions], x: [tx, L sorted recommended solutions]. Before scoring the assembly list, a preset sequence scoring model M2 is trained, and the first assembly solution in the assembly list is evaluated using the sequence scoring model to obtain a first evaluation score S0, and each second assembly solution in the assembly list is evaluated to obtain a second evaluation score S1, S2...Sx corresponding to each second assembly solution. By scoring, it is possible to determine whether to insert the first one in the future to better meet the user's preferences. Subsequently, the maximum score is selected from each second evaluation score, and the maximum score is calculated with the first evaluation score to obtain a calculation result; when the calculation result is greater than the preset insertion threshold, the candidate solution to be inserted corresponding to the maximum score is inserted at the first position of the sorted recommended solution to obtain a recommendation guarantee ticket. According to the above judgment, the user's preferences can be met, and the impact on the existing recommended solutions can also be taken into account. When the calculated result is less than or equal to the preset insertion threshold, it indicates that the candidate solution to be inserted is more in line with the user's travel needs than the sorted recommended solution. Therefore, the first solution of the sorted recommended solution is recommended, which can better meet the user's preferences and facilitate the user's travel.

[0062] like Figure 6 As shown, the embodiment of the present application provides a travel transit plan guaranteed ticket recommendation device 100, which obtains the sorted recommended solutions and the candidate solutions to be inserted through the data acquisition module 110, and provides data support for subsequent insertion recommendation through the acquired solutions; and then uses the data assembly module 120 to assemble the sorted recommended solutions and the candidate solutions to be inserted to obtain an assembly list, wherein the assembly list includes a first assembly plan composed of the sorted recommended solutions, and a plurality of second assembly plans composed of a candidate solution to be inserted and all the sorted recommended solutions, and assembles different plans so that in the subsequent insertion, not only the user's preferences can be taken into account, but also the sorted recommended solutions can be taken into account. The impact of the scheme; then the scheme evaluation module 130 is used to evaluate the first assembly scheme in the assembly list to obtain a first evaluation score, and each second assembly scheme in the assembly list is evaluated to obtain a second evaluation score corresponding to each second assembly scheme, and the scoring is performed to determine whether to insert the first one in the future to better meet the user's preferences; the score calculation module 140 is used to select the maximum score from each second evaluation score, and the maximum score is calculated with the first evaluation score to obtain the calculation result; finally, the scheme insertion module 150 inserts the candidate scheme to be inserted corresponding to the maximum score at the first place of the sorted recommended scheme when the calculation result is greater than the preset insertion threshold, and obtains the recommended minimum ticket. According to the above judgment, the user's preferences can be met, and the impact on the existing recommended schemes can also be taken into account.

[0063] It should be noted that the data acquisition module 110 is connected to the data assembly module 120, the data assembly module 120 is connected to the solution evaluation module 130, the solution evaluation module 130 is connected to the score calculation module 140, and the score calculation module 140 is connected to the solution insertion module 150. The above-mentioned travel transfer solution guaranteed ticket recommendation method is applied to the travel transfer solution guaranteed ticket recommendation device 100, which can determine whether to insert the candidate solution to be inserted by assembling the sorted recommended solutions and the candidate solutions to be inserted, scoring and sorting them, which not only meets the travel needs of users, but also takes into account the user's preferences and the influence of the existing recommended solutions, thereby improving the exposure-purchase rate.

[0064] It should also be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0065] The present application also discloses an electronic device. Figure 7 , Figure 7 The electronic device 500 may include: at least one processor 501 , at least one network interface 504 , a user interface 503 , a memory 505 , and at least one communication bus 502 .

[0066] The communication bus 502 is used to realize the connection and communication between these components.

[0067] The user interface 503 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 503 may also include a standard wired interface and a wireless interface.

[0068] The network interface 504 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0069] Among them, the processor 501 may include one or more processing cores. The processor 501 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 505, and calling data stored in the memory 505. Optionally, the processor 501 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 501 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 501, and it can be implemented separately through a chip.

[0070] Among them, the memory 505 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 505 includes a non-transitory computer-readable storage medium. The memory 505 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 505 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 505 may also be optionally at least one storage device located away from the aforementioned processor 501. Refer to Figure 7 , the memory 505 as a computer storage medium may include an operating system, a network communication module, a user interface module and an application program for a method for recommending a guaranteed ticket for a travel transfer plan.

[0071] exist Figure 7 In the electronic device 500 shown, the user interface 503 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 501 can be used to call the application program stored in the memory 505 for recommending a method for a guaranteed ticket for a travel transfer plan. When executed by one or more processors 501, the electronic device 500 executes one or more methods in the above-mentioned embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited to the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0072] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0073] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0074] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0075] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0076] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0077] The above are only exemplary embodiments of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, those skilled in the art will easily think of other embodiments of the present disclosure.

[0078] This application is intended to cover any variation, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art not described in the present disclosure. The description and examples are to be regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A method for recommending guaranteed tickets for travel transfer schemes, characterized in that: The method comprises: Get the sorted recommended solutions and candidate solutions to be inserted; Assembling the sorted recommended solutions and the candidate solutions to be inserted to obtain an assembly list, wherein the assembly list includes a first assembly solution composed of the sorted recommended solutions, and a plurality of second assembly solutions composed of one candidate solution to be inserted and all the sorted recommended solutions; Evaluate the first assembly scheme in the assembly list to obtain a first evaluation score, and evaluate each of the second assembly schemes in the assembly list to obtain a second evaluation score corresponding to each of the second assembly schemes; Select a maximum score from each of the second evaluation scores, and calculate the maximum score and the first evaluation score to obtain a calculation result; When the calculation result is greater than a preset insertion threshold, the candidate solution to be inserted corresponding to the maximum score is inserted at the first position of the sorted recommended solutions to obtain a recommended minimum ticket.

2. The method according to claim 1, characterized in that The step of evaluating the first assembly scheme in the assembly list to obtain a first evaluation score, and evaluating each of the second assembly schemes in the assembly list to obtain a second evaluation score corresponding to each of the second assembly schemes, includes: The first assembly scheme in the assembly list is evaluated using a preset sequence scoring model to obtain a first evaluation score, and each of the second assembly schemes in the assembly list is evaluated using the sequence scoring model to obtain a second evaluation score corresponding to each of the second assembly schemes.

3. The method according to claim 2, characterized in that The training process of the sequence scoring model includes: Acquire a sample data set, the sample data set comprising a plurality of sequence sample data and an exposure-purchase rate score corresponding to each of the sequence sample data; Scoring the sequence sample data using a preset initial scoring model to obtain a scoring result; Performing a loss function calculation on the scoring result and the exposure-purchase rate score to obtain a value of the loss function; The value of the loss function is used to iteratively adjust the parameters of the initial scoring model, and the sequence scoring model is obtained when the preset training conditions are met.

4. The method according to claim 3, characterized in that The sequence sample data is list-assembled sample sequence data, and the list-assembled sample sequence data includes a sequence data list and label data; The sequence data list represents a list of each sample data sorted in the order of exposure; The flag data indicates whether each of the sample data is purchased.

5. The method according to claim 1, characterized in that Before obtaining the sorted recommended solutions, the method further includes: Obtain L solutions to be sorted; The L to-be-sorted solutions are sorted using a preset sorting algorithm to obtain the sorted recommended solutions, wherein the sorting algorithm is a refined sorting model or a sorting strategy.

6. The method according to claim 1, characterized in that Before obtaining the candidate solution to be inserted, the method further includes: Obtain N candidate solutions; The N candidate solutions are scored and ranked using a preset estimation model, and the x candidate solutions with the highest scores are selected as the candidate solutions to be inserted, wherein the estimation model is a purchase rate estimation model.

7. The method according to claim 1, characterized in that After calculating the maximum score and the first evaluation score to obtain a calculation result, the method further includes: When the calculation result is less than or equal to the preset insertion threshold, the first solution of the sorted recommended solutions is recommended.

8. A device for recommending guaranteed tickets for travel transfer schemes, characterized in that: The device comprises: A data acquisition module (110), used to acquire the sorted recommended solutions and the candidate solutions to be inserted; A data assembly module (120) is used to assemble the sorted recommended solutions and the candidate solutions to be inserted to obtain an assembly list, wherein the assembly list includes a first assembly plan composed of the sorted recommended solutions, and a plurality of second assembly plans composed of one candidate solution to be inserted and all the sorted recommended solutions; A scheme evaluation module (130) is used to evaluate the first assembly scheme in the assembly list to obtain a first evaluation score, and to evaluate each of the second assembly schemes in the assembly list to obtain a second evaluation score corresponding to each of the second assembly schemes; A score calculation module (140), configured to select a maximum score from each of the second evaluation scores, and calculate the maximum score and the first evaluation score to obtain a calculation result; The solution insertion module (150) is used to insert the candidate solution to be inserted corresponding to the maximum score at the first position of the sorted recommended solutions when the calculation result is greater than a preset insertion threshold, so as to obtain a recommended minimum ticket.

9. An electronic device, characterized in that: The electronic device (500) comprises a processor (501), a memory (505), a user interface (503), a communication bus (502) and a network interface (504), wherein the processor (501), the memory (505), the user interface (503) and the network interface (504) are respectively connected to the communication bus (502), the memory (505) is used to store instructions, the user interface (503) and the network interface (504) are used to communicate with other devices, and the processor (501) is used to execute the instructions stored in the memory (505) so that the electronic device (500) executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 7 is executed.

Images