Information recommendation method and device and storage medium
By determining the visit time of the user's visit to the destination, determining the recommendation time and the object to be recommended based on the time, and sending recommendation information to the user terminal, the problem of difficulty in accurately recommending travel products such as hotels in the prior art is solved, and the search efficiency is improved.
Patent Information
- Application Number
- CN202311264173.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-09-26
- Publication Date
- 2025-03-28
AI Technical Summary
The prior art is difficult to accurately recommend travel products such as hotels to users, resulting in inefficient searches.
By determining the visit time of the target user visiting the destination, determining the recommendation time based on the time, and determining the object to be recommended based on the recommendation time, and sending the recommendation information to the terminal device of the target user.
It has achieved targeted service recommendations for target users, improved the search efficiency of travel products such as hotels, and solved the problem of accurate recommendations.
Smart Images
Figure CN119719518A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of computing service technology, and in particular to an information recommendation method, device and storage medium. Background Art
[0002] With the rapid development of Internet technology, its application is becoming more and more extensive. Internet technology is not only limited to work, but also has a significant impact on people's life and study. For example, in life, transportation is becoming more and more convenient, and people are traveling more and more. How to use Internet technology to accurately recommend hotels and other travel products to travel users has become a technical problem that needs to be solved urgently. Summary of the invention
[0003] In order to solve the above technical problems, the present application hopes to provide an information recommendation method, device and storage medium, which solves the problem that hotels and other travel products cannot be accurately recommended to users at present, and proposes a method for reasonably recommending hotels and other travel products to users based on user behavior, thereby improving the search efficiency of hotels and other travel products.
[0004] The technical solution of this application is implemented as follows:
[0005] In a first aspect, an information recommendation method is provided, the method comprising:
[0006] Determine the target user's visit time to the current destination;
[0007] If it is determined that a service recommendation of a target service is required for the target user, determining a recommended time for recommending the target service to the target user based on the visit time;
[0008] Based on the recommendation time, determining at least one object to be recommended; wherein the at least one object to be recommended is used to provide the target service;
[0009] When the time is the recommendation time, recommendation information including at least one of the objects to be recommended is sent to the target terminal device of the target user.
[0010] Optionally, if it is determined that a service recommendation of a target service is required for the target user, before determining a recommendation time for recommending the target service to the target user, the method further includes:
[0011] Obtaining a target prediction model;
[0012] Acquire current behavior characteristic parameters of the target user when visiting the current destination and current service characteristic parameters of the current destination based on the target service;
[0013] The target prediction model is used to predict user behavior of the current behavior characteristic parameters and the current service characteristic parameters to obtain a prediction result; wherein the prediction result is a service recommendation for determining whether the target service needs to be provided to the target user, or a service recommendation for determining whether the target service does not need to be provided to the target user.
[0014] Optionally, if it is determined that a service recommendation of a target service is required for the target user, determining a recommendation time for the target user based on the visit time includes:
[0015] If it is determined that a service recommendation of a target service is required for the target user, determining a first preset time range to which the visit time belongs;
[0016] Determine the historical duration from reaching the current destination to booking the target service within the first preset time range;
[0017] The sum of the visit time and the historical duration is calculated to obtain the recommended time.
[0018] Optionally, determining the historical duration from arriving at the current destination to booking the target service within the first preset time range includes:
[0019] Obtain a reference time set of the target user's arrival at a historical destination within a preset historical time range and the target service within the first preset time range;
[0020] An average value is calculated for the durations in the reference duration set to obtain the historical duration.
[0021] Optionally, determining at least one object to be recommended based on the recommendation time includes:
[0022] When the time is the recommended time, obtaining the current location of the target user;
[0023] If the current location is within the current destination range, determine at least one service provider that provides the target service within a preset range centered on the current location;
[0024] Determine the distance between the current position and each of the service providing objects to obtain at least one target distance;
[0025] Obtaining a service usage rate of the target service of each of the service providing objects;
[0026] Determine the historical average price and the current average price of each of the service provision objects within the second preset time range;
[0027] Determine an evaluation score for each of the service providing objects based on the target distance, the service usage rate, the historical average price and the current average price of each of the service providing objects;
[0028] Based on the evaluation score of at least one of the service providing objects, at least one of the objects to be recommended is determined.
[0029] Optionally, determining the evaluation score of each service provider based on the target distance, the service usage rate, the historical average price and the current average price of each service provider includes:
[0030] Calculating the absolute value of the difference between the historical average price and the current average price of each of the service provision objects to obtain a first value of each of the service provision objects;
[0031] Calculating a preset square root of the target distance of each of the service providing objects to obtain a second value of each of the service providing objects;
[0032] Calculate the product of the first value and the second value of each of the service provision objects to obtain a third value of each of the service provision objects;
[0033] The ratio of the service usage rate of each of the service providing objects to the third value is calculated to obtain an evaluation score of each of the service providing objects.
[0034] Optionally, when the time is the recommendation time, sending recommendation information including at least one of the to-be-recommended objects to the target terminal device of the target user includes:
[0035] When the time is the recommendation time, obtaining at least one access link of the object to be recommended;
[0036] Sorting the access links of at least one of the objects to be recommended in descending order according to the evaluation scores of the objects to be recommended to obtain a target sorting object set;
[0037] Generate the recommendation information based on the target sorting object set;
[0038] The recommendation information is sent to the target terminal device.
[0039] Optionally, the service characteristic parameters include current environment parameters of the destination and historical service usage characteristic parameters of the target service.
[0040] Optionally, the acquiring the target prediction model includes:
[0041] Determine the prediction model to be trained;
[0042] Obtain historical sample data of historical users about the target service;
[0043] Constructing training sample data based on the historical sample data;
[0044] The training sample data is used to perform model training on the prediction model to be trained to obtain the target prediction model.
[0045] In a second aspect, an information recommendation device includes at least: a memory, a processor, and a communication bus; wherein:
[0046] The memory is used to store executable instructions;
[0047] The communication bus is used to realize the communication connection between the processor and the memory;
[0048] The processor is used to execute the information recommendation program stored in the memory to implement the steps of any one of the above-mentioned information recommendation methods.
[0049] In a third aspect, a storage medium stores an information recommendation program, and when the information recommendation program is executed by a processor, the steps of any of the above-mentioned information recommendation methods are implemented.
[0050] The embodiment of the present application provides an information recommendation method, device and storage medium. After the target user's visit time to the current destination is determined by the information recommendation device, if it is determined that a service recommendation of a target service is required for the target user, the recommended time for recommending the target service to the target user is determined based on the visit time, and after at least one object to be recommended is determined based on the recommended time, when the time is the recommended time, recommendation information including at least one object to be recommended is sent to the target terminal device of the target user. In this way, when it is determined that a service recommendation of the target service is required for the target user, the recommended time for recommending the target service is first determined based on the time when the target user visits the destination, and then at least one object to be recommended is determined based on the recommended time, and finally when the time is the recommended time, at least one object to be recommended is sent to the target terminal device, so that targeted service recommendations are made to the target user, which is more personalized, solves the problem that the current problem of not being able to accurately recommend hotels and other travel products to users is solved, and a method for reasonably recommending hotels and other travel products to users based on user behavior is proposed, which improves the search efficiency of hotels and other travel products. BRIEF DESCRIPTION OF THE DRAWINGS
[0051] Figure 1 Schematic diagram of the process of the information recommendation method provided in the embodiment of the present application Figure 1 ;
[0052] Figure 2 Schematic diagram of the process of the information recommendation method provided in the embodiment of the present application Figure 2 ;
[0053] Figure 3 A schematic diagram of the structure of an information recommendation device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0054] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.
[0055] The embodiment of the present application provides an information recommendation method, referring to Figure 1 As shown, the method is applied to an information recommendation device, and the method comprises the following steps:
[0056] Step 101: Determine the arrival time of the target user at the current destination.
[0057] In an embodiment of the present application, the information recommendation device may be a third-party service device for providing information recommendation services, such as a server or other computing device with computing capabilities. The target user is a user who has currently arrived at the current destination and may need the target service. The information recommendation device may determine the time when the target user visits the current destination based on the current motion trajectory of the target user.
[0058] Step 102: If it is determined that a service recommendation of a target service is required for the target user, a recommendation time for recommending the target service to the target user is determined based on the visit time.
[0059] In an embodiment of the present application, when the target user is analyzed and it is determined that the target user needs to use the target service and needs service promotion for the target service, the information recommendation device determines the best recommendation time for recommending the target service to the target user based on the target user's visit time.
[0060] Step 103: Determine at least one object to be recommended based on the recommendation time.
[0061] Wherein, at least one object to be recommended is used to provide a target service.
[0062] In an embodiment of the present application, the information recommendation device performs recommendation analysis according to the recommendation time to determine at least one object to be recommended to the target user.
[0063] Step 104: When the time is the recommendation time, send recommendation information including at least one object to be recommended to the target terminal device of the target user.
[0064] In an embodiment of the present application, when the time is the recommended time, the information recommendation device can use a pre-set target communication method to send at least one object to be recommended to the target terminal device. In this way, the target user can determine an object from at least one object to be recommended to obtain the target service according to actual needs, reducing the search time that the target user needs to search from a large number of objects providing target services, improving the user experience, and improving the search efficiency of objects providing target services.
[0065] The embodiment of the present application provides an information recommendation method. After determining the target user's visit time to the current destination through an information recommendation device, if it is determined that a service recommendation of a target service is required for the target user, a recommended time for recommending the target service to the target user is determined based on the visit time, and after determining at least one object to be recommended based on the recommended time, when the time is the recommended time, recommendation information including at least one object to be recommended is sent to the target terminal device of the target user. In this way, when determining that a service recommendation of the target service is required for the target user, firstly the recommended time for recommending the target service is determined based on the target user's visit time to the destination, and then at least one object to be recommended is determined based on the recommended time, and finally when the time is the recommended time, at least one object to be recommended is sent to the target terminal device, so that targeted service recommendations are made to the target user, which is more personalized, solves the problem that currently the hotel and other travel products cannot be accurately recommended to the user, and proposes a method for reasonably recommending hotels and other travel products to the user based on the user's behavior, thereby improving the search efficiency of hotels and other travel products.
[0066] Based on the above embodiments, the embodiments of the present application provide an information recommendation method, referring to Figure 2 As shown, the method is applied to an information recommendation device, and the method comprises the following steps:
[0067] Step 201: Determine the arrival time of the target user at the current destination.
[0068] In the embodiment of the present application, the target service is a hotel service, and the information recommendation device is used to recommend a hotel to the user as an example for explanation. The current destination is City A, which the user will currently arrive at. At this time, the information recommendation device determines the arrival time of the target user to City A. In some application scenarios, the arrival time can be obtained from the target terminal device of the target user, and the target terminal device can determine the time when the location positioning system of the target terminal device first locates the user in City A as the arrival time; in some other application scenarios, if the target terminal device has functions such as mobile communication, the time when the target terminal device first connects to the mobile communication signal base station belonging to City A can be determined as the arrival time. In some application scenarios, the target service can also be other services provided to users, such as other scheduled services that can be booked, such as scenic spot reservations.
[0069] Step 202: Obtain a target prediction model.
[0070] In an embodiment of the present application, the target prediction model is a prediction model for predicting user behavior that is pre-trained using a model training method. For example, it can be a prediction model for predicting whether a user needs to book a hotel based on the user's behavior.
[0071] Step 203: Acquire current behavior characteristic parameters of the target user when visiting the current destination and current service characteristic parameters of the current destination based on the target service.
[0072] The service characteristic parameters include the current environment parameters of the destination and the historical service usage characteristic parameters of the target service.
[0073] In the embodiment of the present application, the current behavior characteristic parameters of the target user when visiting the current destination may include at least time characteristic parameters of the visit time, characteristic parameters of the relationship between the target user and the current destination, and may even include base station location characteristic parameters of the communication base station to which the target terminal device of the target user is connected. The current environment parameters of the destination may be, for example, weather condition parameters of the destination. The historical service usage characteristic parameters of the target service may be, for example, relevant statistical parameters obtained by providing historical usage of the target service.
[0074] The current behavior characteristic parameters of the target user when visiting the current destination can be obtained from the operator server corresponding to the target terminal device. The current service characteristic parameters based on the target service at the current destination can be some parameters when providing the target service at the current destination, such as the distribution of objects providing the target service, the price of each object, the user experience of each object, the user usage time of each object, etc. The current service characteristic parameters based on the target service can be obtained from the service platform providing the target service.
[0075] Step 204: Use the target prediction model to predict user behavior based on the current behavior characteristic parameters and the current service characteristic parameters to obtain a prediction result.
[0076] The prediction result is to determine whether the target service needs to be recommended to the target user, or to determine whether the target service does not need to be recommended to the target user.
[0077] In an embodiment of the present application, the current behavior characteristic parameters and current service characteristic parameters of the target user are input into the target prediction model, and the current behavior characteristic parameters and current service characteristic parameters are calculated by the target prediction model to implement prediction analysis and obtain prediction results. The prediction result is usually a service recommendation that determines that the target service needs to be provided to the target user or a service recommendation that determines that the target service does not need to be provided to the target user. For example, the first identification result output by the target prediction model, such as 1, can be used to identify the service recommendation that determines that the target service needs to be provided to the user, and the second identification result output by the target prediction model, such as 0, can be used to identify the service recommendation that does not need to be provided to the user.
[0078] It should be noted that step 201 and steps 202 to 204 can be executed simultaneously, steps 202 to 204 can be executed before step 201, and step 201 can be executed before any step before step 204. The specific execution order can be determined by actual conditions and is not specifically limited here.
[0079] Step 205: If it is determined that a service recommendation of a target service is required for the target user, a recommendation time for recommending the target service to the target user is determined based on the visit time.
[0080] In the embodiment of the present application, when the information recommendation device determines that the prediction result is to determine that the target service needs to be recommended to the target user, the information recommendation device can analyze the target user's visit time to determine the best recommended time for the target service recommendation. In some application scenarios, the information recommendation device may determine that the target service needs to be recommended to the target user not through the target prediction model, but based on the user's behavior, such as the user searching for the object providing the target service on the application that provides the reservation target service.
[0081] Step 206: Determine at least one object to be recommended based on the recommendation time.
[0082] Wherein, at least one object to be recommended is used to provide a target service.
[0083] In the embodiment of the present application, the information recommendation device analyzes the recommendation time and determines at least one object to be recommended for the target user.
[0084] Step 207: When the time is the recommendation time, send recommendation information including at least one object to be recommended to the target terminal device of the target user.
[0085] In an embodiment of the present application, when the time is the recommended time, the information recommendation device uses a pre-agreed target communication method to send recommendation information including at least one object to be recommended to the target terminal device, and the target terminal device can display the received at least one object to be recommended in the corresponding display area, so that the target user can view and understand the at least one object to be recommended, and select the object to be recommended that he or she likes to make a reservation. The recommendation information of at least one object to be recommended includes at least an access link for each object to be recommended, so that the user can quickly access each object to be recommended, thus increasing the access efficiency of at least one object to be recommended that provides the target service, improving the reservation rate of each object to be recommended, reducing the time spent by the user in searching for the object that provides the target service, and providing the user's usage experience effect.
[0086] Based on the above embodiments, in other embodiments of the present application, step 205 can be implemented by steps 205a to 205c:
[0087] Step 205a: If it is determined that a service recommendation of a target service is required for the target user, a first preset time range to which the visit time belongs is determined.
[0088] In the embodiment of the present application, a day can be divided into different time periods. In this way, when determining the need to recommend a target service to a target user, the first preset time range to which the visit time belongs is determined.
[0089] Step 205b: Determine the historical duration from the current destination to the predetermined target service within the first preset time range.
[0090] In an embodiment of the present application, a historical duration set is obtained from when each user arrives at the current destination within a first preset time range to the predetermined target service within a period of time, and then the duration elements in the obtained historical duration set are analyzed to determine the historical duration.
[0091] Exemplarily, assuming that the target user arrives at the current destination within the first preset time range of 13:00-14:00, obtain the duration from the arrival time to the hotel booking time of each user who arrives at the current destination within the first preset time range of 13:00-14:00 within a month, and then analyze the corresponding duration of each user within this month to determine the historical duration.
[0092] Step 205c: Calculate the sum of the visit time and the historical duration to obtain the recommended time.
[0093] In the embodiment of the present application, the time after the visit time after the historical duration is determined as the recommended time.
[0094] Based on the above embodiment, in other embodiments of the present application, step 205b can be implemented by steps a11 to a112:
[0095] Step a11: Obtain a reference time set of the target user arriving at a historical destination to a predetermined target service within a preset historical time range, within a first preset time range.
[0096] In the embodiment of the present application, the preset real-time time range is a pre-set time period, for example, it may be a year, a quarter, a month or a week.
[0097] Step a12: Calculate the average value of the durations in the reference duration set to obtain the historical duration.
[0098] In an embodiment of the present application, the sum of the durations in the reference duration set is calculated, the number of elements in the reference duration set is counted, and the ratio of the sum to the number is calculated to obtain the historical duration.
[0099] Based on the above embodiment, in other embodiments of the present application, step 206 can be implemented by steps 206a to 206g:
[0100] Step 206a: When the time is the recommended time, obtain the current location of the target user.
[0101] In an embodiment of the present application, when the current time is the recommended time, the current location of the target user collected by the target terminal device can be sent to the information recommendation device, so that the information recommendation device can determine the current location of the target user. In some application scenarios, the target information recommendation device can also determine the current location of the target user by determining the base station to which the target terminal device is connected at the recommended time.
[0102] Step 206b: If the current location is within the current destination range, determine at least one service provider that provides the target service within a preset range centered on the current location.
[0103] In an embodiment of the present application, the preset range centered on the current location can determine a range centered on the current location with a preset distance as the radius, or a preset number of street ranges centered on the current location. According to the current location of the target user, determine whether the target user is currently still within the geographical scope of the current destination. If the target user is not currently within the geographical scope of the current destination, there is no need to recommend the target user. If the target user is still within the geographical scope of the current destination, determine at least one service provider object that can provide the target service within a preset range with the current location as the center and a preset distance as the radius. Among them, the at least one service provider object can be all objects that provide the target service within the preset range, or it can be an object obtained by filtering all objects that provide the target service within the preset range according to the user's historical usage preferences.
[0104] Step 206c: determine the distance between the current location and each service provider to obtain at least one target distance.
[0105] In the embodiment of the present application, a globe coordinate calculation formula may be used to determine the distance between the current location and each service providing object, thereby obtaining at least one target distance.
[0106] Step 206d: Obtain the service usage rate of the target service of each service provider.
[0107] In the embodiment of the present application, illustratively, for hotel services, the service utilization rate of the hotel determined is the user occupancy rate.
[0108] Step 206e: Determine the historical average price and current average price of each service provider within the second preset time range.
[0109] In the embodiment of the present application, the second preset time range may be an empirical time range obtained based on a large number of experiments, and in actual application, it may be continuously adjusted according to actual conditions.
[0110] Step 206f: Determine the evaluation score of each service provider based on the target distance, service usage rate, historical average price and current average price of each service provider.
[0111] In an embodiment of the present application, the information recommendation device uses a preset score calculation formula to calculate the target distance, service usage rate, historical average price and current average price of each service object to obtain an evaluation score for each service provider.
[0112] Step 206g: Determine at least one object to be recommended based on the evaluation score of at least one service providing object.
[0113] In the embodiment of the present application, the evaluation score of at least one service providing object is analyzed to determine at least one object to be recommended. Generally, the higher the score, the more suitable the recommended object is for the target user.
[0114] Based on the above embodiment, in other embodiments of the present application, step 206f can be implemented by steps b11 to b14:
[0115] Step b11: Calculate the absolute value of the difference between the historical average price and the current average price of each service provider to obtain a first value of each service provider.
[0116] In the embodiment of the present application, the first value of each service provision object=|the historical average price of each service provision object−the current average price of each service provision object|, where || is an absolute value calculation symbol.
[0117] Step b12: Calculate the preset power root of the target distance of each service provider to obtain a second value of each service provider.
[0118] In the embodiment of the present application, the preset power can be set according to actual conditions, for example, the value can be 2, or 3, etc.
[0119] Step b13: Calculate the product of the first value and the second value of each service provision object to obtain a third value of each service provision object.
[0120] Step b14: Calculate the ratio of the service usage rate of each service provider to the third value to obtain an evaluation score of each service provider.
[0121] In the embodiment of the present application, the evaluation score of each service provider = the service usage rate of each service provider / the third value of each service provider.
[0122] Based on the above embodiments, in other embodiments of the present application, step 207 can be implemented by steps 207a to 207d:
[0123] Step 207a: When the time is the recommendation time, obtain an access link of at least one object to be recommended.
[0124] In the embodiment of the present application, the access link can access a predetermined interface of the object to be recommended. In some application scenarios, the predetermined interface can also include the object to be recommended.
[0125] Step 207b: sort the access links of at least one object to be recommended in descending order according to the evaluation scores of the object to be recommended, to obtain a target sorting object set.
[0126] In an embodiment of the present application, access links of at least one object to be recommended are sorted, and the sorting rule is a sorting order from high to low evaluation scores, so that a target sorting object set can be obtained.
[0127] Step 207c: Generate recommendation information based on the target sorting object set.
[0128] In an embodiment of the present application, recommendation information including a target ranking object set is generated.
[0129] Step 207d: Send the recommendation information to the target terminal device.
[0130] In an embodiment of the present application, information interaction can be achieved between the information recommendation device and the target terminal device through a target communication method, wherein the target communication method can be a fifth generation mobile communication technology (5th Generation Mobile Communication Technology, 5G) message communication method.
[0131] Based on the above embodiments, in other embodiments of the present application, step 202 can be implemented by steps 202a to 202d:
[0132] Step 202a: Determine the prediction model to be trained.
[0133] In an embodiment of the present application, the prediction model to be trained may be a classification model, for example, an iterative decision tree model (Gradient Boosting Decision Tree, GBDT).
[0134] Step 202b: Obtain historical sample data of historical users regarding the target service.
[0135] In an embodiment of the present application, historical sample data related to the target service is obtained when historical users preview the target service within a historical time range. The historical sample data includes historical behavior characteristic parameters and historical service characteristic parameters, as well as result data on whether the end user has subscribed to the target service.
[0136] Step 202c: construct training sample data based on historical sample data.
[0137] In an embodiment of the present application, data analysis is performed on the acquired historical sample data, for example, incorrect data is extracted, and the historical sample data is sorted according to preset data sorting rules, etc., to obtain training sample data.
[0138] Step 202d: Use the training sample data to perform model training on the prediction model to be trained to obtain a target prediction model.
[0139] In an embodiment of the present application, the constructed training sample data is used to perform model training on the prediction model to be trained to obtain a target prediction model.
[0140] It should be noted that the embodiments corresponding to steps 202a to 202d can be executed as a separate embodiment. In this way, after the target prediction model is trained, the target prediction model can be stored for subsequent use. In some application scenarios, the target prediction model can be continuously improved through model training as the number of samples increases, to ensure that the target prediction model finally trained is more accurate.
[0141] Based on the above embodiments, the present application provides an information recommendation method, and the specific implementation process of the method can refer to the following steps:
[0142] Step c11: construct training sample data.
[0143] The construction of training sample data includes the following steps:
[0144] Step c111: Obtain sample data.
[0145] When a user is about to visit any city, for example, city C, he can obtain operator data provided by at least one telecom operator and platform service data of an online travel platform.
[0146] Among them, the operator data may include at least the following data: whether it is the weekend when the user first visits, the time period when the user first visits, whether it is a holiday when the user visits, the user's age, the user's gender, whether the user's daily residence is in the same province, whether the user's weekend residence is in the same province, whether the user's workplace is in the same province, the number of user visits, the user's real-time latitude and longitude, base station identity information (Identity document, ID), the city to which the base station belongs, the latitude and longitude of the base station, the user's real-time base station connection ID, etc., among which the time period can be divided into 00:00-6:00, 6:00-12:00, 12:00-18:00 and 18:00-24:00.
[0147] The platform service data includes at least the following data: whether the user has booked a hotel on the day of visit, the hotel's latitude and longitude, the hotel's average daily room price, the average room booking price of each platform user, the average age of hotel customers, the average number of days of stay of hotel customers per stay, the average number of days of stay of each platform user per stay, the average historical occupancy rate of each hotel, etc.
[0148] The platform service data may also include weather data, such as whether it is raining in city C and related data such as temperature and humidity.
[0149] Step c112: construct a training data set.
[0150] The data in step c111 is extracted and combined to obtain the feature fields required for subsequent model training, for example, including whether it is a weekend when the user first arrives, the time period when the user first arrives, whether it is a holiday when the user visits, the user's age, the user's gender, whether the user's work and daily residence is in the same province, whether the user's weekend residence is in the same province, whether the user's work place is in the same province, whether it rains when the user first arrives, the temperature when the user first arrives, and the number of times the user visits. At the same time, the result data of whether the user booked a hotel on the day of the visit is obtained, and the result data is used as the label field.
[0151] Specifically, a feature list can be constructed for each user according to the above 11 feature fields. For any user j, the feature list can be recorded as [a1, a2, …, a 11 ], where a1 represents whether it is the weekend when the user first visits, and the corresponding value of a1 is 1 or 0, 1 represents the weekend, and 0 represents not the weekend; a2 represents the time period when the user first visits, and the corresponding value of a2 can be one of the following values: [1,0,0,0], [0,1,0,0], [0,0,1,0], and [0,0,0,1], where [1,0,0,0] represents the time period from 00:00 to 6:00, [0,1,0,0] represents the time period from 6:00 to 12:00, [0,0,1,0] represents the time period from 12:00 to 18:00, and [1,0,0,0] represents the time period from 18:00 to 24:00. The remaining a3 to a4 are the numerically represented values corresponding to the remaining feature fields in the above 11 feature fields.
[0152] Each constructed feature list will correspond to a label field, which is the field indicating whether the user has booked a hotel on the day of the visit. The label field is numerically represented by 1 if the user has booked a hotel on the day of the visit, and 0 if the user has not booked a hotel on the day of the visit.
[0153] In this way, the constructed feature list and the corresponding label fields constitute the training sample data in the training data set of the user hotel reservation demand prediction model.
[0154] Step c12: Use the training sample data to train the user hotel reservation demand prediction model to obtain a target prediction model.
[0155] Among them, the user hotel reservation demand prediction model can be a GBDT classification model. The training sample data constructed in step c11 is input into the GBDT classification model for model training to obtain the optimal model, namely the target prediction model, and the final target prediction model is saved in a model persistence manner. The target prediction model can be saved in .pkl file format.
[0156] Step c13: Use the target prediction model to predict the user's behavior and obtain the prediction result.
[0157] The data related to the data type of the user as shown in step c11 is obtained to obtain the sample to be predicted, and the sample to be predicted is input into the target prediction model to obtain the prediction result corresponding to the user.
[0158] Step c14: when the prediction result is that the user needs to book a hotel service, determine the best recommended time node for the user.
[0159] First, we count the hotel booking time points of users who visited city C in the past month. Since users arrive at different time points, the urgency of hotel booking is different. Therefore, we can count the hotel booking status of users who arrive at different time points separately. Among them, we can divide a day into 24 hours, and count the users who book a hotel after arriving in city C within the time period [0,1], the time period [1,2], the time period [2,3]..., the time period [23,24], and count how long these users booked the hotel after their visit. At this time, the duration can be accurate to minutes.
[0160] Then count the average time it takes for users who arrived in the time period [0,1], [1,2], [2,3], ..., [23,24] to make a hotel reservation. For example, there are n users who visited and made a hotel reservation in the time period [0,1] in the past month. The time from their arrival to the hotel reservation is t1, t2, t3, ..., t n minutes, then the average time from arrival to hotel booking for users who visit within [0,1] can be calculated by the following formula: In this way, the average duration in other time periods can be obtained in the same way.
[0161] Finally, determine the target time period to which the user's current access time belongs, determine the target average duration corresponding to the target time period, and use the time corresponding to the target average duration after the current access time as the best recommended time node.
[0162] Step c15: Rate and recommend hotels.
[0163] When the time is the best recommended time node, the user's current location is obtained, and the longitude and latitude coordinates can be recorded as (x i ,y i ), get the location of all hotels around the current location (x j ,y j ), then the distance from user i to the hotel is D ijThe calculation method is as follows: First, convert the position coordinates into radian coordinates
[0164]
[0165] Secondly, by formula D ij =R*arcos[cos(Y i )*cos(Y j )*cos(X i -X j )+sin(Y i )*sin(Y j )] Calculate the distance between user i and each hotel, where R = 6371 km, which is the radius of the earth.
[0166] Then, by scoring each hotel, the score calculation formula can be: v j is the average historical occupancy rate of hotel j. It can be seen that v j The larger the value is, the more popular the hotel is, and the higher its recommendation order score will be; D ij is the distance between user i and hotel j at this time. It can be seen that the closer the hotel is to the user, the higher the recommendation order score of the hotel for the user will be; i is the historical average hotel booking price for user i, p j is the average room price of hotel j on the day the user visits. It can be seen that p i With p j The smaller the difference is, the better the recommended order score f is. ij Will be higher.
[0167] After calculating the hotel's score, it is determined whether the user is still in city C when the user's real-time signal base station connection judgment time is the best recommendation time node. If the user's mobile phone signal is still connected to the signal base station of city C at this time, it is determined that the user is still in city C at this time, and the recommended operation of recommending a hotel to the user is performed. Otherwise, it is determined that the user is no longer in city C at this time, and no hotel recommendation is required for him. The recommended operation for users to recommend hotels is as follows: sort the hotels in order of scores from large to small, and recommend the top five hotel information to users through 5G messages, wherein the hotel information recommended to users through 5G messages can include the hotel's pre-recorded video about the hotel introduction and the corresponding hotel booking link, so that users can click to play and watch the hotel's introduction video, and book the hotel they are interested in through the corresponding link. In this way, the recommendation conversion rate of the recommended hotel is improved and the time for users to choose a hotel is reduced.
[0168] It should be noted that, for the description of the same steps and the same contents in this embodiment as those in other embodiments, reference can be made to the description in other embodiments and will not be repeated here.
[0169] The embodiment of the present application provides an information recommendation method. After determining the target user's visit time to the current destination through an information recommendation device, if it is determined that a service recommendation of a target service is required for the target user, a recommended time for recommending the target service to the target user is determined based on the visit time, and after determining at least one object to be recommended based on the recommended time, when the time is the recommended time, recommendation information including at least one object to be recommended is sent to the target terminal device of the target user. In this way, when determining that a service recommendation of the target service is required for the target user, firstly the recommended time for recommending the target service is determined based on the target user's visit time to the destination, and then at least one object to be recommended is determined based on the recommended time, and finally when the time is the recommended time, at least one object to be recommended is sent to the target terminal device, so that targeted service recommendations are made to the target user, which is more personalized, solves the problem that currently the hotel and other travel products cannot be accurately recommended to the user, and proposes a method for reasonably recommending hotels and other travel products to the user based on the user's behavior, thereby improving the search efficiency of hotels and other travel products.
[0170] Based on the above embodiments, the embodiments of the present application provide an information recommendation device, which can be applied to Figures 1-2 In the information recommendation method provided in the corresponding embodiment, refer to Figure 3 As shown, the electronic device 3 may include: a processor 31, a memory 32 and a communication bus 33, wherein:
[0171] A memory 32, for storing executable instructions;
[0172] A communication bus 33, used to realize the communication connection between the processor 31 and the memory 32;
[0173] The processor 31 is used to execute the information recommendation program stored in the memory 32 to implement the following steps:
[0174] Determine the target user's visit time to the current destination;
[0175] If it is determined that a service recommendation of a target service is required for the target user, a recommendation time for recommending the target service to the target user is determined based on the visit time;
[0176] Based on the recommendation time, determining at least one object to be recommended; wherein the at least one object to be recommended is used to provide a target service;
[0177] When the time is the recommendation time, recommendation information including at least one object to be recommended is sent to a target terminal device of a target user.
[0178] In other embodiments of the present application, if the processor executes the step of determining that a service recommendation of a target service is required for the target user, before determining the recommendation time for recommending the target service to the target user, the processor is further configured to execute the following steps:
[0179] Obtaining a target prediction model;
[0180] Obtain current behavior characteristic parameters of the target user when visiting the current destination and current service characteristic parameters of the current destination based on the target service;
[0181] The target prediction model is used to predict user behavior for current behavior feature parameters and current service feature parameters to obtain prediction results; wherein the prediction results are service recommendations that determine the need for target services for target users, or service recommendations that determine the need for target services for target users.
[0182] In other embodiments of the present application, if the processor executes the step of determining that a service recommendation of a target service is required for the target user, and determining a recommended time for recommending the target user based on the visit time, the following steps may be performed:
[0183] If it is determined that a service recommendation of the target service is required for the target user, determining a first preset time range to which the visit time belongs;
[0184] Determine the historical duration of the service from the current destination to the predetermined destination within the first preset time range;
[0185] Calculate the sum of the visit time and the historical duration to get the recommended time.
[0186] In other embodiments of the present application, when the processor executes the step of determining the historical duration from the current destination to the predetermined target service within the first preset time range, it can be implemented by the following steps:
[0187] Obtain a reference time set of the target user's arrival at a historical destination to a predetermined target service within a preset historical time range;
[0188] The duration in the reference duration set is averaged to obtain the historical duration.
[0189] In other embodiments of the present application, when the processor executes the step of determining at least one object to be recommended based on the recommendation time, it can be implemented by the following steps:
[0190] When the time is the recommended time, obtain the current location of the target user;
[0191] If the current location is within the current destination range, determining at least one service provider that provides the target service within a preset range centered on the current location;
[0192] Determine the distance between the current location and each service providing object to obtain at least one target distance;
[0193] Obtaining the service utilization rate of the target service of each service provider;
[0194] Determine the historical average price and the current average price of each service provider within the second preset time range;
[0195] Determine the evaluation score of each service provider based on the target distance, service utilization rate, historical average price and current average price of each service provider;
[0196] At least one object to be recommended is determined based on the evaluation score of at least one service providing object.
[0197] In other embodiments of the present application, when the processor executes the step of determining the evaluation score of each service provider based on the target distance, service usage rate, historical average price and current average price of each service provider, it can be implemented by the following steps:
[0198] Calculate the absolute value of the difference between the historical average price and the current average price of each service provision object to obtain a first value of each service provision object;
[0199] Calculating a preset square root of the target distance of each service provider to obtain a second value of each service provider;
[0200] Calculate the product of the first value and the second value of each service provision object to obtain a third value of each service provision object;
[0201] The ratio of the service utilization rate of each service provider to the third value is calculated to obtain an evaluation score of each service provider.
[0202] In other embodiments of the present application, when the processor executes the step of sending recommendation information including at least one object to be recommended to a target terminal device of a target user when the time is the recommendation time, it can be implemented by the following steps:
[0203] When the time is the recommendation time, obtaining an access link of at least one object to be recommended;
[0204] Sorting the access links of at least one to-be-recommended object in descending order according to the evaluation scores of the to-be-recommended object to obtain a target sorted object set;
[0205] Generate recommendation information based on the target sorting object set;
[0206] Send recommendation information to the target terminal device.
[0207] In other embodiments of the present application, the service characteristic parameters include current environment parameters of the destination and historical service usage characteristic parameters of the target service.
[0208] In other embodiments of the present application, when the processor executes the step of obtaining the target prediction model, it can be implemented by the following steps:
[0209] Determine the prediction model to be trained;
[0210] Obtain historical sample data of historical users about the target service;
[0211] Construct training sample data based on historical sample data;
[0212] The training sample data is used to train the prediction model to be trained to obtain the target prediction model.
[0213] It should be noted that the specific implementation process of the steps performed by the information recommendation device in this embodiment can refer to Figures 1-2 The implementation process of the information recommendation method provided in the corresponding embodiment will not be repeated here.
[0214] The embodiment of the present application provides an information recommendation device. After the target user's visit time to the current destination is determined by the information recommendation device, if it is determined that a service recommendation of a target service is required for the target user, the recommended time for recommending the target service to the target user is determined based on the visit time, and after at least one object to be recommended is determined based on the recommended time, when the time is the recommended time, recommendation information including at least one object to be recommended is sent to the target terminal device of the target user. In this way, when it is determined that a service recommendation of the target service is required for the target user, the recommended time for recommending the target service is first determined based on the time when the target user visits the destination, and then at least one object to be recommended is determined based on the recommended time, and finally when the time is the recommended time, at least one object to be recommended is sent to the target terminal device, so that targeted service recommendations are made to the target user, which is more personalized, solves the problem that currently the hotel and other travel products cannot be accurately recommended to the user, and proposes a reasonable recommendation of hotel and other travel products to the user based on the user's behavior, thereby improving the search efficiency of hotel and other travel products.
[0215] Based on the foregoing embodiments, the embodiments of the present application provide a computer-readable storage medium, referred to as a storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement reference Figures 1-2 The implementation process of the information recommendation method provided in the corresponding embodiment will not be repeated here.
[0216] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of hardware embodiments, software embodiments, or embodiments in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) that contain computer-usable program code.
[0217] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0218] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0219] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0220] The above description is only a preferred embodiment of the present application and is not intended to limit the protection scope of the present application.
Claims
1. An information recommendation method, characterized in that: The method comprises: Determine the target user's visit time to the current destination; If it is determined that a service recommendation of a target service is required for the target user, determining a recommended time for recommending the target service to the target user based on the visit time; Based on the recommendation time, determining at least one object to be recommended; wherein the at least one object to be recommended is used to provide the target service; When the time is the recommendation time, recommendation information including at least one of the objects to be recommended is sent to the target terminal device of the target user.
2. The method according to claim 1, characterized in that If it is determined that a service recommendation of a target service is required for the target user, before determining a recommendation time for recommending the target service to the target user, the method further includes: Obtaining a target prediction model; Acquire current behavior characteristic parameters of the target user when visiting the current destination and current service characteristic parameters of the current destination based on the target service; The target prediction model is used to predict user behavior of the current behavior characteristic parameters and the current service characteristic parameters to obtain a prediction result; wherein the prediction result is a service recommendation for determining that a target service is required for the target user, or a service recommendation for determining that a target service is not required for the target user.
3. The method according to claim 1, characterized in that If it is determined that a service recommendation of a target service is required for the target user, determining a recommendation time for the target user based on the visit time includes: If it is determined that a service recommendation of a target service is required for the target user, determining a first preset time range to which the visit time belongs; Determine the historical duration from reaching the current destination to booking the target service within the first preset time range; The sum of the visit time and the historical duration is calculated to obtain the recommended time.
4. The method according to claim 3, characterized in that The determining of the historical duration from arriving at the current destination to booking the target service within the first preset time range includes: Obtain a reference time set of the target user's arrival at a historical destination within a preset historical time range and the target service reservation within the first preset time range; An average value is calculated for the durations in the reference duration set to obtain the historical duration.
5. The method according to claim 1, characterized in that The determining at least one object to be recommended based on the recommendation time includes: When the time is the recommended time, obtaining the current location of the target user; If the current location is within the current destination range, determine at least one service provider that provides the target service within a preset range centered on the current location; Determine the distance between the current position and each of the service providing objects to obtain at least one target distance; Obtaining a service usage rate of the target service of each of the service providing objects; Determine the historical average price and the current average price of each of the service provision objects within the second preset time range; Determine an evaluation score for each of the service providing objects based on the target distance, the service usage rate, the historical average price and the current average price of each of the service providing objects; Based on the evaluation score of at least one of the service providing objects, at least one of the objects to be recommended is determined.
6. The method according to claim 5, characterized in that The step of determining the evaluation score of each service provider based on the target distance, the service usage rate, the historical average price and the current average price of each service provider includes: Calculating the absolute value of the difference between the historical average price and the current average price of each of the service provision objects to obtain a first value of each of the service provision objects; Calculating a preset square root of the target distance of each of the service providing objects to obtain a second value of each of the service providing objects; Calculate the product of the first value and the second value of each of the service provision objects to obtain a third value of each of the service provision objects; The ratio of the service usage rate of each of the service providing objects to the third value is calculated to obtain an evaluation score of each of the service providing objects.
7. The method according to claim 5, characterized in that The sending, when the time is the recommendation time, recommendation information including at least one of the to-be-recommended objects to the target terminal device of the target user comprises: When the time is the recommendation time, obtaining at least one access link of the object to be recommended; Sorting the access links of at least one of the objects to be recommended in descending order according to the evaluation scores of the objects to be recommended to obtain a target sorting object set; Generating the recommendation information based on the target sorting object set; The recommendation information is sent to the target terminal device.
8. The method according to claim 2, characterized in that: The service characteristic parameters include current environment parameters of the destination and historical service usage characteristic parameters of the target service.
9. The method according to claim 2, characterized in that: The obtaining of the target prediction model comprises: Determine the prediction model to be trained; Obtain historical sample data of historical users about the target service; Constructing training sample data based on the historical sample data; The training sample data is used to perform model training on the prediction model to be trained to obtain the target prediction model.
10. An information recommendation device, characterized in that: The device comprises at least: a memory, a processor and a communication bus; wherein: The memory is used to store executable instructions; The communication bus is used to realize the communication connection between the processor and the memory; The processor is used to execute the information recommendation program stored in the memory to implement the steps of the information recommendation method according to any one of claims 1 to 9.
11. A storage medium, characterized in that: The storage medium stores an information recommendation program, and when the information recommendation program is executed by the processor, the steps of the information recommendation method according to any one of claims 1 to 9 are implemented.