Multi-factor-based self-adaptive service method and device for passengers entering island

Abstract

The invention belongs to the technical field of data processing, and particularly discloses a multi-factor-based island-entering passenger adaptive service method and device, and the method comprises the steps: obtaining the passenger information of a current island-entering vehicle; on the basis of the passenger service information contribution degree model and the passenger information of the current island-entering vehicle, weighted summation is carried out, the interestingness of the passenger of the current island-entering vehicle to various types of service information is determined, and the passenger information comprises various types of passenger characteristics; pushing the service information based on the interestingness of the passenger of the current island entering vehicle in various service information; wherein the passenger service information contribution degree model is used for representing the contribution degree of each type of passenger characteristics to each data set, the contribution degree is used as the weight of weighted summation, and the contribution degree is determined by calculating the information gain between the passenger characteristics and the data sets. According to the method and the device, the service information is pushed according to the interestingness, the method and the device can adapt to different passengers, the matching degree between the passengers and the pushed service information is effectively improved, and the service quality is improved.

Description

Technical Field

[0001] The present application belongs to the field of data processing technology, and more specifically, to a method and device for adaptively providing services to passengers entering an island based on multiple factors. Background Art

[0002] For island cities, when travelers enter the island, information service providers usually proactively push various service information to them. This information may include business information, vacation information, health care information, tourism information, etc. However, the push method commonly used by these information service providers is relatively simple, that is, the same service information is pushed to all travelers indiscriminately, without considering the personalized needs and preferences of each traveler. This push method results in a low match between travelers and the service information received. Many travelers may receive information that does not match their interests, thus affecting their travel experience. How to effectively improve the match between travelers and pushed service information is a technical problem that needs to be solved urgently in this field. Summary of the invention

[0003] In view of the defects of the prior art, the purpose of this application is to effectively improve the matching degree between passengers and pushed service information.

[0004] To achieve the above objectives, in a first aspect, the present application provides a method for adaptively serving passengers entering the island based on multiple factors, the method comprising: Get the passenger information of the vehicles currently entering the island; Based on the passenger service information contribution model and the passenger information of the current vehicles entering the island, a weighted sum is performed to determine the interest of the passengers of the current vehicles entering the island in various types of service information. The passenger information includes multiple types of passenger characteristics; Push service information based on the interest of passengers entering the island in various service information; Among them, the passenger service information contribution model is used to characterize the contribution of various passenger characteristics to each data set. The contribution is used as the weight of the weighted sum. The data set is constructed by counting the passengers' responses to the push of specified service information. The contribution is determined by calculating the information gain between the passenger characteristics and the data set.

[0005] Specifically, in the model building stage (also called the model training stage), by counting the responses of passengers to the push of specified types of service information, a corresponding data set can be constructed, and then by counting the responses of passengers to the push of various types of service information, a data set corresponding to each type of service information can be constructed, and then by calculating the information gain between the specified passenger characteristics and the specified data set, the contribution of the specified passenger characteristics to the specified data set can be determined, and then by calculating the information gain between the characteristics of each type of passenger and each data set, the contribution of each type of passenger characteristics to each data set can be determined, so as to obtain a passenger service information contribution model.

[0006] In the model application stage, the passenger information of the current vehicles entering the island can be obtained. The passenger information includes multiple types of passenger characteristics. One type of passenger characteristics represents one factor that needs to be considered, and multiple types of passenger characteristics represent multiple factors that need to be considered. Then, combined with the passenger service information contribution model, the contribution is used as the weight to perform a weighted summation of the various types of passenger characteristics of the current passengers entering the island. The interest of the passengers of the current vehicles entering the island in various types of service information can be determined, and service information can be pushed based on the interest. It can adapt to different passengers, effectively improve the matching degree between passengers and pushed service information, and improve service quality.

[0007] In a possible implementation, the contribution is obtained through the following steps: Calculate the empirical entropy of the data set; Calculate the empirical conditional entropy of passenger characteristics relative to the dataset; Based on empirical entropy and empirical conditional entropy, the information gain between passenger characteristics and the data set is determined; The information gain between the passenger characteristics and the data set is normalized to obtain the contribution of the passenger characteristics to the data set.

[0008] In a possible implementation, calculating the empirical entropy of the data set specifically includes determining the empirical entropy by the following formula: ; in, Representation dataset The empirical entropy of In order to count the number of passengers A data set constructed based on the response to the push of class service information. Indicates Class service information, Indicates Class service information ( ) The dimension of the response assignment, For the dataset middle The number of dimension samples, The dimension sample belongs to Class service information ( ) Response dimensions, Representation dataset The total number of samples.

[0009] In a possible implementation, the above calculation of the empirical conditional entropy of the passenger characteristics relative to the data set specifically includes determining the empirical conditional entropy by the following formula: ; in, Representation dataset Any feature A in the dataset The empirical conditional entropy of The dimension number representing the value of feature A. represents the dimension of the value assigned to feature A, Representation dataset middle The number of dimension samples, The dimension sample belongs to the first This dimension also belongs to Class service information ( ) Response dimensions, Representation dataset middle The number of dimension samples, The dimension sample belongs to the first dimensions.

[0010] In a possible implementation, the above passenger service information contribution model and the passenger information of the current vehicle entering the island are weighted summed to determine the interest of the passenger of the current vehicle entering the island for various types of service information, specifically including determining the interest by the following formula: ; in, Indicates the number of passengers in the current vehicle entering the island Class service information ( )’s interest; Indicates the characteristics of Category 1 passengers, Indicates the characteristics of the second type of passengers, represents the characteristics of the third type of passengers, and so on. Indicates Passenger characteristics, represents the number of passenger characteristic categories; Represents the first type of passenger feature pair data set The contribution of Represents the second type of passenger feature pair data set The contribution of Represents the third type of passenger feature pair data set The contribution of Indicates Passenger feature pair dataset The contribution of In order to count the number of passengers Class service information ( ) push response data set, A number indicating the service information category.

[0011] In a possible implementation, the service information is pushed based on the interest of passengers on the current inbound vehicles in various types of service information, including: When the difference between different interest levels is greater than a preset threshold, based on a preset multiple and a ratio between different interest levels, target service information is selected from multiple types of service information, and the target service information is pushed, and the ratio between the interest level corresponding to the target service information and the interest level corresponding to the non-target service information is greater than or equal to the preset multiple; When the difference between different interest levels is less than or equal to a preset threshold, various types of service information are pushed.

[0012] In a possible implementation, the method further includes: Update the data set by continuously counting passengers' responses to service information push; Based on the updated data set, the passenger service information contribution model is modified by calculating the information gain between passenger characteristics and the data set.

[0013] It is understandable that by continuously revising the passenger service information contribution model, the accuracy of interest estimation can be continuously guaranteed and the stability of service quality can be ensured.

[0014] In one possible implementation, when the current entry time of vehicles entering the island belongs to the time period corresponding to a specific event (such as the Spring Festival travel event), the data set is constructed by counting the responses of passengers to the push of specified service information for the historical time period corresponding to the specific event.

[0015] It can be understood that a data set is constructed for a specific event, and the passenger service information contribution model corresponding to the specific event is obtained by calculating the information gain between the passenger characteristics and the data set. Then, when the entry time of the current vehicle entering the island belongs to the time period corresponding to the specific event, the passenger service information contribution model corresponding to the specific event can be used to determine the interest of the passengers of the current vehicle entering the island in various types of service information and push service information.

[0016] Therefore, it is possible to adapt to different passengers during the time period corresponding to a specific event, effectively improve the matching degree between passengers and the pushed service information, and improve the service quality under specific events.

[0017] In a second aspect, the present application provides an adaptive service device for island-bound passengers based on multiple factors, including: Passenger information acquisition module, used to obtain passenger information of vehicles currently entering the island; The interest determination module is used to perform weighted summation based on the passenger service information contribution model and the passenger information of the current vehicle entering the island to determine the interest of the passenger of the current vehicle entering the island in various types of service information. The passenger information includes multiple types of passenger characteristics; The service information push module is used to push service information based on the interest of passengers on the current vehicles entering the island in various types of service information; Among them, the passenger service information contribution model is used to characterize the contribution of various passenger characteristics to each data set. The contribution is used as the weight of the weighted sum. The data set is constructed by counting the passengers' responses to the push of specified service information. The contribution is determined by calculating the information gain between the passenger characteristics and the data set.

[0018] In a third aspect, the present application provides an electronic device comprising: at least one memory for storing programs; and at least one processor for executing the programs stored in the memory. When the program stored in the memory is executed, the processor is used to execute the method described in the first aspect or any possible implementation of the first aspect.

[0019] It can be understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.

[0020] In general, the above technical solutions conceived by this application have the following beneficial effects compared with the prior art: (1) Through historical statistical data, combined with information on the frequency of passengers entering the island, the length of time on the island, and the information structure of passengers entering the island, a passenger service information contribution model is formed through normalization processing based on the information entropy contribution model. The passenger information of the current vehicle entering the island can be obtained through the passenger and vehicle ferry ticket booking application. Based on the passenger service information contribution model, the service information category can be determined and pushed according to the passenger information. It can adapt to different passengers and can significantly improve the accuracy (matching degree) of information services and improve service quality.

[0021] (2) Both the model iteration method based on new data and the targeted model training method for specific time or events are considered. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1This is one of the flow charts of the adaptive service method for island passengers based on multiple factors provided in the embodiment of the present application; Figure 2 This is the second flow chart of the adaptive service method for island passengers based on multiple factors provided in the embodiment of the present application; Figure 3 It is a structural schematic diagram of an adaptive service device for island passengers based on multiple factors provided in an embodiment of the present application; Figure 4 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0024] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0025] In the description of the embodiments of the present application, unless otherwise specified, "multiple" means two or more than two. For example, multiple processing units refer to two or more processing units, etc.; multiple elements refer to two or more elements, etc.

[0026] The embodiments of the present application are described below in conjunction with the drawings in the embodiments of the present application.

[0027] Figure 1 This is one of the flow charts of the adaptive service method for island passengers based on multiple factors provided in the embodiment of the present application, such as Figure 1 As shown, the method includes the following steps S101, S102 and S103.

[0028] Step S101, obtaining passenger information of the vehicle currently entering the island; Step S102, based on the passenger service information contribution model and the passenger information of the current vehicle entering the island, a weighted sum is performed to determine the interest of the passenger of the current vehicle entering the island in various types of service information, and the passenger information includes multiple types of passenger characteristics; Step S103, based on the interest of passengers on the current vehicle entering the island in various types of service information, push service information; Among them, the passenger service information contribution model is used to characterize the contribution of various passenger characteristics to each data set. The contribution is used as the weight of the weighted sum. The data set is constructed by counting the passengers' responses to the push of specified service information. The contribution is determined by calculating the information gain between the passenger characteristics and the data set.

[0029] Specifically, in the model building stage (also called the model training stage), by counting the responses of passengers to the push of specified types of service information, a corresponding data set can be constructed, and then by counting the responses of passengers to the push of various types of service information, a data set corresponding to each type of service information can be constructed, and then by calculating the information gain between the specified passenger characteristics and the specified data set, the contribution of the specified passenger characteristics to the specified data set can be determined, and then by calculating the information gain between the characteristics of each type of passenger and each data set, the contribution of each type of passenger characteristics to each data set can be determined, so as to obtain a passenger service information contribution model.

[0030] In the model application stage, the passenger information of the current vehicles entering the island can be obtained. The passenger information includes multiple types of passenger characteristics. One type of passenger characteristics represents one factor that needs to be considered, and multiple types of passenger characteristics represent multiple factors that need to be considered. Then, combined with the passenger service information contribution model, the contribution is used as the weight to perform a weighted summation of the various types of passenger characteristics of the current passengers entering the island. The interest of the passengers of the current vehicles entering the island in various types of service information can be determined, and service information can be pushed based on the interest. It can adapt to different passengers, effectively improve the matching degree between passengers and pushed service information, and improve service quality.

[0031] Figure 2 This is a second flow chart of the adaptive service method for island passengers based on multiple factors provided in the embodiment of the present application, such as Figure 2 As shown, this application provides an adaptive service method for passengers entering the island based on multiple factors, so as to realize passenger services based on historical information about the island and personnel structure information. Through this method, customized information service push can be provided according to the characteristics of the passenger structure, and adaptive service correction and optimization can be performed according to the user response.

[0032] The parameters involved in this algorithm include: historical information on passengers entering the island, structural information on passengers’ companions (people traveling together), and response to passenger service information.

[0033] For a specified historical period, query the historical data of the passenger and vehicle ferry (a ferry that can carry both passengers and vehicles) ticket booking application to obtain passenger information on vehicles loaded on the passenger and vehicle ferry, including historical information on passengers entering the island and information on the structure of passengers' companions. For a specified historical period, query the historical data of the passenger service information push application to obtain the response to the passenger service information. An example of the push method of the passenger service information push application is provided here. The optional push method is SMS push, which is configured with service information and corresponding links. By monitoring the time when the user clicks the link, the passenger's response to this service information push can be obtained.

[0034] The information on the historical entry of passengers to the island (entry to the island can also be called island entry) refers to the relevant information on the historical entry of passengers to the island in the past two years, including the frequency of entry information TraN and the length of time passengers stay on the island TraT. The information on the structure of passengers' companions refers to the age and gender structure of the passengers entering the island this time, represented by TraE. The response to passenger service information refers to whether the passenger takes further access measures after the system pushes the service information to the passenger, represented by TraV, and generally takes two values: yes and no.

[0035] This algorithm consists of three parts. The first part is the model training part. The calculation method is as follows: Step S11, collect a certain amount of data sets, including historical information on passengers entering the island, information on the structure of passengers' companions, and responses to passenger service information.

[0036] Step S12, qualitatively or quantitatively process the historical information of passengers entering the island and the information of the structure of passengers' companions, and the specific method is as follows: Step S121, the island entry frequency information TraN is divided into three levels, when the island entry frequency > 30 times, TraN = high; when 30 times ≥ island entry frequency > 10 times, TraN = medium; when 10 times ≥ island entry frequency ≥ 0 times, TraN = low.

[0037] Step S122, the passenger's time on the island information TraT is divided into three levels, when the time on the island is greater than 30 days, TraT=more; when 30 days≥time on the island>7 days, TraT=medium; when 7 days≥time on the island≥0 days, TraT=less.

[0038] Step S123, the passenger companion structure information TraE includes four categories, namely whether there are elderly people, whether there are middle-aged and young people, whether there are children, and the proportion of women. Among them, if the proportion of women exceeds 50%, it is a high proportion, otherwise it is a low proportion. Based on the above statistics, a one-dimensional matrix is ​​formed , For example: TraE{yes, yes, no, high}. For example, passengers aged between 0 and 10 are children, passengers aged between 10 and 50 are young and middle-aged, and passengers aged over 50 are elderly.

[0039] Step S13, classify the passenger service information response TraV according to the service information type, which can generally be divided into business service information (user Indicates this type of service information), vacation service information (with Indicates this type of service information), health care service information (with Indicates this type of service information), tourism service information (with The TraV value is generated according to the passenger's response to the push of the four types of service information. When the passenger clicks on the corresponding service, TraV is yes, otherwise it is no.

[0040] Step S14, combining the above data to form the following structured data set: , , , ,in, This is a data set constructed by counting passengers' responses to business service information push, as shown in Table 1; This is a dataset constructed by counting travelers' responses to holiday service information push, as shown in Table 2; This is a data set constructed by counting passengers' responses to health care service information push, as shown in Table 3; This is a data set constructed by counting the responses of passengers to the push of tourism service information. The data set is shown in Table 4. Passenger characteristics A~F can be called the 1st to 6th category passenger characteristics. Correspondingly, passenger characteristics A~F can also be expressed as , , …, .

[0041] Table 1 Dataset corresponding to business service information

[0042] Table 2 Dataset corresponding to vacation service information

[0043] Table 3 Dataset corresponding to health care service information

[0044] Table 4 Dataset corresponding to tourism service information

[0045] Step S15: Use the above data to train the information gain model. The specific calculation method is as follows: Step S151, calculate the data set The empirical entropy , where K is The dimension of the response value (the number of possible values ​​is the same as the dimension of the value), The possible values ​​of the response include "yes" and "no", so the dimension of the value assignment is 2, so K=2 in this example. For the dataset Middle The number of samples in each dimension, Representation dataset The total number of samples is calculated as follows: ; Step S152, calculate the feature A in the data set for the data set Empirical conditional entropy , where i is the dimension number of the value assigned to feature A. For example, the maximum value n of the value i of feature A (that is, the dimension of the value assigned to feature A) is 3. The specific calculation formula is as follows: ; in, Representation dataset middle The number of dimension samples, The dimension sample belongs to the first Dimensions also belong to The response dimensions, Representation dataset middle The number of dimension samples, The dimension sample belongs to the first dimensions.

[0046] Step S153, according to the above step S152, calculate the features B~F for the data set The empirical conditional entropy of .

[0047] Step S154, based on all the empirical conditional entropies obtained by calculation, calculate the entropy of each feature for the data set The information gain is calculated as follows: ; Step S155, complete the information gain calculation between all features and data sets according to the above method to form an information entropy contribution model. The information entropy contribution model includes the information gain between various features and each data set, as shown in Table 5.

[0048] Table 5 Information gain between various features and various data sets

[0049] Step S16: After obtaining the above information gain value, the contribution of each feature to the data set can be obtained. For example, , , , , , , it can be seen that feature A is important for the data set The contribution of feature D to the data set is the largest. The contribution of is the smallest. It can be seen that information gain can represent the contribution of features (such as features A~D) to the data set.

[0050] Step S17, the data set The associated information gain values ​​are normalized to form a data set The weight parameter , including weight parameters of various features, Represents the relationship between feature A and data set The weight parameter is obtained by normalizing the information gain value between Represents the feature B and the data set The weight parameter is obtained by normalizing the information gain value between , and so on. Represents the feature F and the data set The weight parameter is obtained by normalizing the information gain value between , and the specific calculation method is as follows: ; Step S18, according to step S17, obtain the weight parameters of all data sets, such as data set The weight parameter (or expressed as ), dataset The weight parameter (or expressed as ), dataset The weight parameter (or expressed as ), dataset The weight parameter (or expressed as ). The weight parameters of all data sets constitute the passenger service information contribution model.

[0051] Step S19, after obtaining the weight parameters of all data sets (passenger service information contribution model), the first part of the model training work is completed.

[0052] The second part is to determine the category of information service to be pushed based on the model trained in the first part. The specific method is as follows: Step S21, firstly, obtain the passenger information of the current vehicle entering the island through the passenger and vehicle ferry ticket booking application, including the passenger's historical entry information and the passenger's companion structure information.

[0053] Step S22, generating a data list according to the following rules and the above information.

[0054] (1) If the frequency of entering the island is high, the value is assigned to 1, the medium value is assigned to 0.8, and the low value is assigned to 0.5; (2) If the time spent on the island is long, the value is assigned to 1, the medium value is assigned to 0.8, and the low value is assigned to 0.5; (3) Whether there are elderly / young people / children, if yes, the value is assigned to 1, if no, the value is assigned to 0.5; (4) If the proportion of women is high, the value is assigned to 1; if the proportion of women is low, the value is assigned to 0.5.

[0055] The specific data generated is shown in Table 6.

[0056] Table 6 Passenger information of vehicles currently entering the island

[0057] Step S23: perform weighted calculation on all data according to the model trained in the first part, and calculate the result express, Passengers expressed The interest level (interest level can also be understood as the matching degree between passengers and corresponding service information), Passengers expressed Interest level, Passengers expressed Interest level, Passengers expressed The interest level is calculated as follows: ; in, express (Feature A) for the dataset The contribution of express (Feature B) for the dataset The contribution of express (Feature C) for the dataset The contribution of express (Feature D) for the dataset The contribution of express (Feature E) for the dataset The contribution of express (Feature F) for the dataset contribution.

[0058] ; in, express (Feature A) for the dataset The contribution of express (Feature B) for the dataset The contribution of express (Feature C) for the dataset The contribution of express (Feature D) for the dataset The contribution of express (Feature E) for the dataset The contribution of express (Feature F) for the dataset contribution.

[0059] ; in, express (Feature A) for the dataset The contribution of express (Feature B) for the dataset The contribution of express (Feature C) for the dataset The contribution of express (Feature D) for the dataset The contribution of express (Feature E) for the dataset The contribution of express (Feature F) for the dataset contribution.

[0060] ; in, express (Feature A) for the dataset The contribution of express (Feature B) for the dataset The contribution of express (Feature C) for the dataset The contribution of express (Feature D) for the dataset The contribution of express (Feature E) for the dataset The contribution of express (Feature F) for the dataset contribution.

[0061] Step S24, based on the calculated Data is selected according to the following rules to form a passenger information push service model: Step S241, when any dimension value ( ) is more than n times the values ​​of the other three dimensions, it is considered that the passenger has a high interest (or acceptance) in the single-dimensional information, and the passenger service information push application will push the information service corresponding to the dimension to the passenger; (the value of n is designed based on experience, and can generally be 2~3) Step S242: when the values ​​of any two dimensions are close to each other and are more than n times higher than the values ​​of the other two dimensions, it is considered that the passenger is highly interested in the information of the two dimensions, and the information services corresponding to the two dimensions are pushed to the passenger through the passenger service information push application; Step S243: when the values ​​of any three dimensions are close to each other and are more than n times higher than the value of the last dimension, it is considered that the passenger has a high interest in the information of the three dimensions, and the information services corresponding to the three dimensions are pushed to the passenger through the passenger service information push application; Step S244, when the values ​​of the four dimensions are close, the information services corresponding to all four dimensions are pushed to the passenger through the passenger service information push application.

[0062] Step S25, pushing corresponding information to the passenger according to the information service push dimension calculated in step S24.

[0063] The third part is to continuously build new data sets based on the completion of the second part and the response of passengers to service information. When the new data set reaches a certain size, the model is corrected according to the method in the first part. The correction method can be periodic correction (such as model correction every 3 months or a year) or event-driven, such as selecting the data set during the Spring Festival travel event for special training to obtain the passenger information push service model for the Spring Festival travel.

[0064] The following is a description of the adaptive service device for island passengers based on multiple factors provided by the present application. The adaptive service device for island passengers based on multiple factors described below and the adaptive service method for island passengers based on multiple factors described above can be referenced to each other.

[0065] Figure 3 is a schematic diagram of the structure of the adaptive service device for island passengers based on multiple factors provided in an embodiment of the present application, such as Figure 3 As shown, the device includes: a passenger information acquisition module 10, an interest level determination module 20 and a service information push module 30. Among them: Passenger information acquisition module 10, used to obtain passenger information of vehicles currently entering the island; The interest determination module 20 is used to perform weighted summation based on the passenger service information contribution model and the passenger information of the current vehicle entering the island to determine the interest of the passenger of the current vehicle entering the island for various types of service information, and the passenger information includes multiple types of passenger characteristics; The service information push module 30 is used to push service information based on the interest of passengers in the current vehicle entering the island for various types of service information; Among them, the passenger service information contribution model is used to characterize the contribution of various passenger characteristics to each data set. The contribution is used as the weight of the weighted sum. The data set is constructed by counting the passengers' responses to the push of specified service information. The contribution is determined by calculating the information gain between the passenger characteristics and the data set.

[0066] It can be understood that the detailed functional implementation of each of the above-mentioned units / modules can be found in the introduction of the aforementioned method embodiment, and will not be repeated here.

[0067] It should be understood that the above-mentioned device is used to execute the method in the above-mentioned embodiment. The implementation principle and technical effect of the corresponding program module in the device are similar to those described in the above-mentioned method. The working process of the device can refer to the corresponding process in the above-mentioned method, which will not be repeated here.

[0068] Based on the method in the above embodiment, an embodiment of the present application provides an electronic device, Figure 4 is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application, such as Figure 4As shown, the electronic device may include: a processor (Processor) 810, a communication interface (Communications Interface) 820, a memory (Memory) 830 and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other through the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to execute the method in the above embodiment.

[0069] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application.

[0070] Based on the method in the above embodiment, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program runs on a processor, the processor executes the method in the above embodiment.

[0071] Based on the method in the above embodiment, an embodiment of the present application provides a computer program product. When the computer program product runs on a processor, the processor executes the method in the above embodiment.

[0072] It is understandable that the processor in the embodiment of the present application may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. The general-purpose processor may be a microprocessor or any conventional processor.

[0073] The method steps in the embodiments of the present application can be implemented by hardware or by a processor executing software instructions. The software instructions can be composed of corresponding software modules, and the software modules can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, mobile hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an ASIC.

[0074] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions may be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

[0075] It should be understood that the various numerical numbers involved in the embodiments of the present application are only used for the convenience of description and are not used to limit the scope of the embodiments of the present application.

[0076] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A multi-factor based adaptive service method for passengers entering the island, characterized in that: include: Get the passenger information of the vehicles currently entering the island; Based on the passenger service information contribution model and the passenger information of the current vehicles entering the island, a weighted sum is performed to determine the interest of the passengers of the current vehicles entering the island in various types of service information. The passenger information includes multiple types of passenger characteristics; Push service information based on the interest of passengers entering the island in various service information; Among them, the passenger service information contribution model is used to characterize the contribution of various passenger characteristics to each data set. The contribution is used as the weight of the weighted sum. The data set is constructed by counting the passengers' responses to the push of specified service information. The contribution is determined by calculating the information gain between the passenger characteristics and the data set.

2. The method for adaptive service for island passengers based on multiple factors according to claim 1 is characterized in that: The contribution is obtained through the following steps: Calculate the empirical entropy of the data set; Calculate the empirical conditional entropy of passenger characteristics relative to the dataset; Based on empirical entropy and empirical conditional entropy, the information gain between passenger characteristics and the data set is determined; The information gain between the passenger characteristics and the data set is normalized to obtain the contribution of the passenger characteristics to the data set.

3. The method for adaptive service for island passengers based on multiple factors according to claim 2 is characterized in that: The calculating the empirical entropy of the data set specifically includes determining the empirical entropy by the following formula: ； in, Representation dataset The empirical entropy of In order to count the number of passengers A data set constructed based on the response to the push of class service information. Indicates Class service information, Indicates The dimension of the value assigned to the class service information response, For the dataset middle The number of dimension samples, The dimension sample belongs to Class service information response dimensions, Representation dataset The total number of samples.

4. The method for adaptive service for island passengers based on multiple factors according to claim 3 is characterized in that: The calculating of the empirical conditional entropy of the passenger characteristics relative to the data set specifically includes determining the empirical conditional entropy by the following formula: ； in, Representation dataset Any feature A in the dataset The empirical conditional entropy of The dimension number representing the value of feature A. represents the dimension of the value assigned to feature A, Representation dataset middle The number of dimension samples, The dimension sample belongs to the first This dimension also belongs to Class service information response dimensions, Representation dataset middle The number of dimension samples, The dimension sample belongs to the feature A dimensions.

5. The method for adaptive service for island passengers based on multiple factors according to claim 1 is characterized in that: The passenger service information contribution model and the passenger information of the current vehicle entering the island are weighted summed to determine the interest of the passenger of the current vehicle entering the island for various types of service information, specifically including determining the interest by the following formula: ； in, Indicates the number of passengers in the current vehicle entering the island The interest level of the service information; Indicates the characteristics of Category 1 passengers, Indicates the characteristics of the second type of passengers, represents the characteristics of the third type of passengers, and so on. Indicates Passenger characteristics, represents the number of passenger characteristic categories; Represents the first type of passenger feature pair data set The contribution of Represents the second type of passenger feature pair data set The contribution of Represents the third type of passenger feature pair data set The contribution of Indicates Passenger feature pair dataset The contribution of In order to count the number of passengers A data set constructed based on the response to the push of class service information. Indicates Class service information.

6. The method for adaptive service for island passengers based on multiple factors according to claim 1 is characterized in that: The service information is pushed based on the interest of passengers on the current vehicle entering the island in various service information, including: When the difference between different interest levels is greater than a preset threshold, based on a preset multiple and a ratio between different interest levels, target service information is selected from multiple types of service information, and the target service information is pushed, and the ratio between the interest level corresponding to the target service information and the interest level corresponding to the non-target service information is greater than or equal to the preset multiple; When the difference between different interest levels is less than or equal to a preset threshold, various types of service information are pushed.

7. The method for adaptive service for island passengers based on multiple factors according to claim 1 is characterized in that: Also includes: Update the data set by continuously counting passengers' responses to service information push; Based on the updated data set, the passenger service information contribution model is modified by calculating the information gain between passenger characteristics and the data set.

8. The method for adaptively serving passengers entering the island based on multiple factors according to any one of claims 1 to 7, characterized in that: When the entry time of the current vehicle entering the island belongs to the time period corresponding to a specific event, the data set is constructed by counting the responses of passengers to the push of designated service information for the historical time period corresponding to the specific event.

9. An adaptive service device for passengers entering the island based on multiple factors, characterized in that: include: Passenger information acquisition module, used to obtain passenger information of vehicles currently entering the island; The interest determination module is used to perform weighted summation based on the passenger service information contribution model and the passenger information of the current vehicle entering the island to determine the interest of the passenger of the current vehicle entering the island in various types of service information. The passenger information includes multiple types of passenger characteristics; The service information push module is used to push service information based on the interest of passengers on the current vehicles entering the island in various types of service information; Among them, the passenger service information contribution model is used to characterize the contribution of various passenger characteristics to each data set. The contribution is used as the weight of the weighted sum. The data set is constructed by counting the passengers' responses to the push of specified service information. The contribution is determined by calculating the information gain between the passenger characteristics and the data set.

10. An electronic device, characterized in that: include: at least one memory for storing a computer program; At least one processor is used to execute the program stored in the memory. When the program stored in the memory is executed, the processor is used to execute the method according to any one of claims 1 to 8.

Images