Travel luggage recommendation method
By obtaining the similarity calculation of the user's travel feature vector and the historical travel itinerary feature vector, the recommendation index is constructed, and the problem of lack of personalization and intelligence in travel luggage preparation is solved, and more accurate luggage recommendation is achieved.
Patent Information
- Application Number
- CN202510524921.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-24
- Publication Date
- 2025-08-08
AI Technical Summary
The lack of personalized and intelligent support for travel luggage preparation in the prior art, which often faces the problem of excessive luggage or missing essential items.
By obtaining the user's travel feature vector, querying the historical travel feature vector in the luggage database, calculating the mean, standard deviation and correlation coefficient of the itinerary feature, constructing a change relationship matrix and an inverse matrix, determining the recommended index of the luggage, and recommending luggage based on the preset threshold.
It improves the intelligence and personalized adaptability of luggage recommendations, reduces the user's decision-making burden, and enhances the practicality and user experience of the recommendation system.
Smart Images
Figure CN120448633A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of intelligent recommendation technology, and in particular to a method for recommending travel luggage. Background Art
[0002] With the rapid development of the tourism industry and the diversification of people's travel needs, luggage preparation has become a crucial and indispensable part of travel planning. Traditional luggage preparation methods rely primarily on user experience and general recommendations, lacking personalized and intelligent support. This often leads to users facing issues such as overpacking, missing essential items, or carrying unnecessary items during travel. Currently, there are several technical solutions on the market related to travel recommendations. However, these technologies primarily focus on itinerary planning, attraction recommendations, and product recommendations, and have not fully considered the intelligent needs of luggage preparation.
[0003] Therefore, there is an urgent need for a method that can intelligently recommend travel luggage based on user itinerary information. Summary of the Invention
[0004] Based on this, it is necessary to provide a travel luggage recommendation method to address the above technical issues.
[0005] In a first aspect, a method for recommending travel luggage is provided, the method comprising:
[0006] Obtaining a feature vector of a pending trip corresponding to the user's pending trip;
[0007] For each piece of luggage to be recommended in the pre-established luggage database, query the first historical travel itinerary corresponding to the piece of luggage to be recommended in the pre-stored correspondence between luggage and travel itineraries;
[0008] Based on the historical orders, obtaining a travel feature vector of the first historical travel itinerary;
[0009] determining a recommendation index corresponding to the luggage to be recommended based on a similarity relationship between the itinerary feature vector of the first historical travel itinerary and the feature vector of the to-be-traveled itinerary;
[0010] According to a preset recommendation quantity threshold, a luggage item with a high recommendation index ranking is selected from all the luggage to be recommended and determined as the target recommended luggage.
[0011] As an optional implementation manner, determining the recommendation index corresponding to the luggage to be recommended based on the similarity relationship between the trip feature vector of the first historical trip and the feature vector of the trip to be traveled includes:
[0012] Determining a feature mean and a feature standard deviation of each trip feature based on the trip feature vector of the first historical travel itinerary;
[0013] Calculating correlation coefficients between the trip features;
[0014] constructing a change relationship matrix based on the standard deviation and the correlation coefficient;
[0015] Based on the change relationship matrix, determining the corresponding determinant and inverse matrix;
[0016] A recommendation index of the luggage to be recommended is determined according to the number of the first historical travel itineraries, the feature vector of the to-be-traveled itinerary, the feature mean, the determinant, and the inverse matrix.
[0017] As an optional implementation, the calculation formula for the characteristic mean of each travel characteristic is:
[0018]
[0019] Among them, μ i (D L ) is the characteristic mean of the i-th trip feature, d is the first historical travel itinerary D L The number of trip feature vectors, D i (j) For the first historical trip D L The i-th trip feature of the j-th trip feature vector in , where n is the number of trip features.
[0020] As an optional implementation, the calculation formula for the characteristic standard deviation of each stroke characteristic is:
[0021]
[0022] Among them, σ i (D L ) is the characteristic standard deviation of trip characteristic i, d is the first historical trip D L The number of trip feature vectors, D i (j) The first historical trip D L The i-th trip feature of the j-th trip feature vector in μ i (D L ) is the feature mean of the i-th feature vector, and n is the number of trip features.
[0023] As an optional implementation, the formula for calculating the correlation coefficient between the travel features is:
[0024]
[0025] Among them, ρ fg For the first historical trip D LThe fth trip feature of the jth trip feature vector in and the g-th travel feature D g (j) The correlation coefficient between f (D L ) is the characteristic mean of the f-th eigenvector, μ g (D L ) is the characteristic mean of the g-th eigenvector, and d is the first historical travel itinerary D L The number of trip feature vectors.
[0026] As an optional implementation, the change relationship matrix is:
[0027]
[0028] Among them, σ1 2 (D L ) is the square of the characteristic standard deviation of the first stroke characteristic, ρ 1n is the correlation coefficient between the first and nth trip features, σ n (D L ) is the characteristic standard deviation of the n-th trip feature.
[0029] As an optional implementation manner, the formula for determining the recommendation index of the luggage to be recommended based on the number of the first historical trips, the feature vector of the trip to be traveled, the feature mean, the determinant, and the inverse matrix is:
[0030]
[0031] Among them, R(A,L) is the recommendation index of the luggage L to be recommended, p is the preset adjustment coefficient, A is the feature vector of the trip to be traveled, μ(D L ) is the characteristic mean μ i (D L ), Σ L -1 is the inverse matrix of the change relationship matrix, |Σ L ∣ is the determinant of the change relationship matrix.
[0032] As an optional implementation, the method further includes:
[0033] For the custom luggage added by the user, obtaining the second historical travel itinerary to which the custom luggage is added;
[0034] Based on the historical order, obtaining a trip feature vector of the second historical travel itinerary;
[0035] For each to-be-processed trip, determining a minimum trip characteristic difference based on the trip characteristic vector of the to-be-processed trip and the trip characteristic vector of the second historical trip;
[0036] If the minimum trip characteristic difference is less than a preset screening parameter, the recommendation priority of the customized baggage for the to-be-processed trip is set to a low priority.
[0037] As an optional implementation manner, for each to-be-processed trip, the formula for determining the minimum trip characteristic difference based on the trip characteristic vector of the to-be-processed trip and the trip characteristic vector of the second historical trip is:
[0038] △=min 1<i<n (X i -U i );
[0039] Among them, △ is the minimum stroke characteristic difference, X i is the i-th trip feature of the trip feature vector X of the trip to be processed, U i is the i-th trip feature of the trip feature vector U of the second historical travel trip.
[0040] As an optional implementation, the method further includes:
[0041] The customized luggage is stored in the luggage database.
[0042] In a second aspect, a computer device is provided, comprising 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 computer program, the method steps described in any one of the first aspects are implemented.
[0043] According to a third aspect, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method steps described in any one of the first aspects are implemented.
[0044] The present application provides a method for recommending travel luggage, the method comprising: obtaining a feature vector of a future trip corresponding to a user's future trip; for each piece of luggage to be recommended in a pre-established luggage database, querying a first historical trip corresponding to the piece of luggage to be recommended from a pre-stored correspondence between luggage and travel itineraries; obtaining a trip feature vector of the first historical trip based on historical orders; determining a recommendation index for the piece of luggage to be recommended based on a similarity relationship between the trip feature vector of the first historical trip and the feature vector of the future trip; and selecting, based on a preset recommendation quantity threshold, a piece of luggage with a high recommendation index from all the pieces of luggage to be recommended and determining it as a target recommended piece of luggage. The technical solution provided by the embodiments of the present application provides at least the following beneficial effects: The travel luggage recommendation method provided by the embodiments of the present application obtains a trip feature vector corresponding to the user's future trip and compares it with the historical trip feature vectors associated with each piece of luggage to be recommended in a pre-established luggage database. Based on the similarity relationship between the two, combined with a preset probability model (e.g., a Gaussian distribution model), a recommendation index for each piece of luggage to be recommended is calculated. Based on a preset recommendation quantity threshold, several pieces of luggage with high recommendation indexes are selected from all the pieces of luggage to be recommended as target recommended pieces of luggage. Compared to existing approaches that rely on users manually selecting luggage or simply filtering based on labels, this application introduces a structured itinerary feature vector modeling approach, combines historical order data to construct a feature mapping relationship between itineraries and luggage, and uses statistical methods (such as mean, standard deviation, correlation coefficient, and covariance matrix) to perform quantitative similarity calculations in the feature space, enabling recommendations to more accurately match the actual needs of the user's current itinerary. This approach effectively improves the intelligence and personalized adaptability of luggage recommendations, reduces the decision-making burden on users during the luggage selection process, and enhances the practicality and user experience of the recommendation system.
[0045] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0047] Figure 1 A flowchart of a travel luggage recommendation method provided in an embodiment of the present application;
[0048] Figure 2A flowchart of a method for determining a luggage recommendation index provided in an embodiment of the present application;
[0049] Figure 3 A flowchart of a method for recommending customized luggage provided in an embodiment of the present application;
[0050] Figure 4 A schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0051] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.
[0052] The following will describe in detail a travel luggage recommendation method provided by an embodiment of the present application in conjunction with specific implementation methods. Figure 1 This is a flowchart of a travel luggage recommendation method provided in an embodiment of the present application, such as Figure 1 The specific steps are as follows:
[0053] Step 101: Obtain a feature vector of a pending trip corresponding to a user's pending trip.
[0054] In practice, the itinerary feature vector may include but is not limited to: departure place, destination, travel time, travel days, travel type (such as business, tourism), climate conditions, user gender, age group, past travel preferences and other fields. The system can encode and vectorize these fields to construct a high-dimensional vector to represent the itinerary to be traveled. For example, the itinerary feature vector X = (X1, X2, X3, ..., X n ), where X1 is the average temperature of the destination in that month, X2 is the temperature difference between the average temperature of the destination and the departure point in that month, X3 is the average precipitation of the destination in that month, X4 is the average wind speed of the destination in that month, X5 is the average UV index of the destination in that month, X6 is the altitude of the destination, X7 is the longitude of the destination, X8 is the latitude of the destination, X9 is whether the departure and destination are in the same country (1 for yes, 0 for no), X 10 Whether the departure and destination use different power plug standards (1 for yes, 0 for no), X 11 Is the driving habits of the departure and destination the same (1 for yes, 0 for no), X 12 is the oldest age among the order passengers, X 13 is the youngest age of the order passengers, X 14 is the number of male travelers, X 15By constructing a multi-dimensional feature vector, the system can comprehensively describe trip characteristics, providing data support for subsequent intelligent baggage recommendations. Furthermore, additional features can be added during subsequent operations, continuously improving the trip feature vector without affecting the overall process.
[0055] Step 102 : For each luggage to be recommended in the pre-established luggage database, query the first historical travel itinerary corresponding to the luggage to be recommended in the pre-stored correspondence between luggage and travel itineraries.
[0056] In practice, a luggage-trip mapping database can be pre-established to record the multiple trips corresponding to a specific luggage item used by a user, i.e., the mapping between luggage and trips. For a specific piece of luggage L to be recommended, the system can retrieve from this database all the trips that the luggage item has been used on in previous orders. These trips constitute the "first historical trip" set for the recommended luggage item L, which is used for subsequent modeling and analysis.
[0057] Step 103: Obtain a travel feature vector of the first historical travel itinerary based on the historical orders.
[0058] In practice, for each first historical trip retrieved in step 102, the system can extract its trip feature vector in a similar manner to step 101. For example, if luggage L was used by user A and user B in two different trips, the system can extract feature vectors for each of these two trips to form a training dataset. These vectors will be used for subsequent statistical analysis and similarity calculations.
[0059] Step 104 : Determine a recommendation index corresponding to the luggage to be recommended based on a similarity relationship between the travel feature vector of the first historical travel itinerary and the feature vector of the travel itinerary to be traveled.
[0060] During implementation, the system may determine the recommendation index corresponding to the luggage to be recommended based on the similarity relationship between the travel feature vector of the first historical travel itinerary and the feature vector of the travel itinerary to be traveled.
[0061] As an optional implementation, Figure 2 A flowchart of a method for determining a luggage recommendation index provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, in step 104, the specific steps of determining the recommendation index corresponding to the luggage to be recommended based on the similarity relationship between the trip feature vector of the first historical trip and the feature vector of the trip to be traveled are as follows:
[0062] Step 201 : Determine the feature mean and feature standard deviation of each trip feature based on the trip feature vector of the first historical travel trip.
[0063] In implementation, the system can determine the mean and standard deviation of each trip feature based on the trip feature vector of the first historical trip. For example, if the trip feature vector of the first historical trip has three feature dimensions: "trip day," "temperature," and "whether it was a business trip," the system can calculate the mean and standard deviation for each of these three dimensions, providing center and scale information for subsequent similarity modeling.
[0064] As an optional implementation, the calculation formula for the feature mean of each trip feature in step 201 is:
[0065]
[0066] Among them, μ i (D L ) is the characteristic mean of the i-th trip feature, d is the first historical travel itinerary D L The number of trip feature vectors, D i (j) For the first historical trip D L The i-th trip feature of the j-th trip feature vector in , where n is the number of trip features.
[0067] As an optional implementation, the calculation formula for the characteristic standard deviation of each travel characteristic in step 201 is:
[0068]
[0069] Among them, σ i (D L ) is the characteristic standard deviation of trip characteristic i, d is the first historical trip D L The number of trip feature vectors, D i (j) For the first historical trip D L The i-th trip feature of the j-th trip feature vector in μ i (D L ) is the feature mean of the i-th feature vector, and n is the number of trip features.
[0070] As an optional implementation, when the characteristic standard deviation σ i (D L ) is 0, you can add an extra small number to it to avoid the exception of the divisor being 0 in the subsequent calculation process. For example, you can take the value of 10 -d .
[0071] Step 202: Calculate the correlation coefficient between the travel features.
[0072] In practice, to build a complete covariance structure, the system can calculate the correlation coefficient between each trip feature, which measures the linear correlation between them. For example, if the system determines that "number of days of travel" and "whether or not to carry large luggage" are positively correlated, a large correlation coefficient indicates a strong relationship between them, which helps optimize the recommendation model.
[0073] As an optional implementation, the formula for calculating the correlation coefficient between the travel features in step 202 is:
[0074]
[0075] Among them, ρ fg For the first historical trip D L The fth trip feature of the jth trip feature vector in and the g-th travel feature D g (j) The correlation coefficient between f (D L ) is the characteristic mean of the f-th eigenvector, μ g (D L ) is the characteristic mean of the g-th eigenvector, and d is the first historical travel itinerary D L The number of trip feature vectors.
[0076] Step 203: construct a change relationship matrix based on the standard deviation and the correlation coefficient.
[0077] During implementation, the system can use the standard deviation and the correlation coefficient to construct a covariance matrix, that is, a change relationship matrix.
[0078] As an optional implementation, the change relationship matrix in step 203 is:
[0079]
[0080] Among them, σ1 2 (D L ) is the square of the characteristic standard deviation of the first stroke characteristic, ρ 1n is the correlation coefficient between the first and nth trip features, σ n (D L ) is the characteristic standard deviation of the n-th trip feature.
[0081] Step 204: Determine the corresponding determinant and inverse matrix based on the change relationship matrix.
[0082] In implementation, the system can calculate the determinant of the change relationship matrix |Σ L ∣, the specific formula is:
[0083]
[0084] Among them, Σ Lij is the matrix Σ L The submatrix after removing the i-th row and j-th column, |Σ Lji ∣ is Σ Lij The determinant of .
[0085] The system can also change the relationship matrix Σ L Perform matrix inversion operation to obtain Σ L -1 .
[0086]
[0087] Among them, the element in row i and column j is (-1) i+j ∣Σ Lji ∣.
[0088] Inverse matrix Σ L -1 Used to measure Mahalanobis distance, |Σ L ∣ is used to normalize the density function. For example: if Σ L It is a 3×3 positive definite symmetric matrix. The system can use numerical algorithms (such as Cholesky decomposition) to efficiently solve its inverse matrix and determinant to ensure computational stability.
[0089] Step 205 : Determine the recommendation index of the luggage to be recommended based on the number of the first historical travel itineraries, the feature vector, the feature mean, the determinant, and the inverse matrix of the travel itinerary.
[0090] During implementation, the system may determine the recommendation index of the luggage to be recommended based on the number of the first historical travel itineraries, the feature vector, the feature mean, the determinant and the inverse matrix of the travel itinerary to be recommended.
[0091] As an optional implementation, in step 205, the formula for determining the recommendation index of the luggage to be recommended based on the number of the first historical trips, the feature vector, the feature mean, the determinant, and the inverse matrix of the trip to be traveled is:
[0092]
[0093] Among them, R(A,L) is the recommendation index of the luggage L to be recommended, p is the preset adjustment coefficient, A is the feature vector of the trip to be traveled, μ(D L ) is the characteristic mean μ i (D L ), Σ L -1 is the inverse matrix of the change relationship matrix, |Σ L ∣ is the determinant of the change relationship matrix.
[0094] Step 105 : According to a preset recommendation quantity threshold, a luggage item with a high recommendation index ranking is selected from all luggage to be recommended, and is determined as the target recommended luggage.
[0095] In practice, the system can sort all the recommended luggage by their recommendation index from high to low, take the top k items (for example, k is a preset recommendation threshold, such as k = 3), mark the corresponding luggage as "target recommended luggage," and display them to the user on the user interface. This ensures accurate recommendations while avoiding information overload. For example, after the user enters their itinerary, the system can compare the recommendation indices of luggage A, B, C, and D, which are 0.85, 0.63, 0.60, and 0.20, respectively. If the threshold k = 2, then A and B are recommended.
[0096] As an optional implementation, Figure 3 This is a flowchart of a method for recommending customized luggage provided in an embodiment of the present application, such as Figure 3 The specific steps are as follows:
[0097] Step 301: For the customized luggage added by the user, obtain the second historical travel itinerary to which the customized luggage is added.
[0098] In practice, when a user manually adds custom luggage through the system (for example, adding a "yoga mat" or "painting tools"), the system can record the travel itinerary information associated with the custom luggage as its second historical travel itinerary. This itinerary information includes the departure point, destination, purpose, travel time, etc. entered by the user, which is used to mark the usage scenario characteristics of the luggage. For example, if a user adds a "beach tent" as custom luggage during a "seaside recuperation" trip, this trip will become the second historical travel itinerary corresponding to the custom luggage "beach tent".
[0099] Step 302: Obtain a trip feature vector of a second historical travel itinerary based on the historical orders.
[0100] In practice, the system can perform structured vectorization on the second historical travel itinerary recorded in step 301 to generate a trip feature vector, which serves as a reference vector for subsequent recommendation control. The feature vector may include: climate type, season, travel days, destination type (urban / natural), user tags, etc.
[0101] Step 303 : For each to-be-processed trip, determine the minimum trip characteristic difference based on the trip characteristic vector of the to-be-processed trip and the trip characteristic vector of the second historical trip.
[0102] In practice, the system can generate corresponding trip feature vectors for all pending trips (i.e., current or future recommended candidate trips). These are then compared with the trip feature vectors of the second historical trip along different dimensions. The difference between the two trip features can be calculated using Euclidean distance, Mahalanobis distance, or other distance functions. The minimum of all trip feature differences is then taken, representing the matching degree of the custom baggage item in the most similar trip.
[0103] As an optional implementation, in step 303, for each to-be-processed trip, the formula for determining the minimum trip characteristic difference based on the trip characteristic vector of the to-be-processed trip and the trip characteristic vector of the second historical trip is:
[0104] △=min 1<i<n (X i -U i );
[0105] Among them, △ is the minimum stroke characteristic difference, X i is the i-th trip feature of the trip feature vector X of the trip to be processed, U i is the i-th trip feature of the trip feature vector U of the second historical travel trip.
[0106] Step 304: If the minimum trip characteristic difference is less than the preset screening parameter, the recommendation priority of the customized baggage for the pending trip is set to low priority.
[0107] During implementation, technicians can pre-set the screening parameters for difference screening in the system, for example, set it to 0.4. When the minimum trip feature difference is less than 0.4, it indicates that there is a candidate trip whose feature difference with the original trip is sufficiently small (i.e., very close). The system adopts a conservative strategy when processing customized luggage. When the luggage is added by the user, in order to avoid adaptation errors caused by incorrect recommendations, the system can introduce a vector similarity constraint. That is, only when the minimum difference between the feature vectors of the current candidate trip and the first trip bound to the customized luggage is less than a set threshold ε (e.g., 0.01), the luggage item is recommended with a low priority for the trip, thereby limiting the system's over-generalized recommendations for customized luggage while taking into account the matching degree.
[0108] As an optional implementation, the system can store the customized luggage in the luggage database.
[0109] This embodiment of the present application provides a method for recommending travel luggage, comprising: obtaining a feature vector of a future trip corresponding to a user's future trip; for each piece of luggage to be recommended in a pre-established luggage database, querying a previously stored mapping between luggage and travel itineraries for a first historical trip corresponding to the piece of luggage to be recommended; obtaining a trip feature vector for the first historical trip based on historical bookings; determining a recommendation index for the piece of luggage to be recommended based on the similarity between the trip feature vector of the first historical trip and the feature vector of the future trip; and selecting, based on a preset recommendation quantity threshold, the luggage item with the highest recommendation index from all the pieces of luggage to be recommended and determining it as the target recommended luggage. This embodiment of the present application dynamically analyzes the user's itinerary characteristics through an algorithmic model to accurately recommend a suitable luggage list, preventing users from missing essential items due to lack of experience or information asymmetry. Initially, the system requires only minimal data to initiate recommendations. Furthermore, the recommendation algorithm can be continuously optimized based on the user's actual use of the recommendations, improving recommendation accuracy and user satisfaction. This reduces the time and effort users spend on luggage preparation, helps users quickly organize luggage that meets their travel needs, and enhances travel convenience. It can also provide differentiated luggage recommendations based on different travel scenarios (such as business trips, family travel, outdoor adventures, etc.) to meet the personalized needs of users and provide users with a more efficient and considerate travel luggage preparation experience.
[0110] It should be understood that although Figures 1 to 3 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. In addition, Figures 1 to 3 At least part of the steps may include multiple steps or multiple stages. These steps or stages are not necessarily performed at the same time, but can be performed at different times. The order of execution of these steps or stages is not necessarily one by one, but can be performed in turn or alternately with other steps or at least part of the steps or stages in other steps.
[0111] It can be understood that the same / similar parts between the various embodiments of the above method in this specification can be referred to each other, and each embodiment focuses on the differences from other embodiments. For related parts, please refer to the description of other method embodiments.
[0112] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program that can be run on the processor. When the processor executes the computer program, the method steps of dynamically identifying flight changes are implemented. Figure 4A schematic diagram of the structure of a computer device provided in an embodiment of the present application is shown in FIG. Figure 4 As shown, the computer device may include a processor 401, a system bus 402, a non-volatile storage medium 403, an internal memory 404, a network interface 405, a display screen 406, and an input device 407. The non-volatile storage medium 403 stores an operating system 4031 and a computer program 4032. The processor 401 is configured to execute the computer program 4032 to implement the aforementioned method steps for luggage recommendation. The system bus 402 is configured to connect the processor 401, the non-volatile storage medium 403, the internal memory 404, the network interface 405, the display screen 406, and the input device 407, ensuring efficient communication between these components. The internal memory 404 is configured to temporarily store running programs and data, enabling the processor 401 to quickly access required information, thereby improving overall system performance. The network interface 405 (e.g., a network card) enables the computer device to connect to a local area network or the Internet, enabling data transmission and remote communication. The display screen 406 is configured to present the luggage recommendation information processed by the computer device to the user in graphical or textual form. The input device 407 (such as a keyboard, mouse, touch screen, etc.) is used to allow the user to input travel itinerary data and luggage recommendation commands into the computer device to achieve interactive operations with the computer device.
[0113] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
[0114] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.
[0115] It should also be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data for analysis, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.
[0116] Each embodiment in this specification is described in a related manner. Similar parts between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the system embodiment is generally similar to the method embodiment, so the description is relatively simple. For related parts, refer to the description of the method embodiment.
[0117] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0118] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.
Claims
1. A travel luggage recommendation method, characterized in that: The method comprises: Obtaining a feature vector of a pending trip corresponding to the user's pending trip; For each piece of luggage to be recommended in a pre-established luggage database, query the first historical travel itinerary corresponding to the piece of luggage to be recommended in the pre-stored correspondence between luggage and travel itineraries; Based on the historical orders, obtaining a travel feature vector of the first historical travel itinerary; determining a recommendation index corresponding to the luggage to be recommended based on a similarity relationship between the itinerary feature vector of the first historical travel itinerary and the feature vector of the to-be-traveled itinerary; According to a preset recommendation quantity threshold, a luggage item with a high recommendation index ranking is selected from all the luggage to be recommended and determined as the target recommended luggage.
2. The method according to claim 1, characterized in that The determining, based on the similarity relationship between the itinerary feature vector of the first historical travel itinerary and the itinerary feature vector of the to-be-traveled travel itinerary, of a recommendation index corresponding to the luggage to be recommended includes: Determining a feature mean and a feature standard deviation of each trip feature based on the trip feature vector of the first historical travel itinerary; Calculating correlation coefficients between the trip features; constructing a change relationship matrix based on the standard deviation and the correlation coefficient; Based on the change relationship matrix, determining the corresponding determinant and inverse matrix; A recommendation index of the luggage to be recommended is determined according to the number of the first historical travel itineraries, the feature vector of the to-be-traveled itinerary, the feature mean, the determinant, and the inverse matrix.
3. The method according to claim 2, characterized in that The calculation formula of the characteristic mean of each stroke characteristic is: Among them, μ i (D L ) is the characteristic mean of the i-th trip feature, d is the first historical travel itinerary D L The number of trip feature vectors, D i (j) For the first historical trip D L The i-th trip feature of the j-th trip feature vector in , where n is the number of trip features.
4. The method according to claim 2, characterized in that The calculation formula for the characteristic standard deviation of each stroke characteristic is: Among them, σ i (D L ) is the characteristic standard deviation of trip characteristic i, d is the first historical trip D L The number of trip feature vectors, D i (j) For the first historical trip D L The i-th trip feature of the j-th trip feature vector in μ i (D L ) is the feature mean of the i-th feature vector, and n is the number of trip features.
5. The method according to claim 2, characterized in that The formula for calculating the correlation coefficient between the travel features is: Among them, ρ fg For the first historical trip D L The fth trip feature of the jth trip feature vector in and the g-th travel feature D g (j) The correlation coefficient between f (D L ) is the characteristic mean of the f-th eigenvector, μ g (D L ) is the characteristic mean of the g-th eigenvector, and d is the first historical travel itinerary D L The number of trip feature vectors.
6. The method according to claim 2, characterized in that The change relationship matrix is: Among them, σ1 2 (D L ) is the square of the characteristic standard deviation of the first stroke characteristic, ρ 1n is the correlation coefficient between the first and nth trip features, σ n (D L ) is the characteristic standard deviation of the n-th trip feature.
7. The method according to claim 2, characterized in that The formula for determining the recommendation index of the luggage to be recommended based on the number of the first historical travel itineraries, the feature vector of the to-be-traveled itinerary, the feature mean, the determinant, and the inverse matrix is: Among them, R(A,L) is the recommendation index of the luggage L to be recommended, p is the preset adjustment coefficient, A is the feature vector of the trip to be traveled, μ(D L ) is the characteristic mean μ i (D L ), Σ L -1 is the inverse matrix of the change relationship matrix, |Σ L ∣ is the determinant of the change relationship matrix.
8. The method according to claim 1, characterized in that The method further comprises: For the custom luggage added by the user, obtaining the second historical travel itinerary to which the custom luggage is added; Based on the historical order, obtaining a trip feature vector of the second historical travel itinerary; For each to-be-processed trip, determining a minimum trip characteristic difference based on the trip characteristic vector of the to-be-processed trip and the trip characteristic vector of the second historical trip; If the minimum trip characteristic difference is less than a preset screening parameter, the recommendation priority of the customized baggage for the to-be-processed trip is set to a low priority.
9. The method according to claim 8, characterized in that For each to-be-processed trip, the formula for determining the minimum trip characteristic difference based on the trip characteristic vector of the to-be-processed trip and the trip characteristic vector of the second historical trip is: △=min 1<i<n (X i -U i ); Among them, △ is the minimum stroke characteristic difference, X i is the i-th trip feature of the trip feature vector X of the trip to be processed, U i is the i-th trip feature of the trip feature vector U of the second historical travel trip.
10. The method according to claim 8, characterized in that The method further comprises: The customized luggage is stored in the luggage database.