Automatic generation method of travel guide based on artificial intelligence

Abstract

The present invention relates to an automatic generation method of a travel guide based on artificial intelligence, by establishing a "holiday-emotion" model, based on the basic information of the user and the input number of days to play: the "holiday-emotion" model automatically predicts the user's daily emotional index data during the play period, after the user inputs the destination city of the tour, automatically pre-screens the scenic spots of the city and analyzes in real time to obtain the recommended degree of the scenic spots, then matches the scenic spots with the corresponding recommended degree of play according to the user's daily emotional index data, selects nearby hotels according to the matched scenic spots, finally grabs the purchase link of the scenic spot ticket and the booking link of the recommended hotel, uses the AI ​​generation model to generate a text introduction of the travel schedule according to the departure date, return date, and target scenic spot, and generates a paid summary order in combination with the number of people playing. The method saves time and effort, is simple and fast, realizes the user's "one-click" travel guide generation and scenic spot and hotel reservation, and can greatly improve the user's travel experience.

Description

Technical Field

[0001] The present invention relates to the field of Internet information technology processing, and in particular to a method for automatically generating a travel guide based on artificial intelligence. Background Art

[0002] With the booming tourism industry and the increasing demand for personalized tourism, how to quickly and accurately provide travelers with personalized travel guides has become an important issue in the development of the industry. Traditional travel guides mostly rely on manual search of travel information on the Internet, and then spend time compiling travel itineraries based on the required information. This process is time-consuming and difficult to update in real time, making it difficult to meet users' growing demand for personalization and immediacy. Even if users find or make a suitable travel guide, they still need to book tickets and hotel rooms for each attraction one by one, which is time-consuming and laborious.

[0003] In recent years, although some travel recommendation systems based on rules or simple algorithms have emerged, these systems often only recommend a single attraction or the most popular attractions, lack sufficient flexibility and depth, and fail to effectively integrate different types of tourism resource information, such as attractions, hotels, etc.

[0004] Based on the above reasons, there is an urgent need to provide an intelligent tool to automatically generate travel guides for users and automatically generate summary orders for attractions and hotels based on the travel guides. Summary of the invention

[0005] Based on the above statements, the present invention provides an artificial intelligence-based automatic travel guide generation method, which can automatically generate a travel guide for a user and automatically generate a summary order of attractions and hotels based on the travel guide.

[0006] The technical solution of the present invention to solve the above technical problems is as follows:

[0007] The method for automatically generating travel guides based on artificial intelligence is as follows: obtaining tourist attractions in each city, and obtaining the levels and themes of these tourist attractions, and then calibrating the physical exertion values ​​and fun values ​​of these tourist attractions; building a "holiday-emotion" model based on big data, and the "holiday-emotion" model is used to predict the user's daily mood index under different holiday durations; obtaining the user's basic information and guiding the user to input the number of days for travel, the number of visitors and the destination city, and pre-screening the available attractions from the tourist attractions in the destination city according to the level and theme of the attractions, and using the "holiday-emotion" model to predict the user's daily mood index during the travel according to the number of days for travel entered by the user; determining the emotional matching degree of each available attraction according to the theme, physical exertion value and fun value, and placing the available attractions one by one The emotional matching degree is matched with the emotional index of each day during the tour. If the match is successful, the current attractions for selection will be marked as target attractions until the number of target attractions is equal to the number of days for the tour; a travel strategy AI generation model is established, and the name of the target attraction, departure date and return date are input into the travel strategy AI generation model. The travel strategy AI generation model automatically captures the information of the target attraction and obtains the recommended hotels nearby based on the geographical location of the target attraction, and captures the purchase link of the target attraction ticket and the booking link of the recommended hotel; the travel strategy AI generation model generates a text introduction to the travel itinerary based on the departure date, return date and target attraction, and generates a payment summary order based on the number of visitors to prompt the user whether to confirm the travel itinerary. After the user chooses to confirm, the order payment page will be jumped to prompt the user to pay.

[0008] As a preferred solution: when calibrating the physical exertion value of a tourist attraction, the tourist theme of the tourist attraction is obtained and the comment text of the tourist attraction is captured, key words are extracted from the comment text, and the physical exertion value of the tourist attraction is determined according to the theme of the tourist attraction and the frequency of occurrence of the key words in the comment text; when calibrating the fun value of the tourist attraction, the tourist theme of the tourist attraction is obtained and the comment text of the tourist attraction is captured, key words are extracted from the comment text, and the fun value of the tourist attraction is determined according to the theme of the tourist attraction and the frequency of occurrence of the key words in the comment text.

[0009] As a preferred solution, the "holiday-emotion" model presets standard emotion index data for different travel days, that is, the emotion index for each day. When predicting the user's emotion index during the travel period, the standard emotion index data matching the travel days is first obtained, and then the weather data for each day during the travel period is captured, and the emotion index for each day is corrected according to the weather conditions.

[0010] As a preferred solution: when determining the emotional matching degree of a scenic spot, a preset calculation formula for the recommended play value is used: D=k(aN+bM), where k is the calculation coefficient corresponding to the play theme, N is the physical consumption value, M is the fun value, and a and b are preset calculation coefficients; the correspondence between the recommended value of the scenic spot and the emotional index is pre-calibrated, that is, the recommended play value of the scenic spot, and the applicable emotional index is automatically output after the recommended play value of the scenic spot is calculated.

[0011] As a preferred solution, the "holiday-emotion" model presets standard emotion index data for tourists of different age groups on different days of play. After guiding the user to input the number of days of play, the number of visitors, and whether there are elderly people or children, the standard emotion index data of the corresponding age group under the current number of days of play is retrieved according to the user's age group and whether there are elderly people or children. The average emotion index of all members per day during the play period is then calculated, and finally the comprehensive emotion index of all members during the play period is considered for subsequent attraction matching.

[0012] As a preferred solution: after the user has confirmed the travel itinerary and paid the fees several times, the frequency of different attraction themes and hotel grades in the travel itinerary is recorded and analyzed, and the user's travel and consumption preferences are derived based on this; the next time a travel itinerary is generated, attractions and hotels are selected based on the user's travel and consumption preferences.

[0013] As a preferred solution: When generating paid summary orders, the travel guide AI generation model captures the purchase links of attraction tickets and recommended hotel booking links on different platforms and compares prices to find the one with the lowest price and captures the combined itinerary paid summary order.

[0014] Compared with the prior art, the technical solution of the present application has the following beneficial technical effects: the present application establishes a "holiday-emotion" model, and uses it according to the user's basic information and the number of days of play input: the "holiday-emotion" model automatically predicts the user's daily emotional index data during the play period. After the user enters the tourist destination city, the city's scenery is automatically pre-screened and analyzed in real time to obtain the recommended degree of the scenic spot. Then, the scenic spots with corresponding recommended degrees of play are matched according to the user's daily emotional index data. According to the matched scenic spots and nearby hotels, the purchase link of the scenic spot ticket and the booking link of the recommended hotel are finally captured, and the AI ​​generation model is used to generate a text introduction of the travel itinerary according to the departure date, return date, and target scenic spot, and a paid summary order is generated in combination with the number of visitors. Using this method, users do not need to check online information, make their own guides, and book scenic spot tickets and hotels by themselves. It saves time and effort, is simple and fast, and enables users to generate "one-click" travel guides, which can greatly enhance the user's travel experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 4 is a flow chart of the method in this embodiment. DETAILED DESCRIPTION

[0016] The method for automatically generating a travel guide based on artificial intelligence is as follows: obtaining tourist attractions in each city, and obtaining the levels of these tourist attractions (such as 3A scenic spots, 4A scenic spots, 5A scenic spots), tourist themes (such as mountain climbing, boat riding, commercial streets, amusement parks, self-driving tours, forest scenic spots, etc.), and then calibrating the physical exertion values ​​and fun values ​​of these tourist attractions.

[0017] A questionnaire was conducted on the tourism volunteers, and the volunteers were asked about their mood, vitality, interest and other information every day during the tour, and the daily emotional index was scored accordingly. For example, the mood column has several options such as "excited", "good", "average", and "depressed"; the vitality column has several options such as "abundant", "good", "average", and "lack"; the interest column has several options such as "still want to continue playing", "can play or not", and "don't want to continue playing". The emotional index can be obtained by scoring the volunteers according to the combination of their choices in the three columns (each combination corresponds to a score). The basic big data was obtained by surveying a large number of volunteers through the above method.

[0018] Since the user's emotional index will change every day during the tour, and sometimes the change is large, it is necessary to select attractions and specify a travel schedule for the user based on the change in the user's emotional index, so that the user can get a better travel experience.

[0019] In this embodiment, based on basic big data, a "holiday-emotion" model is built according to the age, number of travel days and daily emotion index of a large number of volunteers during the travel period. After training and optimization, the "holiday-emotion" model can automatically calculate the daily emotion index of users according to different user ages and different number of travel days.

[0020] The user's basic information is obtained and the user is guided to input the number of days for travel, the number of tourists and the destination city. The attractions for selection are pre-screened from the tourist attractions in the destination city according to the level and the theme of the attractions. In this embodiment, attractions with higher levels and containing different themes are given priority, which can ensure a better travel experience and indicate the richness of the travel content.

[0021] The "holiday-emotion" model is used to predict the user's emotion index for each day during the game according to the user's age and the number of days entered.

[0022] Determine the emotional matching degree of each optional scenic spot according to the play theme, physical consumption value and fun value, match the emotional matching degree of the optional scenic spot with the emotional index of each day during the play one by one, and if the match is successful, mark the current optional scenic spot as the target scenic spot until the number of target scenic spots is equal to the number of days of play; establish a travel strategy AI generation model, input the name of the target scenic spot, departure date and return date into the travel strategy AI generation model, the travel strategy AI generation model automatically captures the information of the target scenic spot and obtains the recommended hotels nearby according to the geographical location of the target scenic spot, and captures the purchase link of the target scenic spot ticket and the booking link of the recommended hotel; the travel strategy AI generation model generates a text introduction of the travel itinerary based on the departure date, return date and target scenic spot, and generates a payment summary order based on the number of visitors to prompt the user whether to confirm this travel itinerary. After the user chooses to confirm, the order payment page will be jumped to prompt the user to pay.

[0023] In this embodiment: when calibrating the physical exertion value of a tourist attraction, the tourist attraction theme is obtained and the comments of the tourist attraction are captured, and keywords are extracted from the comments, and the physical exertion value of the tourist attraction is determined according to the theme of the tourist attraction and the proportion of the keywords in the comments. For example, if the total number of captured comments is X, and the number of keywords such as "tired", "sour", and "shaking" is J, then the physical exertion value N = f1J / X, where f1 is the value corresponding to the current tourist attraction theme.

[0024] When calibrating the fun value of a tourist attraction, the tourist attraction's theme is obtained and the comments of the tourist attraction are captured. Keywords are extracted from the comments, and the fun value of the tourist attraction is determined based on the theme of the tourist attraction and the frequency of the keywords in the comments. For example, if the total number of captured comments is X, and the number of keywords such as "fun", "interesting", and "stimulating" is K, then the fun value M = f2K / X, where f2 is the value corresponding to the current tourist theme.

[0025] In this embodiment, the "holiday-emotion" model is preset with standard emotion index data for different play days, that is, the emotion index for each day. When predicting the emotion index of the user during the play period, the standard emotion index data matching the play days is first obtained, and then the weather data of each day during the play period is captured, and the emotion index of each day is corrected according to the weather conditions. For example, sunny days, cloudy days and rainy days are respectively preset with corresponding correction coefficients, and the corrected emotion index can be calculated using the correction coefficient and the standard emotion index.

[0026] In this embodiment: when determining the emotional matching degree of the scenic spot, the calculation formula for the recommended play value is preset: D=w(aN+bM), where w is the calculation coefficient corresponding to the play theme, N is the physical consumption value, M is the fun value, and a and b are preset calculation coefficients; the correspondence between the recommended value of the scenic spot and the emotional index is pre-calibrated, that is, the recommended play value of the scenic spot (for example, when the user's emotional index is high, it is randomly matched with a play scenic spot, when the user's emotional index is average, it is matched with an scenic spot with an average play recommendation value, for example, when the user's emotional index is low, it is matched with an scenic spot with a high play recommendation value, so that the user can have), and after calculating the recommended play value of the scenic spot, the applicable emotional index is automatically output.

[0027] In this embodiment, the "holiday-emotion" model presets standard emotion index data for tourists of different age groups on different days of play, and guides the user to input the number of days of play, the number of visitors, and whether there are elderly people or children. Then, the standard emotion index data of the corresponding age group under the current number of days of play is retrieved according to the user's age group and whether there are children or elderly people; then the average emotion index of all members during each day of play is calculated, and finally the comprehensive emotion index of all members during the play period is considered, and subsequent scenic spot matching is carried out based on this.

[0028] In this embodiment: after the user confirms the travel itinerary and pays the fee for many times, the frequency of different scenic spot themes and hotel grades in the travel itinerary is recorded and analyzed, and the user's travel and consumption preferences are obtained based on this; when generating a travel itinerary next time, scenic spots and hotels are selected based on the user's travel preferences and consumption preferences. This measure can make the system generate travel itineraries that are more and more in line with the user's preferences, further improving the user experience.

[0029] In this embodiment, when generating a paid summary order, the travel guide AI generation model grabs the purchase links of scenic spot tickets and recommended hotel booking links on different platforms and compares the prices, finds the one with the lowest price and grabs the combined itinerary paid summary order. This measure can reduce the user's travel expenses.

[0030] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for automatically generating travel strategies based on artificial intelligence, characterized in that: Obtain tourist attractions in each city, and obtain the levels and themes of these tourist attractions, and then calibrate the physical exertion value and fun value of these tourist attractions; build a "holiday-emotion" model based on big data, and the "holiday-emotion" model is used to predict the user's daily mood index under different holiday durations; obtain basic user information and guide the user to enter the number of days to play, the number of visitors and the destination city, pre-screen the available attractions from the tourist attractions in the destination city according to the level and theme of the attractions, and use the "holiday-emotion" model to predict the user's daily mood index during the play according to the number of days to play entered by the user; determine the mood matching degree of each available attraction according to the theme, physical exertion value and fun value, and compare the mood matching degree of the available attractions with the tourist attractions one by one The emotional index of each day during the period is matched. If the match is successful, the current selectable attractions will be marked as target attractions until the number of target attractions equals the number of days of play; a travel strategy AI generation model is established, and the name of the target attraction, departure date and return date are input into the travel strategy AI generation model. The travel strategy AI generation model automatically captures the information of the target attraction and obtains the recommended hotels nearby based on the geographical location of the target attraction, and captures the purchase link of the target attraction ticket and the booking link of the recommended hotel; the travel strategy AI generation model generates a text introduction to the travel itinerary based on the departure date, return date and target attraction, and generates a payment summary order based on the number of visitors to prompt the user whether to confirm the travel itinerary. After the user chooses to confirm, the order payment page will be jumped to prompt the user to pay.

2. The method for automatically generating travel strategies based on artificial intelligence according to claim 1 is characterized in that: When calibrating the physical exertion value of a tourist attraction, the tourist theme of the tourist attraction is obtained and the comment text of the tourist attraction is captured, and keywords are extracted from the comment text. The physical exertion value of the tourist attraction is determined according to the theme of the tourist attraction and the frequency of occurrence of the keywords in the comment text; When calibrating the fun value of a tourist attraction, the tourist theme of the tourist attraction is obtained and the comment text of the tourist attraction is captured, and keywords are extracted from the comment text. The fun value of the tourist attraction is determined based on the theme of the tourist attraction and the frequency of occurrence of the keywords in the comment text.

3. The method for automatically generating a travel guide based on artificial intelligence according to claim 1 is characterized in that: The "holiday-emotion" model presets standard emotion index data for different play days, that is, the emotion index for each day. When predicting the emotion index during the user's play period, the standard emotion index data matching the play days is first obtained, and then the weather data for each day during the play period is captured, and the emotion index of each day is corrected according to the weather conditions.

4. The method for automatically generating travel strategies based on artificial intelligence according to claim 1 is characterized in that: When determining the emotional matching degree of a scenic spot, the calculation formula for the recommended play value is preset: D=k(aN+bM), where k is the calculation coefficient corresponding to the play theme, N is the physical consumption value, M is the fun value, and a and b are preset calculation coefficients; the correspondence between the recommended value of the scenic spot and the emotional index is pre-calibrated, that is, the recommended play value of the scenic spot, and the applicable emotional index is automatically output after the recommended play value of the scenic spot is calculated.

5. The method for automatically generating a travel guide based on artificial intelligence according to claim 1 is characterized in that: The "holiday-emotion" model presets standard emotion index data for tourists of different age groups on different days of play. After guiding the user to input the number of days of play, the number of visitors, and whether there are elderly people or children, the standard emotion index data of the corresponding age group under the current number of days of play is retrieved according to the user's age group and whether there are children or elderly people; the average emotion index of all members during each day of play is calculated, and finally the comprehensive emotion index of all members during the play period is considered, and subsequent scenic spot matching is carried out accordingly.

6. The method for automatically generating a travel guide based on artificial intelligence according to claim 1 is characterized in that: After the user has confirmed the travel itinerary and paid the fees several times, the frequency of different attraction themes and hotel grades in the travel itinerary is recorded and analyzed, and the user's travel and consumption preferences are derived based on this. The next time a travel itinerary is generated, attractions and hotels are selected based on the user's travel preferences and consumption preferences.

7. The method for automatically generating travel strategies based on artificial intelligence according to claim 1 is characterized by: When generating paid summary orders, the travel guide AI generation model captures the purchase links of attraction tickets and recommended hotel booking links on different platforms and compares the prices to find the one with the lowest price and captures the paid summary order for the combined itinerary.

Images