Tourism information recommendation method and system based on big data platform

Abstract

The invention discloses a tourism information recommendation method and system based on a big data platform, and relates to the technical field of big data information recommendation, and the tourism information recommendation method based on the big data platform specifically comprises the steps: 1, collecting the historical behavior information of a user based on the big data platform, extracting the historical tourism tendency data features of the user, the method comprises a first step of connecting with a user behavior information data source through a big data platform, collecting user historical behavior information and extracting data features of user historical tourism tendency, and a second step of determining relevance in tourism information through data analysis based on the user historical tourism tendency data features, screening associated users of users with tourism information to be pushed, and pushing the tourism information to be pushed. And step 3, analyzing a historical travel tendency data feature set of the associated users, comprehensively setting travel tendency indexes, and determining a travel information pushing scheme. The method can improve recommendation accuracy, optimize user experience and industry operation efficiency, and promote reasonable configuration of travel resources.

Description

Technical Field

[0001] The present invention belongs to the technical field of big data information recommendation, and in particular relates to a tourism information recommendation method and system based on a big data platform. Background Art

[0002] With the rapid development of Internet technology, the tourism industry is undergoing an unprecedented digital and information transformation. This transformation has not only changed the operating model of the tourism industry, but also greatly improved the experience and satisfaction of tourists. Traditional tourism information recommendation methods, such as manual screening, advertising or simple search matching, can no longer meet the growing diversified and personalized needs of modern tourists.

[0003] With the improvement of people's living standards and the accumulation of travel experience, modern tourists' demand for travel information is becoming increasingly diversified and personalized. They not only want to obtain basic information about tourist destinations, transportation, accommodation, etc., but also want to understand in-depth content such as local culture, customs, and food. At the same time, they pay more attention to the personalization and uniqueness of the travel experience and hope to customize the travel itinerary according to their interests and preferences.

[0004] Therefore, there is an urgent need for a tourism information recommendation method based on a big data platform. By making full use of users' historical behavior data and using data mining and analysis, users' travel preferences and consumption habits can be discovered, thereby providing users with more accurate and personalized tourism information recommendations. Summary of the invention

[0005] The purpose of the present invention is to provide a tourism information recommendation method and system based on a big data platform, which is used to solve the technical problem in the prior art that it is impossible to make tourism information recommendation solutions match users' travel preferences and consumption habits by making full use of users' historical behavior data.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] The tourism information recommendation method based on the big data platform includes:

[0008] Step 1: Collect user historical behavior information based on the big data platform and extract user historical travel tendency data features;

[0009] By connecting the big data platform with the user behavior information data source, the user historical behavior information is collected. The user historical behavior information includes user location information data, user preference information data, user consumption habit information data and user reservation information data. The duration T is determined as a cycle, and each cycle is divided into m time periods. The duration of each time period is t. Within the n cycle duration, real-time or historical user behavior information data is collected from various user behavior information data sources. The data mining technology is used to extract the data features of the user's historical travel tendency from the user's historical behavior information.

[0010] Step 2: Based on the historical travel tendency data characteristics of users, determine the relevance of travel information through data analysis, and select the related users of the users to whom the travel information is to be pushed;

[0011] Step 3: Analyze the historical travel tendency data feature set of related users, comprehensively set travel tendency indicators, and determine the travel information push plan.

[0012] Furthermore, data mining technology is used to extract data features of users' historical travel tendencies from their historical behavior information. The specific method is as follows:

[0013] The set D(A)={u(A), o(A), c(A)} is used to represent the historical travel tendency data feature set of user A, where u(A) represents the travel destination feature data set of user A, u(A)={(s(A,j), r(A,j))|j=1,...,g}, where j represents user A browsing the jth tourist area, g represents user A browsing g tourist destinations in total, s(A,j) represents user A browsing the jth tourist area type, r(A,j) represents the number of times user A browsed the jth tourist area, the reservation time feature data refers to the date that appears more than L times in the user's reservation information data, and this type of date is recorded as the reservation date of user A, and the consumption feature data refers to the amount of consumption of the user for each travel obtained by analyzing the user's consumption habit information data.

[0014] Furthermore, the associated users of the user whose travel information is to be pushed are screened, and the specific method is as follows:

[0015] Identify the users to whom travel information is to be pushed, establish a user travel similarity index based on the historical travel tendency data of the users to be pushed and the historical travel tendency data of other users, set a user travel similarity index threshold Q, calculate the user travel similarity index between the users to be pushed and other users, filter out user travel similarity indexes that are less than or equal to the threshold Q, and determine other users who meet the condition, and record such users as associated users of the users to whom travel information is to be pushed.

[0016] Furthermore, based on the historical travel tendency data of the user to be pushed and the historical travel tendency data of other users, a user travel similarity index is established. The specific method is as follows:

[0017] The formula Xs(A, B) = [kag*ag(A)-ag(B)|+klo*lo(A, B)]*sd(A, B) is used to represent the user travel similarity index, where Xs(A, B) represents the user travel similarity index between A and B, A represents the user A to be pushed, B represents other user B, ag(A) represents the age of user A, ag(B) represents the age of user B, lo(A, B) represents the location distance between user A and user B, sd(A, B) represents the travel consumption tendency similarity index between user A and user B, and k ag Represents the user age weight coefficient, k lo Represents the user distance weight coefficient. The smaller the value of Xs(A, B), the higher the correlation degree of the travel tendency data between user A and user B.

[0018] Furthermore, the tourism consumption tendency similarity index includes:

[0019] Using the formula represents the similarity index of tourism consumption tendency, where sd(A, B) represents the similarity index of tourism consumption tendency between A and B, A represents the user A to be pushed, B represents other user B, i represents the type of the i-th tourist area, H represents a total of H types of tourist destination feature data, A(i) represents the number of times the user A to be pushed goes to the i-th tourist area within a cycle, B(i) represents the number of times the user B goes to the i-th tourist area within a cycle, mA(i) represents the average consumption amount of the user A to be pushed going to the i-th tourist area, mB(i) represents the average consumption amount of the user B going to the i-th tourist area, v(i) represents the characteristic factor of the i-th tourist area, a is a positive integer, k c represents the user travel frequency weight coefficient, k m Represents the user's tourism consumption weight coefficient.

[0020] Furthermore, the historical travel tendency data feature set of the associated users is analyzed. The specific method is as follows:

[0021] Based on the historical travel tendency data feature set of the user to whom the travel information is to be pushed, the reservation time feature data and consumption feature data of the user to whom the travel information is to be pushed are clarified, the associated users of the user A to whom the travel information is to be pushed are determined, and the reservation time feature data and consumption feature data between the user A and its associated users are compared;

[0022] The comparison method of the reservation time feature data is to set a time suitability threshold F, and compare the reservation time feature data between the user A to be pushed the travel information and each of its associated users in turn. When the user A and its associated user C have more than or equal to F identical reservation dates, the historical travel tendency data feature set of the associated user C is retained. When the user A and its associated user C have less than F identical reservation dates, the historical travel tendency data feature set of the associated user C is screened out.

[0023] The method for comparing consumption characteristic data is to set a consumption contrast threshold M, compare the consumption characteristic data between user A whose travel information is to be pushed and his associated users in turn, record the average consumption amount of the user's travel as Xf, and compare the average consumption amount between user A and his associated user C through the average consumption amount contrast formula. When X(A, C) is greater than or equal to M, retain the historical travel tendency data feature set of associated user C; when X(A, C) is less than M, filter out the historical travel tendency data feature set of associated user C.

[0024] Furthermore, the average consumption amount comparison formula includes:

[0025] Using the formula It represents the average consumption amount contrast formula, where X(A, C) represents the average consumption amount contrast between user A and user C, Xf(A) represents the average consumption amount of user A's travel, Xf(C) represents the average consumption amount of user C's travel, and a is a positive integer.

[0026] Furthermore, a comprehensive tourism tendency index is set up, and the specific method is as follows:

[0027] Determine all associated users who are retained after comparing the scheduled time feature data with the consumption feature data, record them as preferred users, and comprehensively consider the historical travel tendency data feature set of the preferred users and the historical travel tendency data feature set of the users to be pushed, and set the travel tendency index;

[0028] Using the formula represents the tourism tendency index, where Yr(A) represents the tourism tendency index of user A to be pushed, r(A, j) represents the number of times user A browses the jth tourist area, F(A, j) represents the average number of times user A's preferred users browse the jth tourist area, Am(j) represents the per capita consumption amount in tourist area j, Xf(A) represents the average consumption amount of user A's travel, and k 1 represents the browsing tendency weight coefficient of the recommended user, k 2 Indicates the weight coefficient of the browsing tendency of preferred users.

[0029] The present invention also provides a tourism information recommendation system based on a big data platform, which is applied to a tourism information recommendation method based on a big data platform, including:

[0030] The user historical travel tendency data extraction module collects user historical behavior information based on the big data platform and extracts the user's historical travel tendency data features;

[0031] The associated user determination module determines the association in the travel information through data analysis based on the user's historical travel tendency data characteristics, and selects the associated users of the user to whom the travel information is to be pushed;

[0032] The travel tendency index setting module analyzes the historical travel tendency data feature set of related users, comprehensively sets the travel tendency index, and determines the travel information push plan.

[0033] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0034] 1. The present invention compares and analyzes the historical travel tendency data of the user to be pushed with the historical data of other users, and establishes a user travel similarity index. This index comprehensively considers multiple dimensions such as the user's age, location distance, and travel consumption tendency, which helps to fully reflect the similarity of travel preferences between users. By calculating the user travel similarity index, the associated users who are closest to the travel tendency of the user to be pushed can be screened out to ensure that the pushed travel information meets the actual needs of the user to be pushed;

[0035] 2. The present invention can screen out preferred users with similar travel preferences and consumption habits as the users to be pushed by comparing the scheduled time characteristic data and consumption characteristic data of the users to be pushed and their associated users. By using the historical travel tendency data feature set of the preferred users and the data of the users to be pushed for comprehensive analysis, the real needs and preferences of the users to be pushed can be captured more accurately, which helps to improve the accuracy of the recommendation.

[0036] 3. The present invention sets the tourism tendency index by comprehensively considering multiple factors such as the number of user browsing times, the average number of browsing times of preferred users, the average consumption amount per capita in the tourist area and the average consumption amount of users. It not only considers the historical tourism tendency data of the user to be pushed, but also introduces the data of the preferred users for comprehensive analysis, so as to have a more comprehensive understanding of the personalized needs of the user to be pushed. By calculating the tourism tendency indicators of users and different tourist areas, it is possible to quantify the user's preference for each tourist area, which helps to improve the user's satisfaction with the tourism recommendation information. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 A step diagram of a travel information recommendation method based on a big data platform is shown;

[0039] Figure 2 A step diagram of a method for setting a tourism trend indicator is shown;

[0040] Figure 3 The flowchart of the tourism information recommendation system based on the big data platform is shown. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] Embodiment 1: Figure 1 , Figure 2 The tourism information recommendation method based on the big data platform shown in the figure specifically includes the following steps:

[0043] Step 1: Collect user historical behavior information based on the big data platform and extract user historical travel tendency data features.

[0044] Collect user historical behavior information by connecting with the user behavior information data source through the big data platform. The user behavior information data source includes the user mobile device data source, the user social media data source, the user payment system data source and the user online reservation system data source. The user's historical movement trajectory is obtained through the user mobile device data source, for example: the user's location information is obtained through the built-in GPS, Wi-Fi and Bluetooth of the user's smart phone and tablet device, and the user's movement trajectory is recorded. From the user mobile device data source, the user's location is obtained by using technologies such as GPS, Wi-Fi and Bluetooth, and recorded as user location information data. From the user social media data source, the user's sharing, commenting, liking and other behaviors are analyzed to identify the names of attractions and regions that the user has browsed many times to obtain user preference information data. From the user payment system data source, the user's consumption records are extracted, especially those related to tourism, such as air tickets and hotel ticket consumption information, to obtain user consumption habit information data. From the user online reservation system data source, the user's reservation record, including the reservation time, location, and service type, is obtained, and recorded as user reservation information data;

[0045] Determine that a cycle duration is T, and divide each cycle duration into m time periods, and the duration of each time period is t. Within the n cycle duration, collect real-time or historical user behavior information data from various user behavior information data sources, and use data mining technology to extract data features of user historical travel tendencies from user location information data, user preference information data, user consumption habit information data, and user reservation information data;

[0046] The set D(A) = {u(A), o(A), c(A)} is used to represent the historical travel tendency data feature set of user A, including travel destination feature data, reservation time feature data, and consumption feature data, where u(A) represents the travel destination feature data set of user A. The travel destination feature data includes the type of travel area, the location distance of the travel area, and the per capita consumption amount of the travel area. The type of travel area refers to the cultural type travel area, the ecological type travel area, and the leisure type travel area classified by the travel theme. u(A) = {(s(A, j), r(A, j ))|j=1,...,g}wherein j indicates that user A browsed the jth tourist area, g indicates that user A browsed g tourist destinations in total, s(A,j) indicates that user A browsed the jth tourist area type, r(A,j) indicates the number of times user A browsed the jth tourist area, the reservation time characteristic data refers to the date that appears more than L times in the user's reservation information data, and such dates are recorded as the reservation dates of user A, such as weekends, holidays and other fixed dates. The consumption characteristic data refers to the amount of consumption of the user for each travel obtained by analyzing the user's consumption habit information data.

[0047] Step 2: Based on the characteristics of the user's historical travel tendency data, determine the relevance of travel information through data analysis, and screen the related users of the users to whom the travel information is to be pushed.

[0048] Identify the users to whom travel information is to be pushed, establish a user travel similarity index based on the historical travel tendency data of the users to be pushed and the historical travel tendency data of other users, and judge the degree of association between users based on the user travel similarity index. The specific calculation formula of the user travel similarity index is as follows:

[0049] Xs(A,B)=[kag*ag(A)-ag(B)|+klo*lo(A,B)]*sd(A,B);

[0050] Among them, Xs(A, B) represents the user travel similarity index between A and B, A represents the user A to be pushed, B represents other user B, ag(A) represents the age of user A, ag(B) represents the age of user B, lo(A, B) represents the location distance between user A and user B, sd(A, B) represents the travel consumption tendency similarity index between user A and user B, k ag Represents the user age weight coefficient, k lo Represents the user distance weight coefficient. The smaller the value of Xs(A, B), the higher the correlation degree of the travel tendency data between user A and user B.

[0051] Furthermore, the specific calculation formula of the similarity index of the travel consumption tendency of user A and user B is as follows:

[0052]

[0053] Wherein, sd(A, B) represents the similarity index of tourism consumption tendency between A and B, A represents the user A to be pushed, B represents other user B, i represents the type of the i-th tourist area, H represents a total of H types of tourist destination feature data, A(i) represents the number of times the user A to be pushed goes to the i-th tourist area within a cycle, B(i) represents the number of times the user B goes to the i-th tourist area within a cycle, mA(i) represents the average consumption amount of the user A to be pushed when going to the i-th tourist area, mB(i) represents the average consumption amount of the user B when going to the i-th tourist area, v(i) represents the characteristic factor of the i-th tourist area, when A(i)-B(i) is greater than 0, v(i) is equal to 1, when A(i)-B(i) is less than 0, v(i) is equal to 0, a is a positive integer, k c represents the user travel frequency weight coefficient, k m Represents the user's tourism consumption weight coefficient.

[0054] Set a user travel similarity index threshold Q, calculate the user travel similarity index between the user to be pushed and other users, filter out user travel similarity indexes that are less than or equal to the threshold Q, and determine other users who meet the condition, and record such users as associated users of the user whose travel information is to be pushed.

[0055] Step 3: Analyze the historical travel tendency data feature set of related users, comprehensively set travel tendency indicators, and determine the travel information push plan.

[0056] Based on the historical travel tendency data feature set of the user to whom the travel information is to be pushed, the reservation time feature data and consumption feature data of the user to whom the travel information is to be pushed are clarified, the associated users of the user A to whom the travel information is to be pushed are determined, and the reservation time feature data and consumption feature data between the user A and its associated users are compared;

[0057] The comparison method of the reservation time feature data is to set a time suitability threshold F, and compare the reservation time feature data between the user A to be pushed the travel information and each of its associated users in turn. When the user A and its associated user C have more than or equal to F identical reservation dates, the historical travel tendency data feature set of the associated user C is retained. When the user A and its associated user C have less than F identical reservation dates, the historical travel tendency data feature set of the associated user C is screened out.

[0058] The comparison method of consumption characteristic data is to set the consumption contrast threshold M, and compare the consumption characteristic data between the user A to be pushed travel information and each of its associated users in turn, and record the average consumption amount of the user's travel as Xf. Compare the average consumption amount between user A and his associated user C. When X(A, C) is greater than or equal to M, retain the historical travel tendency data feature set of associated user C. When X(A, C) is less than M, filter out the historical travel tendency data feature set of associated user C.

[0059] Determine all the associated users who are retained after comparing the scheduled time feature data with the consumption feature data, record them as preferred users, and comprehensively consider the historical travel tendency data feature set of the preferred users and the historical travel tendency data feature set of the users to be pushed to set the travel tendency index. The specific formula of the travel tendency index is as follows:

[0060]

[0061] Where Yr(A) represents the travel tendency index of user A to be pushed, r(A, j) represents the number of times user A browses the jth tourist area, F(A, j) represents the average number of times user A's preferred users browse the jth tourist area, Am(j) represents the per capita consumption amount in tourist area j, Xf(A) represents the average consumption amount of user A's travel, and k 1 represents the browsing tendency weight coefficient of the recommended user, k 2 Indicates the weight coefficient of the browsing tendency of preferred users.

[0062] The tourism tendency indexes of the user and different tourist areas are calculated, and E tourist areas with the highest tourism tendency indexes are selected as push objects. In this embodiment, E is set to be 5.

[0063] Embodiment 2: Figure 3 The tourism information recommendation system based on the big data platform shown in the figure specifically includes:

[0064] The user historical travel tendency data extraction module is used to connect with the user behavior information data source through the big data platform to collect the user's historical behavior information. The user behavior information data source includes the user mobile device data source, the user social media data source, the user payment system data source and the user online reservation system data source. The user's historical movement trajectory is obtained through the user mobile device data source, for example: the user's location information is obtained through the built-in GPS, Wi-Fi and Bluetooth of the user's smart phone and tablet device, and the user's movement trajectory is recorded. From the user mobile device data source, the user's location is obtained by using technologies such as GPS, Wi-Fi and Bluetooth, and recorded as user location information data. From the user social media data source, the user's sharing, commenting, liking and other behaviors are analyzed to identify the names of attractions and regions that the user has browsed many times to obtain user preference information data. From the user payment system data source, the user's consumption records, especially those related to travel, such as air tickets and hotel ticket consumption information, are extracted to obtain user consumption habit information data. From the user online reservation system data source, the user's reservation record, including the reservation time, location, and service type, is obtained, and recorded as user reservation information data;

[0065] Determine that a cycle duration is T, and divide each cycle duration into m time periods, and the duration of each time period is t. Within the n cycle duration, collect real-time or historical user behavior information data from various user behavior information data sources, and use data mining technology to extract data features of user historical travel tendencies from user location information data, user preference information data, user consumption habit information data, and user reservation information data;

[0066] The set D(A) = {u(A), o(A), c(A)} is used to represent the historical travel tendency data feature set of user A, including travel destination feature data, reservation time feature data, and consumption feature data, where u(A) represents the travel destination feature data set of user A. The travel destination feature data includes the type of travel area, the location distance of the travel area, and the per capita consumption amount of the travel area. The type of travel area refers to the cultural type travel area, the ecological type travel area, and the leisure type travel area classified by the travel theme. u(A) = {(s(A, j), r(A, j ))|j=1,...,g}wherein j indicates that user A browsed the jth tourist area, g indicates that user A browsed g tourist destinations in total, s(A,j) indicates that user A browsed the jth tourist area type, r(A,j) indicates the number of times user A browsed the jth tourist area, the reservation time characteristic data refers to the date that appears more than L times in the user's reservation information data, and such dates are recorded as the reservation dates of user A, such as weekends, holidays and other fixed dates. The consumption characteristic data refers to the amount of consumption of the user for each travel obtained by analyzing the user's consumption habit information data.

[0067] The associated user determination module is used to identify the users to whom travel information is to be pushed, establish a user travel similarity index based on the historical travel tendency data of the users to be pushed and the historical travel tendency data of other users, and judge the degree of association between users based on the user travel similarity index. The specific calculation formula of the user travel similarity index is as follows:

[0068] Xs(A,B)=[kag*ag(A)-ag(B)|+klo*lo(A,B)]*sd(A,B);

[0069] Among them, Xs(A, B) represents the user travel similarity index between A and B, A represents the user A to be pushed, B represents other user B, ag(A) represents the age of user A, ag(B) represents the age of user B, lo(A, B) represents the location distance between user A and user B, sd(A, B) represents the travel consumption tendency similarity index between user A and user B, k ag Represents the user age weight coefficient, k lo Represents the user distance weight coefficient. The smaller the value of Xs(A, B), the higher the correlation degree of the travel tendency data between user A and user B.

[0070] Furthermore, the specific calculation formula for the similarity index of the travel consumption tendency of user A and user B within a period is as follows:

[0071]

[0072] Wherein, sd(A, B) represents the similarity index of tourism consumption tendency between A and B, A represents the user A to be pushed, B represents other user B, i represents the type of the i-th tourist area, H represents a total of H types of tourist destination feature data, A(i) represents the number of times the user A to be pushed goes to the i-th tourist area within a cycle, B(i) represents the number of times the user B goes to the i-th tourist area within a cycle, mA(i) represents the average consumption amount of the user A to be pushed when going to the i-th tourist area, mB(i) represents the average consumption amount of the user B when going to the i-th tourist area, v(i) represents the characteristic factor of the i-th tourist area, when A(i)-B(i) is greater than 0, v(i) is equal to 1, when A(i)-B(i) is less than 0, v(i) is equal to 0, a is a positive integer, k c represents the user travel frequency weight coefficient, k m Represents the user's tourism consumption weight coefficient.

[0073] Set a user travel similarity index threshold Q, calculate the user travel similarity index between the user to be pushed and other users, filter out user travel similarity indexes that are less than or equal to the threshold Q, and determine other users who meet the condition, and record such users as associated users of the user whose travel information is to be pushed.

[0074] A travel tendency index setting module, based on the historical travel tendency data feature set of the user to be pushed travel information, clarifies the booking time feature data and consumption feature data of the user to be pushed travel information, determines the associated users of user A to be pushed travel information, and compares the booking time feature data and consumption feature data between user A and its associated users;

[0075] The comparison method of the reservation time feature data is to set a time suitability threshold F, and compare the reservation time feature data between the user A to be pushed the travel information and each of its associated users in turn. When the user A and its associated user C have more than or equal to F identical reservation dates, the historical travel tendency data feature set of the associated user C is retained. When the user A and its associated user C have less than F identical reservation dates, the historical travel tendency data feature set of the associated user C is screened out.

[0076] The comparison method of consumption characteristic data is to set the consumption contrast threshold M, compare the consumption characteristic data between the user A to be pushed travel information and each of its associated users in turn, record the average consumption amount of the user's travel as Xf, and use the formula Compare the average consumption amount between user A and his associated user C. When X(A, C) is greater than or equal to M, retain the historical travel tendency data feature set of associated user C. When X(A, C) is less than M, filter out the historical travel tendency data feature set of associated user C.

[0077] Determine all the associated users who are retained after comparing the scheduled time feature data with the consumption feature data, record them as preferred users, and comprehensively consider the historical travel tendency data feature set of the preferred users and the historical travel tendency data feature set of the users to be pushed to set the travel tendency index. The specific formula of the travel tendency index is as follows:

[0078]

[0079] Where Yr(A) represents the travel tendency index of user A to be pushed, r(A, j) represents the number of times user A browses the jth tourist area, F(A, j) represents the average number of times user A's preferred users browse the jth tourist area, Am(j) represents the per capita consumption amount in tourist area j, Xf(A) represents the average consumption amount of user A's travel, and k 1 represents the browsing tendency weight coefficient of the recommended user, k 2 Indicates the weight coefficient of the browsing tendency of preferred users.

[0080] Calculate the user's travel tendency index for different tourist areas, and select E tourist areas with the highest travel tendency index as push targets.

[0081] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

[0082] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A tourism information recommendation method based on a big data platform, characterized in that: include: Step 1: Collect user historical behavior information based on the big data platform and extract user historical travel tendency data features; By connecting the big data platform with the user behavior information data source, the user historical behavior information is collected. The user historical behavior information includes user location information data, user preference information data, user consumption habit information data and user reservation information data. The duration T is determined as a cycle, and each cycle is divided into m time periods. The duration of each time period is t. Within the n cycle duration, real-time or historical user behavior information data is collected from various user behavior information data sources. The data mining technology is used to extract the data features of the user's historical travel tendency from the user's historical behavior information. Step 2: Based on the historical travel tendency data characteristics of users, determine the relevance of travel information through data analysis, and select the related users of the users to whom the travel information is to be pushed; Step 3: Analyze the historical travel tendency data feature set of related users, comprehensively set travel tendency indicators, and determine the travel information push plan.

2. The method for recommending tourism information based on a big data platform according to claim 1, characterized in that: Using data mining technology, we extract the data features of users’ historical travel tendencies from their historical behavior information. The specific method is as follows: The set D(A)={u(A), o(A), c(A)} is used to represent the historical travel tendency data feature set of user A, where u(A) represents the travel destination feature data set of user A, u(A)={(s(A,j), r(A,j))|j=1,...,g}, where j represents user A browsing the jth tourist area, g represents user A browsing g tourist destinations in total, s(A,j) represents user A browsing the jth tourist area type, r(A,j) represents the number of times user A browsed the jth tourist area, the reservation time feature data refers to the date that appears more than L times in the user's reservation information data, and this type of date is recorded as the reservation date of user A, and the consumption feature data refers to the amount of consumption of the user for each travel obtained by analyzing the user's consumption habit information data.

3. The method for recommending tourism information based on a big data platform according to claim 1, characterized in that: Filter the associated users of the user whose travel information is to be pushed. The specific method is: Identify the users to whom travel information is to be pushed, establish a user travel similarity index based on the historical travel tendency data of the users to be pushed and the historical travel tendency data of other users, set a user travel similarity index threshold Q, calculate the user travel similarity index between the users to be pushed and other users, filter out user travel similarity indexes that are less than or equal to the threshold Q, and determine other users who meet the condition, and record such users as associated users of the users to whom travel information is to be pushed.

4. The method for recommending tourism information based on a big data platform according to claim 3 is characterized in that: The user travel similarity index is established based on the historical travel tendency data of the user to be pushed and the historical travel tendency data of other users. The specific method is as follows: The formula Xs(A, B) = [kag*ag(A)-ag(B)|+klo*lo(A, B)]*sd(A, B) is used to represent the user travel similarity index, where Xs(A, B) represents the user travel similarity index between A and B, A represents the user A to be pushed, B represents other user B, ag(A) represents the age of user A, ag(B) represents the age of user B, lo(A, B) represents the location distance between user A and user B, sd(A, B) represents the travel consumption tendency similarity index between user A and user B, and k ag Represents the user age weight coefficient, k lo Represents the user distance weight coefficient. The smaller the value of Xs(A, B), the higher the correlation degree of the travel tendency data between user A and user B.

5. The method for recommending tourism information based on a big data platform according to claim 4, characterized in that: Tourism consumption tendency similarity index, including: Using the formula represents the similarity index of tourism consumption tendency, where sd(A, B) represents the similarity index of tourism consumption tendency between A and B, A represents the user A to be pushed, B represents other user B, i represents the type of the i-th tourist area, H represents a total of H types of tourist destination feature data, A(i) represents the number of times the user A to be pushed goes to the i-th tourist area within a cycle, B(i) represents the number of times the user B goes to the i-th tourist area within a cycle, mA(i) represents the average consumption amount of the user A to be pushed going to the i-th tourist area, mB(i) represents the average consumption amount of the user B going to the i-th tourist area, v(i) represents the characteristic factor of the i-th tourist area, a is a positive integer, k c represents the user travel frequency weight coefficient, k m Represents the user's tourism consumption weight coefficient.

6. The method for recommending tourism information based on a big data platform according to claim 1, characterized in that: Analyze the historical travel tendency data feature set of the associated users. The specific method is as follows: Based on the historical travel tendency data feature set of the user to whom the travel information is to be pushed, the reservation time feature data and consumption feature data of the user to whom the travel information is to be pushed are clarified, the associated users of the user A to whom the travel information is to be pushed are determined, and the reservation time feature data and consumption feature data between the user A and its associated users are compared; The comparison method of the reservation time feature data is to set a time suitability threshold F, and compare the reservation time feature data between the user A to be pushed the travel information and each of its associated users in turn. When the user A and its associated user C have more than or equal to F identical reservation dates, the historical travel tendency data feature set of the associated user C is retained. When the user A and its associated user C have less than F identical reservation dates, the historical travel tendency data feature set of the associated user C is screened out. The method for comparing consumption characteristic data is to set a consumption contrast threshold M, compare the consumption characteristic data between user A whose travel information is to be pushed and his associated users in turn, record the average consumption amount of the user's travel as Xf, and compare the average consumption amount between user A and his associated user C through the average consumption amount contrast formula. When X(A, C) is greater than or equal to M, retain the historical travel tendency data feature set of associated user C; when X(A, C) is less than M, filter out the historical travel tendency data feature set of associated user C.

7. The method for recommending tourism information based on a big data platform according to claim 6, characterized in that: The average consumption amount comparison formula includes: Using the formula It represents the average consumption amount contrast formula, where X(A, C) represents the average consumption amount contrast between user A and user C, Xf(A) represents the average consumption amount of user A's travel, Xf(C) represents the average consumption amount of user C's travel, and a is a positive integer.

8. The method for recommending tourism information based on a big data platform according to claim 1, characterized in that: Comprehensively set tourism tendency indicators, the specific methods are as follows: Determine all associated users who are retained after comparing the scheduled time feature data with the consumption feature data, record them as preferred users, and comprehensively consider the historical travel tendency data feature set of the preferred users and the historical travel tendency data feature set of the users to be pushed, and set the travel tendency index; Using the formula represents the tourism tendency index, where Yr(A) represents the tourism tendency index of the user A to be recommended, r(A, j) represents the number of times user A browses the jth tourist area, F(A, j) represents the average number of times the preferred users of user A browse the jth tourist area, Am(j) represents the per capita consumption amount in tourist area j, Xf(A) represents the average consumption amount of user A's travel, k1 represents the browsing tendency weight coefficient of the user to be recommended, and k2 represents the browsing tendency weight coefficient of the preferred user.

9. A tourism information recommendation system based on a big data platform, applied to the tourism information recommendation method based on a big data platform according to any one of claims 1 to 8, characterized in that: include: The user historical travel tendency data extraction module collects user historical behavior information based on the big data platform and extracts the user's historical travel tendency data features; The associated user determination module determines the association in the travel information through data analysis based on the user's historical travel tendency data characteristics, and selects the associated users of the user to whom the travel information is to be pushed; The travel tendency index setting module analyzes the historical travel tendency data feature set of related users, comprehensively sets the travel tendency index, and determines the travel information push plan.

Images