Travel route recommendation method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a travel route recommendation method and device, electronic equipment and a storage medium, and belongs to the technical field of artificial intelligence. The method comprises the following steps: acquiring historical travel data of a target user, and determining a preference feature vector of the target user according to the historical travel data; obtaining travel starting and ending point information of the target user, determining a plurality of to-be-recommended travel routes according to the travel starting and ending point information, and determining destination feature vectors corresponding to the to-be-recommended travel routes; determining user preference scores of the to-be-recommended travel routes according to the preference feature vectors and the destination feature vectors; and screening a plurality of target travel routes from the to-be-recommended travel routes according to the user preference scores, and recommending the target travel routes to the target user. According to the embodiment of the invention, the accuracy and diversity of travel route recommendation can be improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a travel route recommendation method and device, an electronic device, and a storage medium. Background Art

[0002] When users are traveling by car, the travel APP in the car provides great convenience for users to plan travel routes. The positioning of the RV itself is a means of transportation for travel, so for the APP installed in the RV, the travel route recommendation has higher requirements for real-time capture of user needs.

[0003] Conventional travel route recommendations are usually based on the user's historical behavior, preferences, and static features of the destination. It is difficult to capture the user's attention to different features. The relevant recommendation algorithms often generate recommendation results based on fixed feature weights and cannot flexibly adjust the model's attention to different features, making it difficult to capture the complex relationship between user preferences and destinations, resulting in inaccurate recommendation results or failure to meet user needs. In addition, existing recommendation algorithms tend to recommend similar types of travel routes, resulting in a lack of diversity in recommendation results and failure to meet user needs for different types of routes. In addition, the recommendation results generated by existing recommendation algorithms often lack explanatory power, making it difficult for users to understand the basis for the recommendation, which affects the user's trust in the recommendation system. The above problems need to be solved urgently. Summary of the invention

[0004] The main purpose of the embodiments of the present application is to propose a travel route recommendation method and device, an electronic device and a storage medium, aiming to improve the accuracy and diversity of travel route recommendations.

[0005] To achieve the above-mentioned purpose, a first aspect of an embodiment of the present application proposes a travel route recommendation method, the method comprising:

[0006] Acquire historical travel data of a target user, and determine a preference feature vector of the target user according to the historical travel data;

[0007] Acquire the starting and ending point information of the target user's travel, determine a plurality of to-be-recommended travel routes according to the starting and ending point information of the travel, and determine a destination feature vector corresponding to each of the to-be-recommended travel routes;

[0008] Determining a user preference score for each of the to-be-recommended travel routes according to the preference feature vector and the destination feature vector;

[0009] A plurality of target travel routes are obtained by screening the travel routes to be recommended according to the user preference scores, and the target travel routes are recommended to the target user.

[0010] In some embodiments, determining the preference feature vector of the target user according to the historical travel data includes:

[0011] Determine multiple historical travel locations of the target user and corresponding travel times and stay durations according to the historical travel data;

[0012] Determine a plurality of travel styles corresponding to each of the historical travel locations, and determine a location preference score for each of the historical travel locations based on the number of trips and the length of stay;

[0013] Determining a style preference weight for each travel style according to the location preference score of the historical travel location corresponding to each travel style;

[0014] The preference feature value of each travel style dimension is determined according to the style preference weight of each travel style, and the preference feature vector is generated according to the preference feature value of each travel style dimension.

[0015] In some embodiments, determining a plurality of recommended travel routes according to the travel starting and ending point information includes at least one of the following:

[0016] Acquire a preset travel route database, and match the travel start and end point information in the travel route database to obtain a plurality of travel routes to be recommended;

[0017] The travel start and end point information is input into a pre-trained path planning model to obtain a plurality of travel routes to be recommended.

[0018] In some embodiments, determining the destination feature vector corresponding to each of the to-be-recommended travel routes includes:

[0019] Determining a plurality of intermediate destinations of the travel route to be recommended, and determining scenic spot information and travel style information of each of the intermediate destinations;

[0020] Determining a location score of the intermediate destination according to the scenic spot information, and determining a travel style multidimensional vector of the intermediate destination according to the travel style information;

[0021] Using the travel style multidimensional vector as a key vector and the place rating as a value vector to obtain a key-value vector pair of the intermediate destination;

[0022] The destination feature vector is generated according to the key-value vector pair of each of the intermediate destinations.

[0023] In some embodiments, determining the user preference score of each of the to-be-recommended travel routes according to the preference feature vector and the destination feature vector comprises:

[0024] Determining key vectors and value vectors of multiple intermediate destinations of the to-be-recommended travel route according to the destination feature vector;

[0025] Taking the preference feature vector as a query vector, calculating the similarity between the query vector and each of the key vectors, and determining the attention score of each of the corresponding intermediate destinations according to the similarity;

[0026] Normalizing the attention scores to obtain attention weights of the intermediate destinations;

[0027] The numerical vector is weightedly summed according to the attention weight to obtain the user preference score of the travel route to be recommended.

[0028] In some embodiments, the step of selecting a plurality of target travel routes from the to-be-recommended travel routes according to the user preference score comprises at least one of the following:

[0029] Comparing the user preference score with a preset first threshold, and if the user preference score is greater than or equal to the first threshold, determining the corresponding to-be-recommended travel route as the target travel route;

[0030] The recommended travel routes are sorted from large to small according to the user preference scores, and a number of the recommended travel routes with higher sorting values ​​are selected as the target travel routes.

[0031] In some embodiments, recommending the target travel route to the target user includes:

[0032] Determining scenic spot information and travel style information of multiple intermediate destinations corresponding to the target travel route;

[0033] Determine a plurality of scenic spot tags according to the scenic spot information, generate a scenic spot preview screen corresponding to each of the scenic spot tags, and then embed the scenic spot preview screen into the scenic spot tag;

[0034] determining a plurality of travel style tags according to the travel style information;

[0035] A travel route recommendation scheme is generated according to the target travel route, the scenic spot tag and the travel style tag, and the travel route recommendation scheme is pushed to the target user.

[0036] To achieve the above-mentioned purpose, a second aspect of an embodiment of the present application provides a travel route recommendation device, the device comprising:

[0037] A preference feature vector determination module, used to obtain historical travel data of a target user, and determine a preference feature vector of the target user according to the historical travel data;

[0038] A destination feature vector determination module is used to obtain the starting and ending point information of the target user's travel, determine a plurality of to-be-recommended travel routes according to the starting and ending point information of the travel, and determine the destination feature vector corresponding to each of the to-be-recommended travel routes;

[0039] A user preference score determination module, used to determine the user preference score of each of the to-be-recommended travel routes according to the preference feature vector and the destination feature vector;

[0040] The travel route screening module is used to screen a plurality of target travel routes from the travel routes to be recommended according to the user preference scores, and recommend the target travel routes to the target user.

[0041] To achieve the above-mentioned purpose, the third aspect of an embodiment of the present application proposes an electronic device, which includes a memory, a processor, a program stored on the memory and executable on the processor, and a data bus for realizing connection and communication between the processor and the memory. When the program is executed by the processor, the travel route recommendation method described in the first aspect above is implemented.

[0042] To achieve the above-mentioned purpose, the fourth aspect of an embodiment of the present application proposes a storage medium, which is a computer-readable storage medium used for computer-readable storage, and the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the travel route recommendation method as described in the first aspect above.

[0043] The travel route recommendation method and device, electronic device and storage medium proposed in the present application obtain the historical travel data of the target user, determine the preference feature vector of the target user based on the historical travel data, then obtain the travel start and end point information of the target user, determine multiple travel routes to be recommended based on the travel start and end point information, and determine the destination feature vector corresponding to each travel route to be recommended, and then determine the user preference score of each travel route to be recommended based on the preference feature vector and the destination feature vector, and finally select and obtain several target travel routes from the travel routes to be recommended based on the user preference score, and recommend the target travel route to the target user. The embodiment of the present application determines the preference feature vector based on the historical travel data of the user, determines the travel route to be recommended and the corresponding destination feature vector based on the travel start and end point information of the user, and determines the user preference score of each travel route to be recommended based on the preference feature vector and the destination feature vector, thereby selecting the target travel route for recommendation based on the user preference score, and can provide the user with a travel route that meets the user's preferences and real-time needs, improves the accuracy and diversity of travel route recommendations, and also improves the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a flow chart of a travel route recommendation method provided in an embodiment of the present application;

[0045] Figure 2 yes Figure 1 A flow chart of step S101 in FIG.

[0046] Figure 3 yes Figure 1 A flow chart of step S102 in FIG.

[0047] Figure 4 yes Figure 1 Another flow chart of step S102 in ;

[0048] Figure 5 yes Figure 1 A flow chart of step S103 in FIG.

[0049] Figure 6 yes Figure 1 A flow chart of step S104 in FIG.

[0050] Figure 7 yes Figure 1 Another flow chart of step S104 in ;

[0051] Figure 8 is a schematic diagram of the structure of a travel route recommendation device provided in an embodiment of the present application;

[0052] Fig. 9 is a schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application;

[0053] Fig.10 It is a schematic diagram of the structure of the storage medium provided in an embodiment of the present application. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0055] It should be noted that, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification, claims and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0056] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0057] First, some nouns involved in this application are analyzed:

[0058] Artificial intelligence (AI) is a new technical science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. AI is a branch of computer science that attempts to understand the essence of intelligence and produce a new type of intelligent machine that can respond in a similar way to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing and expert systems. AI can simulate the information process of human consciousness and thinking. AI is also a theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0059] Natural language processing (NLP): NLP uses computers to process, understand and apply human languages ​​(such as Chinese, English, etc.). NLP is a branch of artificial intelligence and an interdisciplinary subject between computer science and linguistics. It is often referred to as computational linguistics. Natural language processing includes grammatical analysis, semantic analysis, and text understanding. Natural language processing is often used in technical fields such as machine translation, handwritten and printed character recognition, speech recognition and text-to-speech conversion, information intent recognition, information extraction and filtering, text classification and clustering, public opinion analysis and opinion mining. It involves data mining related to language processing, machine learning, knowledge acquisition, knowledge engineering, artificial intelligence research, and linguistic research related to language computing.

[0060] Information Extraction: A text processing technology that extracts specified types of entity, relationship, event and other factual information from natural language text and forms structured data output. Information extraction is a technology that extracts specific information from text data. Text data is composed of some specific units, such as sentences, paragraphs, and chapters. Text information is composed of some small specific units, such as characters, words, phrases, sentences, paragraphs, or a combination of these specific units. Extracting noun phrases, names, place names, etc. from text data are all text information extraction. Of course, the information extracted by text information extraction technology can be various types of information.

[0061] User Preferences: Refers to the rational and biased choices made by users when considering goods and services. It is the comprehensive result of user cognition, psychological feelings and rational economic trade-offs. In daily life, the concept of "preference" is widely used consciously and unconsciously. People often use their preferences to assist in daily decision-making, or simply use preferences to express preferences and biased opinions, without involving actual choices. In more cases, we need to face and deal with more complex preference analysis situations, such as conditional preferences, that is, users need to meet certain conditions and prerequisites when making biased choices and intentional judgments.

[0062] Attention Mechanism: Originated from the study of human vision. In cognitive science, due to the bottleneck of information processing, humans selectively focus on a part of all information while ignoring other visible information. The above mechanism is usually called the attention mechanism. The attention mechanism allows the model to dynamically focus on different parts of the input sequence when processing sequence data. In deep learning, the attention mechanism is often used to increase the model's attention to the input data, thereby improving the performance and accuracy of the model.

[0063] When users are traveling by car, the travel APP in the car provides great convenience for users to plan travel routes. The positioning of the RV itself is a means of transportation for travel, so for the APP installed in the RV, the travel route recommendation has higher requirements for real-time capture of user needs.

[0064] Conventional travel route recommendations are usually based on the user's historical behavior, preferences, and static features of the destination. It is difficult to capture the user's attention to different features. The relevant recommendation algorithms often generate recommendation results based on fixed feature weights, and cannot flexibly adjust the model's attention to different features, making it difficult to capture the complex relationship between user preferences and destinations, resulting in inaccurate recommendation results or failure to meet user needs. In addition, existing recommendation algorithms tend to recommend similar types of travel routes, resulting in a lack of diversity in recommendation results and failure to meet user needs for different types of routes. In addition, the recommendation results generated by existing recommendation algorithms often lack explanatory power, making it difficult for users to understand the basis for the recommendation, which affects the user's trust in the recommendation system.

[0065] Based on this, the embodiments of the present application provide a travel route recommendation method and device, an electronic device and a storage medium, aiming to improve the accuracy and diversity of travel route recommendations.

[0066] The travel route recommendation method and device, electronic device and storage medium provided in the embodiments of the present application are specifically described through the following embodiments. First, the travel route recommendation method in the embodiments of the present application is described.

[0067] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0068] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0069] The travel route recommendation method provided in the embodiment of the present application relates to the field of artificial intelligence technology. The travel route recommendation method provided in the embodiment of the present application can be applied to a terminal, can be applied to a server side, or can be software running in a terminal or a server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, or a server cluster or a distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the travel route recommendation method, etc., but is not limited to the above forms.

[0070] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0071] It should be noted that in each specific implementation of the present application, when it comes to the need to perform relevant processing based on data related to user identity or characteristics such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first, and the collection, use, and processing of these data will comply with relevant laws, regulations, and standards. In addition, when the embodiment of the present application needs to obtain the user's sensitive personal information, the user's separate permission or consent will be obtained through a pop-up window or by jumping to a confirmation page. After clearly obtaining the user's separate permission or consent, the necessary user-related data for the normal operation of the embodiment of the present application will be obtained.

[0072] Figure 1 is an optional flowchart of the travel route recommendation method provided in the embodiment of the present application. Figure 1 The method may include but is not limited to steps S101 to S104.

[0073] Step S101, obtaining historical travel data of a target user, and determining a preference feature vector of the target user according to the historical travel data;

[0074] Step S102, obtaining the starting and ending point information of the target user's travel, determining a plurality of travel routes to be recommended according to the starting and ending point information, and determining the destination feature vector corresponding to each of the travel routes to be recommended;

[0075] Step S103, determining the user preference score of each to-be-recommended travel route according to the preference feature vector and the destination feature vector;

[0076] Step S104, screening out a number of target travel routes from the travel routes to be recommended according to the user preference scores, and recommending the target travel routes to the target user.

[0077] Step S101 of the embodiment of the present application extracts the user's preference feature vector based on the user's historical travel data, thereby realizing user representation learning; step S102 determines multiple travel routes to be recommended based on the user's travel start and end point information, and extracts the destination feature vector of the intermediate destination based on the travel route to be recommended, thereby realizing destination representation learning; step S103 introduces an attention mechanism, and determines the user's attention to different destination features by calculating the similarity between the preference feature vector and the destination feature vector, thereby obtaining the user preference score of the travel route to be recommended; step S104 selects the target travel route based on the user preference score, and generates the final travel route recommendation result. The embodiment of the present application determines the preference feature vector based on the user's historical travel data, determines the travel route to be recommended and the corresponding destination feature vector based on the user's travel start and end point information, and determines the user preference score of each travel route to be recommended based on the preference feature vector and the destination feature vector, thereby selecting the target travel route for recommendation based on the user preference score, and can provide users with travel routes that meet user preferences and real-time needs, improve the accuracy and diversity of travel route recommendations, and also improve user experience.

[0078] See also Figure 2 In some embodiments, step S101 may include but is not limited to steps S1011 to S1014:

[0079] Step S1011, determining multiple historical travel locations of the target user and the corresponding travel times and stay durations according to the historical travel data;

[0080] Step S1012, determining a number of travel styles corresponding to each historical travel location, and determining a location preference score for each historical travel location based on the number of trips and the length of stay;

[0081] Step S1013, determining the style preference weight of each travel style according to the location preference score of the historical travel locations corresponding to each travel style;

[0082] Step S1014, determining the preference feature value of each travel style dimension according to the style preference weight of each travel style, and generating a preference feature vector according to the preference feature value of each travel style dimension.

[0083] In step S1011 of some embodiments, a plurality of historical travel locations visited by the target user in the historical period, as well as the number of trips to each historical travel location and the duration of stay for each trip can be counted based on the historical travel data.

[0084] In step S1012 of some embodiments, text description information of each historical travel location is captured on the Internet through crawler technology, and information is extracted from the text description information through natural language processing technology to obtain several travel styles of each historical travel location, such as natural scenery, outdoor activities, historical culture, food experience, etc.; the number of trips and the length of stay are weighted and summed according to preset weight coefficients to obtain the location preference score of each historical travel location.

[0085] In step S1013 of some embodiments, the travel style and location preference score of each historical travel location have been determined in the previous step. This step counts the sum of the location preference scores of multiple historical travel locations of the same travel style to obtain the style preference score of each travel style, and then normalizes the style preference score to obtain the style preference weight of each travel style. It should be noted that the same historical travel location can have multiple corresponding travel styles at the same time, and the same travel style can also have multiple corresponding historical travel locations at the same time. Therefore, when counting, the location preference score of each historical travel location needs to be added to the style preference scores of all corresponding travel styles.

[0086] In step S1014 of some embodiments, the preference characteristic values ​​of each travel style dimension are determined according to the style preference weights of each travel style, and a preference characteristic vector is generated according to the preference characteristic values ​​of each travel style dimension. For example, if the style preference weights of user A for natural scenery, outdoor activities, historical culture, and food experience are 0.4, 0.3, 0.3, and 0, respectively, then it can be determined that the preference characteristic values ​​of the four travel style dimensions of natural scenery, outdoor activities, historical culture, and food experience are 0.4, 0.3, 0.3, and 0, respectively, and the generated preference characteristic vector can be expressed as (0.4, 0.3, 0.3, 0).

[0087] It can be understood that the embodiment of the present application extracts the target user's preference feature vector based on the target user's historical travel data, and can determine the target user's preference degree for different travel styles, thereby facilitating the subsequent determination of the user preference score for each recommended travel route, thereby improving the accuracy of travel route recommendations.

[0088] See also Figure 3 In some embodiments, the step of determining a plurality of recommended travel routes according to the travel starting and ending point information in step S102 may include, but is not limited to, at least one of steps S1021 to S1022:

[0089] Step S1021, obtaining a preset travel route database, matching the travel route database according to the travel start and end point information, and obtaining a plurality of travel routes to be recommended;

[0090] Step S1022, input the travel start and end point information into a pre-trained path planning model to obtain multiple travel routes to be recommended.

[0091] In step S1021 of some embodiments, multiple existing travel routes may be collected in advance, and these travel routes may be marked according to the travel starting and ending points to obtain multiple sample travel routes with different travel starting and ending points, thereby forming a travel route database; after obtaining the travel starting and ending point information input by the target user, the travel route database is traversed according to the travel starting and ending point information to obtain multiple recommended travel routes corresponding to the travel starting and ending point information.

[0092] In step S1022 of some embodiments, a path planning model may be constructed in advance based on the traffic network information, and combined with the travel start and end point information input by the target user, path planning is performed based on the path planning model to obtain multiple recommended travel routes. It should be noted that the process of constructing a path planning model is a well-known technology in the art, which is not the focus of the embodiments of this application and will not be described in detail here.

[0093] The travel route recommendation method of the embodiment of the present application is mainly used in self-driving travel scenarios. Considering that in self-driving travel scenarios, users often first select the starting point and end point of the travel and then plan the route. Therefore, the embodiment of the present application needs to plan and recommend a specific travel route after determining the user's travel starting and ending point information.

[0094] It can be understood that the embodiment of the present application, by constructing a travel route database / path planning model, can quickly determine multiple travel routes to be recommended based on the travel route database / path planning model after obtaining the travel start and end point information input by the user, thereby improving the diversity of travel route recommendations.

[0095] See also Figure 4In some embodiments, the step of determining the destination feature vector corresponding to each recommended travel route in step S102 may include, but is not limited to, steps S1023 to S1026:

[0096] Step S1023, determining multiple intermediate destinations of the travel route to be recommended, and determining the scenic spot information and travel style information of each intermediate destination;

[0097] Step S1024, determining the location score of the intermediate destination according to the scenic spot information, and determining the travel style multidimensional vector of the intermediate destination according to the travel style information;

[0098] Step S1025, using the travel style multidimensional vector as a key vector and the place score as a value vector to obtain a key-value vector pair of the intermediate destination;

[0099] Step S1026: Generate a destination feature vector based on the key-value vector pairs of each intermediate destination.

[0100] In step S1023 of some embodiments, for the travel routes to be recommended determined in the aforementioned steps, multiple intermediate destinations of each travel route to be recommended are determined, and text description information of each intermediate destination is captured on the Internet through crawler technology. Information is extracted from the text description information through natural language processing technology, so that several travel styles of each intermediate destination and several attractions in the area can be determined, and the attraction information and travel style information of each intermediate destination can be obtained.

[0101] In step S1024 of some embodiments, based on the scenic spot information of the intermediate destination, the scenic spot scores of each scenic spot of the intermediate destination are obtained from the preset tourism platform, and then the average value is taken as the location score of the intermediate destination; based on the travel style information of the intermediate destination, several destination travel styles corresponding to the intermediate destination can be determined, and the characteristic value of the travel style dimension corresponding to each destination travel style is set to 1, and the characteristic values ​​of the remaining travel style dimensions are set to 0, so as to obtain the travel style multidimensional vector of the intermediate destination. For example, if the travel style information of a certain intermediate destination X indicates that the travel style of the intermediate destination X includes natural scenery and historical culture, then the characteristic value of the travel style dimension of natural scenery and historical culture is set to 1, and the characteristic value of the travel style dimension of outdoor activities and food experience is set to 0. Similarly, according to the dimensional order of natural scenery, outdoor activities, historical culture and food experience, the obtained travel style multidimensional vector can be expressed as (1, 0, 1, 0).

[0102] In steps S1025 and S1026 of some embodiments, the travel style multidimensional vector of the intermediate destination is used as the key vector K, and the place score of the intermediate destination is used as the numerical vector V to obtain the key-value vector pair {K, V} of the intermediate destination; according to the key-value vector pairs of each intermediate destination i of the recommended travel route, the destination feature vector {K i , V i}, i∈N, N represents the set of intermediate destinations of the travel route to be recommended.

[0103] In an embodiment of the present application, a key-value vector pair of each intermediate destination is formed according to the travel styles and place scores of multiple intermediate destinations of the travel route to be recommended, thereby obtaining a destination feature vector of the travel route to be recommended. Based on the destination feature vector, the travel style characteristics and place score characteristics of the intermediate destinations of the travel route to be recommended can be accurately represented, which facilitates the subsequent determination of the user preference score of each travel route to be recommended in combination with the preference feature vector of the target user, thereby further improving the accuracy of the travel route recommendation.

[0104] See also Figure 5 In some embodiments, step S103 may include but is not limited to steps S1031 to S1034:

[0105] Step S1031, determining key vectors and value vectors of multiple intermediate destinations of the travel route to be recommended according to the destination feature vector;

[0106] Step S1032, taking the preference feature vector as a query vector, calculating the similarity between the query vector and each key vector, and determining the attention score of each corresponding intermediate destination according to the similarity;

[0107] Step S1033, normalizing the attention scores to obtain the attention weights of each intermediate destination;

[0108] Step S1034, performing weighted summation on the numerical vector according to the attention weight to obtain the user preference score of the travel route to be recommended.

[0109] The embodiment of the present application calculates user preference scores based on the attention mechanism. Assuming that user A likes natural scenery and outdoor activities, while user B likes historical culture and food experience, the attention mechanism can make it possible to pay more attention to the information on the two travel style dimensions of natural scenery and outdoor activities when recommending a travel route to user A, and pay more attention to the information on the two travel style dimensions of historical culture and food experience when recommending a travel route to user B. In this way, a travel route that is more suitable for natural scenery and outdoor activities can be recommended to user A, and a travel route that is more suitable for historical culture and food experience can be recommended to user B.

[0110] In the attention mechanism-related algorithms, the input information includes:

[0111] Q (query vector): represents the information to be queried.

[0112] K (key vector): represents the information used to calculate the correlation.

[0113] V (numeric vector): represents the values ​​weighted according to the relevance weights.

[0114] In step S1031 of some embodiments, the destination feature vector {K i , V i}, read out the key vector K of each intermediate destination i i and a numerical vector V i .

[0115] In step S1032 of some embodiments, the preference feature vector of the target user obtained in the above step is used as the query vector Q, and the query vector Q and the key vector K are calculated. i The similarity between the query vector Q and the key vector K is used to determine the attention score of each intermediate destination. i The similarity can be determined by dot product, additive attention or other similarity algorithms, which is not limited in this embodiment of the present application.

[0116] In some optional embodiments, the attention score of the intermediate destination may be determined by the following formula:

[0117]

[0118] Among them, Attention(Q,K i ) represents the attention score of the intermediate destination i, d Ki Denotes the key vector K i Dimension.

[0119] In step S1033 of some embodiments, after the attention scores of the intermediate destinations are calculated, the attention scores are normalized to obtain the attention weights of the intermediate destinations. The specific formula is as follows:

[0120] Attention_Weights(Q,K i )=softmax(Attention(Q,K i ))

[0121] Among them, Attention_Weights(Q,K i ) represents the attention weight of the intermediate destination i.

[0122] In step S1034 of some embodiments, the attention weight is applied to the numerical vector V, and the weighted sum obtained is the user preference score of the recommended travel route, that is, the information that the recommendation algorithm pays attention to. The specific calculation formula is as follows:

[0123]

[0124] The output Attention_Output is the user preference score of the travel route to be recommended.

[0125] In some optional embodiments, a multi-head attention mechanism can be used to calculate multiple groups of attention in parallel to improve the expressiveness and generalization capabilities of the model.

[0126] It can be appreciated that the embodiment of the present application calculates the user preference score of the travel route to be recommended based on the attention mechanism, which can pay more attention to the user's preferred travel style, improve the accuracy of the user preference score calculation, and further improve the accuracy of the travel route recommendation.

[0127] See also Figure 6 In some embodiments, the step of selecting a plurality of target travel routes from the recommended travel routes according to the user preference scores in step S104 may include, but is not limited to, at least one of steps S1041 to S1042:

[0128] Step S1041, comparing the user preference score with a preset first threshold, and if the user preference score is greater than or equal to the first threshold, determining the corresponding recommended travel route as the target travel route;

[0129] Step S1042, sorting the recommended travel routes from large to small according to the user preference scores, and selecting a number of recommended travel routes with higher ranking values ​​as target travel routes.

[0130] In step S1041 of some embodiments, a preset first threshold is obtained, the user preference score is compared with the preset first threshold, and the recommended travel route with a user preference score greater than or equal to the first threshold is used as the target travel route.

[0131] In step S1042 of some embodiments, the recommended travel videos are sorted from large to small according to the user preference scores, and then M (eg, 3) recommended routes with top ranking values ​​are selected as target travel routes.

[0132] See also Figure 7 In some embodiments, the step of recommending the target travel route to the target user in step S104 may include, but is not limited to, steps S1043 to S1046:

[0133] Step S1043, determining the scenic spot information and travel style information of multiple intermediate destinations corresponding to the target travel route;

[0134] Step S1044, determining a plurality of scenic spot tags according to the scenic spot information, generating a scenic spot preview screen corresponding to each scenic spot tag, and then embedding the scenic spot preview screen into the scenic spot tag;

[0135] Step S1045, determining a plurality of travel style tags according to the travel style information;

[0136] Step S1046, generating a travel route recommendation plan based on the target travel route, scenic spot tags, and travel style tags, and pushing the travel route recommendation plan to the target user.

[0137] In step S1043 of some embodiments, for the target travel route determined in the aforementioned steps, multiple intermediate destinations of the target travel route are determined, and text description information of each intermediate destination is captured on the Internet through crawler technology. Information is extracted from the text description information through natural language processing technology, and several travel styles of each intermediate destination and several attractions in the area can be determined, so that the attraction information and travel style information of each intermediate destination can be obtained.

[0138] In step S1044 of some embodiments, all attractions on the target travel route are obtained based on the summary of attraction information of each intermediate destination, and then the attractions are screened based on the popularity or rating of each attraction on the travel platform to obtain multiple target attractions and generate corresponding multiple attraction tags. At the same time, for each attraction tag, a corresponding attraction preview screen is generated based on the video and / or picture of the relevant attraction (authorized by the copyright owner), and then the attraction preview screen is embedded in the attraction tag. In this way, when the target user views the recommended target travel route, he can intuitively see the main attractions on the target travel route, and can also view the attraction preview screen by clicking the attraction tag, which facilitates the user to choose the travel route, further improving the user experience and satisfaction.

[0139] In step S1045 of some embodiments, multiple travel styles corresponding to the target travel route are determined based on the travel style information, thereby forming multiple travel style labels. In this way, when the target user views the recommended target travel route, he can intuitively see the travel style label of the target travel route, which is convenient for the user to choose the travel route, further improving the user experience and satisfaction.

[0140] In step S1046 of some embodiments, a travel route recommendation plan is generated based on the target travel route, attraction tags, and travel style tags, and the travel route recommendation plan is pushed to the target user. The target user can select a travel route based on information such as the attraction tags, travel style tags, and user preference scores of each target travel route.

[0141] In the embodiment of the present application, by generating attraction labels and travel style labels, the characteristics of the recommended target travel route can be displayed to the user more clearly and intuitively, making it easier for the user to understand the basis of the recommendation and choose a travel route, thereby improving the user's trust in the recommendation system and user experience.

[0142] It can be appreciated that the embodiment of the present application determines a preference feature vector based on the user's historical travel data, determines a travel route to be recommended and a corresponding destination feature vector based on the user's travel start and end point information, determines a user preference score for each travel route to be recommended based on the preference feature vector and the destination feature vector, and thus selects a target travel route for recommendation based on the user preference score, thereby providing the user with a travel route that meets the user's preferences and real-time needs, thereby improving the accuracy and diversity of travel route recommendations and also improving the user experience.

[0143] Compared with the prior art, the embodiments of the present application also have the following advantages:

[0144] 1) It can dynamically adjust the recommendation results according to the user's real-time needs and preferences, making the recommendations more personalized and in line with the user's personalized preferences and needs;

[0145] 2) It can clearly show the degree of attention paid by the model to different features during the recommendation process, increase the interpretability of the recommendation system, and make it easier for users to understand the basis of the recommendation results.

[0146] 3) The introduction of the attention mechanism makes the recommendation algorithm more flexible and adaptable. It can dynamically adjust the recommendation results according to different users and different situations to adapt to the ever-changing needs and preferences of users.

[0147] 4) The introduction of the attention mechanism can focus more on the features and information that users care about, thereby improving the accuracy of recommendations, reducing irrelevant recommendation results, and improving user satisfaction.

[0148] 5) Personalized recommendation results can improve user experience, making it easier for users to find travel routes that suit their interests and needs, thereby increasing users' trust in the recommendation system.

[0149] 6) By providing more personalized and accurate recommendation results, it can attract more users, improve user retention and conversion rates, and enhance the market competitiveness of the recommendation system.

[0150] See also Figure 8 The present application also provides a travel route recommendation device, which can implement the above travel route recommendation. The travel route recommendation device includes:

[0151] A preference feature vector determination module is used to obtain historical travel data of a target user and determine a preference feature vector of the target user based on the historical travel data;

[0152] The destination feature vector determination module is used to obtain the starting and ending point information of the target user's travel, determine multiple recommended travel routes based on the starting and ending point information, and determine the destination feature vector corresponding to each recommended travel route;

[0153] A user preference score determination module, used to determine the user preference score of each to-be-recommended travel route according to the preference feature vector and the destination feature vector;

[0154] The travel route screening module is used to screen several target travel routes from the travel routes to be recommended according to the user preference scores, and recommend the target travel routes to the target user.

[0155] The specific implementation of the travel route recommendation device is basically the same as the specific implementation of the above-mentioned travel route recommendation method, and will not be repeated here.

[0156] The embodiment of the present application also provides an electronic device, the electronic device comprising: a memory, a processor, a program stored in the memory and executable on the processor, and a data bus for realizing connection and communication between the processor and the memory, and the program is executed by the processor to realize the above-mentioned travel route recommendation method. The electronic device can be any intelligent terminal including a tablet computer, a car computer, etc.

[0157] See also Fig. 9 , Fig. 9 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0158] The processor 901 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0159] The memory 902 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 902 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 902, and the processor 901 calls and executes the travel route recommendation method of the embodiment of this application;

[0160] Input / output interface 903, used to implement information input and output;

[0161] Communication interface 904, used to realize communication interaction between the device and other devices, which can be realized by wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.);

[0162] A bus 905 that transmits information between various components of the device (e.g., the processor 901, the memory 902, the input / output interface 903, and the communication interface 904);

[0163] The processor 901 , the memory 902 , the input / output interface 903 and the communication interface 904 are connected to each other in communication within the device via a bus 905 .

[0164] See also Fig.10 The embodiment of the present application also provides a storage medium, which is a computer-readable storage medium used for computer-readable storage. The storage medium stores one or more programs 1001, and the one or more programs 1001 can be executed by one or more processors to implement the above-mentioned travel route recommendation method.

[0165] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0166] The travel route recommendation method and device, electronic device and storage medium provided in the embodiments of the present application obtain the historical travel data of the target user, determine the preference feature vector of the target user based on the historical travel data, then obtain the travel start and end point information of the target user, determine multiple travel routes to be recommended based on the travel start and end point information, and determine the destination feature vector corresponding to each travel route to be recommended, and then determine the user preference score of each travel route to be recommended based on the preference feature vector and the destination feature vector, and finally select and obtain several target travel routes from the travel routes to be recommended based on the user preference score, and recommend the target travel route to the target user. The embodiments of the present application determine the preference feature vector based on the historical travel data of the user, determine the travel route to be recommended and the corresponding destination feature vector based on the travel start and end point information of the user, and determine the user preference score of each travel route to be recommended based on the preference feature vector and the destination feature vector, so as to select the target travel route for recommendation based on the user preference score, and can provide the user with travel routes that meet the user's preferences and real-time needs, improve the accuracy and diversity of travel route recommendations, and also improve the user experience.

[0167] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0168] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0169] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0170] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0171] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0172] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0173] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units 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 interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0174] The units described above 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.

[0175] 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.

[0176] 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 storage medium. 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, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0177] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A travel route recommendation method, characterized in that: The method comprises: Acquire historical travel data of a target user, and determine a preference feature vector of the target user according to the historical travel data; Acquire the starting and ending point information of the target user's travel, determine a plurality of to-be-recommended travel routes according to the starting and ending point information of the travel, and determine a destination feature vector corresponding to each of the to-be-recommended travel routes; Determining a user preference score for each of the to-be-recommended travel routes according to the preference feature vector and the destination feature vector; A plurality of target travel routes are obtained by screening the travel routes to be recommended according to the user preference scores, and the target travel routes are recommended to the target user.

2. The travel route recommendation method according to claim 1, characterized in that: The determining the preference feature vector of the target user according to the historical travel data comprises: Determine multiple historical travel locations of the target user and corresponding travel times and stay durations according to the historical travel data; Determine a plurality of travel styles corresponding to each of the historical travel locations, and determine a location preference score for each of the historical travel locations based on the number of trips and the length of stay; Determining a style preference weight for each travel style according to the location preference score of the historical travel location corresponding to each travel style; The preference feature value of each travel style dimension is determined according to the style preference weight of each travel style, and the preference feature vector is generated according to the preference feature value of each travel style dimension.

3. The travel route recommendation method according to claim 1, characterized in that: The determining of a plurality of recommended travel routes according to the travel starting and ending point information includes at least one of the following: Acquire a preset travel route database, and match the travel start and end point information in the travel route database to obtain a plurality of travel routes to be recommended; The travel start and end point information is input into a pre-trained path planning model to obtain a plurality of travel routes to be recommended.

4. The travel route recommendation method according to claim 1, characterized in that: The determining of the destination feature vector corresponding to each of the to-be-recommended travel routes includes: Determining a plurality of intermediate destinations of the travel route to be recommended, and determining scenic spot information and travel style information of each of the intermediate destinations; Determining a location score of the intermediate destination according to the scenic spot information, and determining a travel style multidimensional vector of the intermediate destination according to the travel style information; Using the travel style multidimensional vector as a key vector and the place rating as a value vector to obtain a key-value vector pair of the intermediate destination; The destination feature vector is generated according to the key-value vector pair of each of the intermediate destinations.

5. The travel route recommendation method according to claim 1, characterized in that: The determining of the user preference score of each of the to-be-recommended travel routes according to the preference feature vector and the destination feature vector comprises: Determining key vectors and value vectors of multiple intermediate destinations of the to-be-recommended travel route according to the destination feature vector; Taking the preference feature vector as a query vector, calculating the similarity between the query vector and each of the key vectors, and determining the attention score of each of the corresponding intermediate destinations according to the similarity; Normalizing the attention scores to obtain attention weights of the intermediate destinations; The numerical vector is weightedly summed according to the attention weight to obtain the user preference score of the travel route to be recommended.

6. The travel route recommendation method according to claim 1, characterized in that: The step of selecting a plurality of target travel routes from the to-be-recommended travel routes according to the user preference scores includes at least one of the following: Comparing the user preference score with a preset first threshold, and if the user preference score is greater than or equal to the first threshold, determining the corresponding to-be-recommended travel route as the target travel route; The recommended travel routes are sorted from large to small according to the user preference scores, and a number of the recommended travel routes with higher sorting values ​​are selected as the target travel routes.

7. The travel route recommendation method according to any one of claims 1 to 6, characterized in that: The recommending the target travel route to the target user includes: Determining scenic spot information and travel style information of multiple intermediate destinations corresponding to the target travel route; Determine a plurality of scenic spot tags according to the scenic spot information, generate a scenic spot preview screen corresponding to each of the scenic spot tags, and then embed the scenic spot preview screen into the scenic spot tag; determining a plurality of travel style tags according to the travel style information; A travel route recommendation scheme is generated according to the target travel route, the scenic spot tag and the travel style tag, and the travel route recommendation scheme is pushed to the target user.

8. A travel route recommendation device, characterized in that: The device comprises: A preference feature vector determination module, used to obtain historical travel data of a target user, and determine a preference feature vector of the target user according to the historical travel data; A destination feature vector determination module is used to obtain the starting and ending point information of the target user's travel, determine a plurality of to-be-recommended travel routes according to the starting and ending point information of the travel, and determine the destination feature vector corresponding to each of the to-be-recommended travel routes; A user preference score determination module, used to determine the user preference score of each of the to-be-recommended travel routes according to the preference feature vector and the destination feature vector; The travel route screening module is used to screen a plurality of target travel routes from the travel routes to be recommended according to the user preference scores, and recommend the target travel routes to the target user.

9. An electronic device, characterized in that: The electronic device includes a memory, a processor, a program stored in the memory and executable on the processor, and a data bus for realizing connection and communication between the processor and the memory. When the program is executed by the processor, the steps of the travel route recommendation method as described in any one of claims 1 to 7 are realized.

10. A storage medium, the storage medium being a computer-readable storage medium, used for computer-readable storage, characterized in that: The storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the travel route recommendation method as described in any one of claims 1 to 7.

Images