Hotel preloading method and device based on user behaviors, medium and program product

Abstract

The invention discloses a hotel preloading method and device based on user behaviors, a medium and a program product. The method comprises the following steps: acquiring a hotel card list in a visual area of a user, and extracting feature data of each hotel in the visual area; loading an offline prediction model, inputting the extracted feature data of each hotel into the prediction model, outputting the click probability of each hotel by the prediction model, selecting the hotel with the maximum click probability, and recording the index of the hotel; and searching the target hotel in the visual area according to the index, and completing personalized hotel preloading based on the user behavior. According to the method, the hotel characteristics in the visual area of the user are extracted by taking the click behavior of the user as a reference, then the prediction model for the user is constructed by utilizing a multi-layer perceptron technology, the hotel click probability in the visual area of the user is output by the prediction model, and related contents are pre-loaded according to the probability value sequence, so that the personalized problem is effectively solved, and the user experience is improved. The hit rate is improved, resource waste is reduced, and user experience is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer deep learning, and in particular to a hotel preloading method, device, medium and program product based on user behavior. Background Art

[0002] With the development of artificial intelligence, many software can already predict people's choices and implement preloading functions. For example, in terms of people's choice of hotels, current mini programs generally adopt a fixed preloading strategy, that is, according to the configuration, hotels in the user's visible area are selected for data preloading. This is usually based on the assumption of user interface design, that is, users are more likely to pay attention to and click on hotels in these locations first.

[0003] This fixed preloading strategy has certain problems: (1) Lack of personalization. This preloading strategy is consistent for all users and does not take into account the user's personal preferences and historical behavior. Therefore, for users who prefer different locations or specific types of hotels, the preloaded content may not match their interests. (2) Low hit rate. Users may consider many factors when choosing a hotel, such as the hotel's price, rating, distance, etc. If preloading is performed based solely on the hotel's visible area location, the user may not be interested in the preloaded content, thereby reducing the hit rate and wasting a certain amount of bandwidth and computing resources. (3) Lack of dynamic adjustment. This preloading strategy ignores the context of the user's click and cannot update and optimize the preloading strategy based on real-time user behavior.

[0004] It can be seen that the existing preloading strategy for hotel selection is still insufficient and needs to be further improved. How to create a new hotel preloading method, device, medium and program product based on user behavior, so that it can calculate the click probability for each hotel in the user's current visual area based on the user's click behavior characteristics and use the prediction model built by multi-layer perceptron technology, sort the results according to the probability value, preload personalized related content, improve the hit rate, reduce user waiting time, and improve user experience, has become a goal that the industry urgently needs to improve. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a hotel preloading method based on user behavior, so that it can calculate the click probability for hotels in the user's current visual area based on the user's click behavior characteristics and use a prediction model constructed by multi-layer perceptron technology, sort the results according to the probability values, preload personalized related content, improve the hit rate, reduce user waiting time, and improve user experience, thereby overcoming the shortcomings of existing preloading strategies.

[0006] In order to solve the above technical problems, in a first aspect, the present invention provides a hotel preloading method based on user behavior, comprising the following steps:

[0007] Get the hotel card list in the user's visible area and extract the feature data of each hotel in the visible area;

[0008] Load the offline prediction model, input the extracted feature data of each hotel into the prediction model, calculate and output the click probability of each hotel by the prediction model, select the hotel with the highest click probability, and record its index;

[0009] The target hotel is searched in the visible area according to the index of the hotel with the highest click probability, and personalized hotel preloading based on the user behavior is completed.

[0010] As a further improvement, the feature data includes: basic features, which are for the user's i-th click, based on the user's previous i-1 click records, and extract one or more features of the hotel rating, minimum price, distance, appearance ratio in the historical click records and visible area position in the visible area at the i-th click, among which the rating, minimum price and distance features are fluctuation ratios calculated based on the user's historical click records.

[0011] As a further improvement, the feature data also includes: an interactive feature, which is a feature formed by the interaction of two basic features; and a relative feature, which is a relative feature between various hotels within the user's visible area.

[0012] As a further improvement, the preloading method also includes: each time the user completes a hotel card click, the feature data on the hotel card clicked is recorded in the cache, and when the next visible area hotel card click probability is predicted, the latest cache information is read, and feature data is extracted based on this information, thereby achieving dynamic adjustment of user click features and prediction model input features.

[0013] Further improvement, the construction process of the prediction model includes the following steps:

[0014] By adding embedding code, we can collect information about hotel cards in the user's visible area and the clicks of users every time an online user searches for a hotel.

[0015] The collected data is preprocessed to remove missing values ​​and outliers;

[0016] For the pre-processed data, the hotel data is grouped according to the user ID, and each user's group includes a set of clicked hotels and a set of unclicked hotels in the user's visible area;

[0017] For each user group, extract the characteristic data of each hotel in turn;

[0018] The data after feature extraction is divided into training set, validation set and test set. A multi-layer neural network is constructed based on the multi-layer perceptron, and dropout is added for model training. During the training process, the model hyperparameters are adjusted according to the verification results of the validation set and the test set to form an offline prediction model.

[0019] A further improvement is that, for the pre-processed data, the hotel data is grouped according to user ID, and each user's group includes a clicked hotel set and a non-clicked hotel set in the user's visible area. The step also includes: checking the non-clicked hotel set, and if there is a hotel in the user's historical click list, the hotel is removed from the non-clicked hotel set.

[0020] As a further improvement, the training set includes positive samples and negative samples. When constructing the positive samples of the training set, it is ensured that the ratio of the visible area position of the hotel card matches the actual click ratio of online users. The negative samples of the training set are randomly selected from the set of hotels that are not clicked in the user's visible area to ensure a balance in the number of positive and negative samples.

[0021] In a second aspect, an embodiment of the present invention further provides an electronic device, the electronic device comprising:

[0022] at least one processor; and,

[0023] a memory communicatively connected to the at least one processor; wherein,

[0024] The memory stores instructions that can be executed by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor implements the hotel preloading method based on user behavior as described in any one of the first aspect or any one of the implementations of the first aspect.

[0025] In the third aspect, an embodiment of the present invention further provides a non-transitory computer-readable storage medium, which stores computer instructions, and when the computer instructions are executed by at least one processor, the at least one processor executes the user behavior-based hotel preloading method in the aforementioned first aspect or any implementation of the first aspect.

[0026] In a fourth aspect, an embodiment of the present invention further provides a computer program product, comprising a computer program stored on a non-transitory computer-readable storage medium, wherein the computer program comprises program instructions, and when the program instructions are executed by a computer, the computer executes the user behavior-based hotel preloading method in the aforementioned first aspect or any implementation of the first aspect.

[0027] After adopting such a design, the present invention has at least the following advantages:

[0028] The preloading method based on user behavior of the present invention takes the user's clicking behavior as a benchmark, extracts the hotel features in the user's visual area, and then uses the multi-layer perceptron technology to build a prediction model for the user. The prediction model outputs the click probability of the hotel in the user's visual area, and preloads related content according to the probability value sorting. This can not only solve the problem of lack of personalized prediction, but also effectively improve the hit rate, reduce resource waste, reduce user waiting time, and improve user experience.

[0029] The hotel features of each user's click behavior are recorded in real time and cached. When predicting hotels in the next visible area, the feature input is dynamically adjusted according to the click context, which effectively solves the problem of lack of dynamic adjustment in the existing preloading method and makes the preloading method of the present invention more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The above is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0031] Figure 1 It is a flow logic diagram of the preloading method according to an embodiment of the present invention.

[0032] Figure 2 A schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0033] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0034] Refer to the attached Figure 1 As shown, the hotel preloading method based on user behavior in this embodiment includes a training phase and a prediction phase.

[0035] (I) The training phase is used to train and obtain an offline prediction model for the user. It specifically includes the following steps:

[0036] Data collection and preparation: By adding embedding codes, we collect hotel cards in the user's visible area and user clicks every time an online user searches for a hotel.

[0037] Data cleaning and preprocessing: The collected data is preprocessed to remove missing values ​​and outliers.

[0038] Feature extraction: For the preprocessed data, the hotel data is grouped according to the user ID. Each user's group includes a set of clicked hotels and a set of unclicked hotels in the user's visible area. For each user's group, the feature data of each hotel is extracted in turn.

[0039] It also includes the step of filtering easily confused data, specifically, checking the unclicked hotel set, if there is a hotel in the user's historical click list, then remove the hotel from the unclicked hotel set to prevent confusing the model effect.

[0040] In addition, the feature data extracted from each hotel has three types of features: (1) Basic features, which are based on the user's i-th click and the user's previous i-1 click records as the benchmark, and extract one or more features of the hotel rating, minimum price, distance, appearance ratio in the historical click records, and position in the visible area at the time of the i-th click, where the rating, minimum price, and distance features are the fluctuation ratios calculated based on the user's historical click records. (2) Interactive features, which are features formed by the interaction of two basic features. For example, the interactive feature of rating and price is price / rating, and this feature value can, to a certain extent, represent the user's consideration of the cost-effectiveness of the two features. This interactive feature can better understand the user's comprehensive preference for hotel features. (3) Relative features, which are relative features between hotels in the user's visible area. For example, the relative ranking of prices between hotels. The feature set composed of the above basic features, interactive features, and relative features can more effectively extract hotel features and better form a complete feature set for each hotel.

[0041] Model training: Divide the data after feature extraction into training set, validation set and test set, construct a multi-layer neural network based on MLP multi-layer perceptron, add dropout to prevent overfitting, and perform model training. During the training process, adjust the model hyperparameters according to the verification results of the validation set and test set to form an offline prediction model for users.

[0042] The training set includes positive samples and negative samples. When constructing the positive samples of the training set, it is ensured that the ratio of the visible area position of the hotel card matches the actual click ratio of online users. The negative samples of the training set are randomly selected from the set of hotels that are not clicked in the user's visible area to ensure a balance in the number of positive and negative samples.

[0043] (ii) The prediction stage is used to predict the user's hotel reservation based on the trained offline prediction model and complete preloading.

[0044] The preloading method in this embodiment takes the user using the WeChat applet to realize hotel reservation as an example, and describes in detail the hotel preloading method based on user behavior, which specifically includes the following steps:

[0045] S1. When it is detected that a user enters the WeChat applet, a list of hotel cards in the user's visible area is obtained, and feature data of each hotel in the visible area is extracted, including basic features, interactive features, and relative features.

[0046] S2. Create an AI reasoning session and load the trained offline prediction model;

[0047] S3. Convert the feature data of each hotel extracted in step S1 into the input format of the prediction model, input the feature data into the prediction model, calculate and output the click probability of each hotel by the prediction model, select the hotel with the highest click probability, and record its index;

[0048] S4. Searching for a target hotel in the visible area according to the index of the hotel with the highest click probability, and completing the personalized hotel preloading based on the user behavior.

[0049] S5. Each time a user completes a click on a hotel card, the feature data on the hotel card clicked is recorded in the cache. When predicting the probability of clicking on a hotel card in the visible area next time, the latest cache information is read and used as a basis for extracting feature data, thereby achieving dynamic adjustment of user click features and prediction model input features.

[0050] See attached Figure 2 As shown, an embodiment of the present invention further provides an electronic device 10, the electronic device 10 comprising:

[0051] at least one processor; and,

[0052] a memory communicatively connected to the at least one processor; wherein,

[0053] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the user behavior-based preloading method in the aforementioned method embodiment.

[0054] An embodiment of the present invention further provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable the computer to execute the user behavior-based preloading method in the aforementioned method embodiment.

[0055] An embodiment of the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer executes the user behavior-based preloading method in the aforementioned method embodiment.

[0056] Reference below Figure 2, which shows a schematic diagram of the structure of an electronic device 10 suitable for implementing an embodiment of the present invention. The electronic device in the embodiment of the present invention may include, but is not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 2 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0057] like Figure 2 As shown, the electronic device 10 may include a processing device (e.g., a central processing unit, a graphics processor, etc.) 101, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 102 or a program loaded from a storage device 108 to a random access memory (RAM) 103. In the RAM 103, various programs and data required for the operation of the electronic device 10 are also stored. The processing device 101, the ROM 102, and the RAM 103 are connected to each other via a bus 104. An input / output (I / O) interface 105 is also connected to the bus 104.

[0058] Typically, the following devices may be connected to the I / O interface 105: an input device 106 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 107 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 108 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 109. The communication device 109 may allow the electronic device 10 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 2 1 shows an electronic device 10 having various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0059] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 109, or installed from the storage device 108, or installed from the ROM 102. When the computer program is executed by the processing device 101, the above-mentioned functions defined in the method of the embodiment of the present invention are executed.

[0060] It should be noted that the computer-readable medium of the present invention can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal can take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0061] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0062] The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: obtains at least two Internet Protocol addresses; sends a node evaluation request including the at least two Internet Protocol addresses to a node evaluation device, wherein the node evaluation device selects an Internet Protocol address from the at least two Internet Protocol addresses and returns it; receives the Internet Protocol address returned by the node evaluation device; wherein the obtained Internet Protocol address indicates an edge node in a content distribution network.

[0063] Alternatively, the computer-readable medium carries one or more programs, and when the one or more programs are executed by the electronic device, the electronic device: receives a node evaluation request including at least two Internet Protocol addresses; selects an Internet Protocol address from the at least two Internet Protocol addresses; and returns the selected Internet Protocol address; wherein the received Internet Protocol address indicates an edge node in a content distribution network.

[0064] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as python, Java, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0065] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0066] The units involved in the embodiments of the present invention may be implemented by software or hardware, wherein the name of a unit does not limit the unit itself in some cases.

[0067] It should be noted that the above embodiments and features therein may be combined with each other if there is no conflict.

[0068] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A hotel preloading method based on user behavior, characterized in that: The following steps are involved: Get the hotel card list in the user's visible area and extract the feature data of each hotel in the visible area; Load the offline prediction model, input the extracted feature data of each hotel into the prediction model, calculate and output the click probability of each hotel by the prediction model, select the hotel with the highest click probability, and record its index; The target hotel is searched in the visible area according to the index of the hotel with the highest click probability, and personalized hotel preloading based on the user behavior is completed.

2. The hotel preloading method based on user behavior according to claim 1 is characterized in that: The characteristic data includes: The basic features are, for the i-th click of the user, based on the user's previous i-1 click records, and one or more features of the hotel rating, minimum price, distance, appearance ratio in the historical click records and position in the visible area at the i-th click are extracted, among which the rating, minimum price and distance features are fluctuation ratios calculated based on the user's historical click records.

3. The hotel preloading method based on user behavior according to claim 2 is characterized in that: The characteristic data also includes: Interactive features are features composed of the interaction of two basic features; Relative characteristics are the relative characteristics between various hotels within the user's visible area.

4. The hotel preloading method based on user behavior according to claim 3 is characterized in that: The preloading method also includes: each time a user completes a hotel card click, the feature data on the hotel card clicked is recorded in a cache, and when predicting the probability of clicking a hotel card in the visible area next time, the latest cache information is read and used as a reference to extract feature data, thereby achieving dynamic adjustment of user click features and prediction model input features.

5. The hotel preloading method based on user behavior according to any one of claims 1 to 4, characterized in that: The construction process of the prediction model includes the following steps: By adding embedding code, we can collect information about hotel cards in the user's visible area and the clicks of users every time an online user searches for a hotel. The collected data is preprocessed to remove missing values ​​and outliers; For the pre-processed data, the hotel data is grouped according to the user ID, and each user's group includes a set of clicked hotels and a set of unclicked hotels in the user's visible area; For each user group, extract the characteristic data of each hotel in turn; The data after feature extraction is divided into training set, validation set and test set. A multi-layer neural network is constructed based on the multi-layer perceptron, and dropout is added for model training. During the training process, the model hyperparameters are adjusted according to the verification results of the validation set and the test set to form an offline prediction model.

6. The hotel preloading method based on user behavior according to claim 5 is characterized in that: The step of grouping the hotel data according to the user ID for the pre-processed data, wherein each user's group includes a set of clicked hotels and a set of unclicked hotels in the user's visible area, further includes: The unclicked hotel set is checked, and if there is a hotel in the user's historical click list, the hotel is removed from the unclicked hotel set.

7. The hotel preloading method based on user behavior according to claim 5 is characterized in that: The training set includes positive samples and negative samples. When constructing the positive samples of the training set, it is ensured that the position ratio of the hotel card visible area matches the actual click ratio of online users. The negative samples of the training set are randomly selected from the set of hotels that are not clicked in the user's visible area to ensure a balance in the number of positive samples and negative samples.

8. An electronic device, characterized in that: The electronic device includes: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor executes the hotel preloading method based on user behavior as described in any one of claims 1 to 7.

9. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium stores computer instructions, which, when executed by at least one processor, cause the at least one processor to perform the hotel preloading method based on user behavior as described in any one of claims 1 to 7.

10. A computer program product, characterized in that The computer program product comprises a computer program / instruction, which, when executed by a processor, implements the hotel preloading method based on user behavior as described in any one of claims 1 to 7.

Images