Deep learning-based tourism management market demand prediction method and system

Abstract

The invention provides a tourism management market demand prediction method and system based on deep learning. The method comprises the steps of obtaining tourism related information corresponding to a preset tourism management platform; the tourism management platform is used for managing a plurality of preset scenic spots; performing sentiment analysis on the tourism related information based on a natural language processing technology to obtain positive sentiment information, negative sentiment information and tourist interested information; obtaining travel information corresponding to the travel related information; the travel information comprises multiple pieces of travel preference information and a travel group type corresponding to each piece of travel preference information; acquiring scenic spot access information of each preset scenic spot corresponding to the tourism related information; generating market demand information in a preset time period according to the positive emotion information, the negative emotion information, the tourist interested information, the travel preference information, the travel group type and the scenic spot access information; the market demand information at least comprises tourist number information, popular time information and tourist preference information of the preset scenic spot.

Description

Technical Field

[0001] The present invention belongs to the field of deep learning, and specifically relates to a tourism management market demand forecasting method and system based on deep learning. Background Art

[0002] With the rapid development of tourism and the continuous advancement of information technology, tourism management platforms play an important role in optimizing tourism resource allocation, improving tourist experience, and achieving sustainable tourism development. However, traditional tourism market demand forecasting methods have many limitations. They mainly rely on historical statistical data and simple trend analysis, and fail to fully utilize and deeply explore the value of massive tourism information. These methods often ignore important factors such as tourists' emotions, travel preferences, and group characteristics, resulting in poor accuracy and timeliness of forecast results. Traditional tourism market demand forecasting methods mainly rely on a single data source and a simple statistical model, which makes it difficult to capture complex market demand changes.

[0003] Therefore, there is an urgent need for a tourism management market demand forecasting method based on deep learning to address the shortcomings of traditional methods and contribute to the sustainable development of the tourism industry. Summary of the invention

[0004] This application provides a tourism management market demand forecasting method and system based on deep learning, which aims to solve the many limitations of traditional tourism market demand forecasting methods, which mainly rely on historical statistical data and simple trend analysis, and fail to fully utilize and deeply explore the value of massive tourism information. These methods often ignore important factors such as tourists' emotions, travel preferences, and group characteristics, resulting in poor accuracy and timeliness of forecast results. Traditional tourism market demand forecasting methods mainly rely on a single data source and simple statistical models, and it is difficult to capture the problem of complex market demand changes.

[0005] In a first aspect, the present application provides a method for predicting tourism management market demand based on deep learning, the method comprising: Obtaining tourism-related information corresponding to a preset tourism management platform; the tourism management platform is used to manage multiple preset scenic spots; Perform sentiment analysis on the tourism-related information based on natural language processing technology to obtain positive sentiment information, negative sentiment information and information of interest to tourists; Acquire travel information corresponding to the tourism-related information; the travel information includes a plurality of travel preference information and a travel group type corresponding to each of the travel preference information; Acquire scenic spot visit information of each preset scenic spot corresponding to the tourism related information; Market demand information in a preset time period is generated based on the positive emotion information, negative emotion information, tourist interest information, travel preference information, travel group type and attraction visit information; the market demand information at least includes the tourist quantity information, popular time information and tourist preference information of the preset attractions.

[0006] In some embodiments, the sentiment analysis of the tourism-related information based on natural language processing technology to obtain positive sentiment information, negative sentiment information and information of interest to tourists includes: extracting text information, audio information and image information of the tourism-related information; obtaining text sentiment information corresponding to the text information; obtaining expression information corresponding to the image information, and obtaining image sentiment information based on the expression information; obtaining tone information corresponding to the audio information, and obtaining audio sentiment information based on the tone information; obtaining sentiment themes corresponding to the text sentiment information, image sentiment information and audio sentiment information; and generating the positive sentiment information, negative sentiment information and information of interest to tourists based on the text sentiment information, image sentiment information, audio sentiment information and sentiment themes.

[0007] Exemplarily, before generating the positive emotion information, negative emotion information and information of interest to tourists based on the text emotion information, image emotion information and audio emotion information, it also includes: inputting the text emotion information, image emotion information and audio emotion information into a preset emotion intensity scoring model respectively, and obtaining emotion scoring information corresponding to the text emotion information, image emotion information and audio emotion information respectively; obtaining social network interaction information of the preset tourists from the tourism-related information; constructing a social relationship map of the preset tourists based on the social network interaction information; used to analyze the interaction of the preset tourists at each of the preset attractions; generating the positive emotion information and negative emotion information based on the text emotion information, image emotion information, audio emotion information and the corresponding emotion scoring information; and generating information of interest to tourists based on the social relationship map and the emotion scoring information.

[0008] In some embodiments, the obtaining of travel information corresponding to the tourism-related information; the travel information including multiple travel preference information and travel group types corresponding to each of the travel preference information, includes: performing named entity recognition on the tourism-related information, obtaining the tourist personal information, destination, transportation method and travel time corresponding to the destination of the preset tourist; obtaining the click information and browsing information of the preset tourist on the tourism management platform; inputting the tourist personal information into a preset machine learning model, and outputting the travel group type; generating the travel preference information based on the click information, browsing information and the transportation method and travel time corresponding to the destination.

[0009] In some embodiments, the obtaining of the scenic spot visit information of each preset scenic spot corresponding to the tourism-related information includes: obtaining multiple movement trajectory information corresponding to the tourism-related information; inputting the multiple movement trajectory information into a preset clustering model, and the clustering model outputs the unpopular area information and the popular area information of each preset scenic spot; and generating the scenic spot visit information of each preset scenic spot based on the unpopular area information and the popular area information.

[0010] Exemplarily, before inputting the multiple movement trajectory information into the preset clustering model, it also includes: obtaining image information and image upload timestamp corresponding to the tourism-related information; inputting the image information into a preset convolutional neural network model, and outputting location information corresponding to the image information and tourist expression information in the image information; and inputting the tourist expression information, location information and multiple movement trajectory information into the preset clustering model, and outputting unpopular area information and popular area information of each of the preset attractions.

[0011] In some embodiments, the market demand information in a preset time period is generated based on the positive emotion information, negative emotion information, information of interest to tourists, travel preference information, travel group type and attraction visit information, including: obtaining historical travel information and corresponding historical market demand information corresponding to each of the preset attractions; the historical travel information includes historical positive emotion information, historical negative emotion information, historical information of interest, historical preference information, historical travel group type and historical attraction visit information; constructing a gated recurrent neural network model based on the historical market demand information and historical travel information; inputting the positive emotion information, negative emotion information, information of interest to tourists, travel preference information, travel group type and attraction visit information into the gated recurrent neural network model, and the gated recurrent neural network model outputs the market demand information.

[0012] In some embodiments, before generating the market demand information in a preset time period based on the positive emotion information, negative emotion information, information of interest to tourists, travel preference information, travel group type and scenic spot visit information, it also includes: obtaining weather information and holiday information corresponding to the scenic spot visit information; and generating the market demand information in a preset time period based on the positive emotion information, negative emotion information, information of interest to tourists, travel preference information, travel group type, scenic spot visit information, weather information and holiday information.

[0013] In some embodiments, after generating the market demand information in a preset time period based on the positive emotion information, negative emotion information, tourist interest information, travel preference information, travel group type and attraction visit information, it also includes: generating resource configuration information corresponding to each of the preset attractions based on the market demand information; the resource configuration information includes at least transportation capacity adjustment information, optimized accommodation and catering service information and travel route design information; and completing resource configuration for each of the preset attractions based on the resource configuration information.

[0014] In the second aspect, the present application provides a tourism management market demand forecasting system based on deep learning, the system comprising: An information acquisition unit, used to acquire tourism-related information corresponding to a preset tourism management platform; the tourism management platform is used to manage a plurality of preset scenic spots; A sentiment analysis unit, used to perform sentiment analysis on the tourism-related information based on natural language processing technology to obtain positive sentiment information, negative sentiment information and information of interest to tourists; A travel acquisition unit, configured to acquire travel information corresponding to the tourism-related information; the travel information includes a plurality of travel preference information and a travel group type corresponding to each of the travel preference information; A visit acquisition unit, used to acquire scenic spot visit information of each preset scenic spot corresponding to the tourism related information; A demand generation unit is used to generate market demand information in a preset time period based on the positive emotion information, negative emotion information, tourist interest information, travel preference information, travel group type and scenic spot visit information; the market demand information at least includes the tourist quantity information, popular time information and tourist preference information of the preset scenic spot.

[0015] In a third aspect, the present application further provides a computer device, including: Memory and processor; The memory is used to store computer programs; The processor is used to execute the computer program and implement the steps of the tourism management market demand forecasting method based on deep learning as described in the first aspect above when executing the computer program.

[0016] In a fourth aspect, the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the processor implements the steps of the tourism management market demand forecasting method based on deep learning as described in the first aspect above.

[0017] The present application provides a method and system for forecasting tourism management market demand based on deep learning, which aims to improve the accuracy and timeliness of tourism market demand forecasting through more comprehensive data analysis and processing technology. The core of this method is to integrate and make full use of multi-source information, and through complex data mining and analysis, capture more subtle market dynamics, especially tourists' emotions, travel preferences and group characteristics, etc. These factors are often ignored or difficult to effectively use in traditional tourism market demand forecasting.

[0018] First, relevant tourism information is obtained from the tourism management platform, which is responsible for managing multiple preset attractions. Tourism-related information may include but is not limited to tourists' comments, user feedback on social media, weather forecasts, holiday information, and activity information of scenic spots. Natural language processing (NLP) technology is used to perform sentiment analysis on the above tourism-related information to identify positive sentiment information, negative sentiment information, and specific information of interest to tourists. This part can help understand tourists' true feelings and points of interest about a certain attraction, and provide more delicate behavioral data support for subsequent predictions. The travel information contained in the tourism-related information is analyzed to extract different travel preferences (such as self-driving tours, natural scenery, historical culture, etc.) and their corresponding travel group types (such as families, couples, lonely travelers, etc.). This step helps to segment the market and better understand the differences in needs of different groups. The historical visit records of each preset attraction, such as the number of tourists, stay time, visit frequency, etc., are collected as an important basis for model training. Based on the above positive sentiment information, negative sentiment information, tourist interest information, travel preference information, travel group type, and attraction visit information, a deep learning algorithm is used to predict market demand information within a preset time period. Specific market demand information includes but is not limited to the expected number of tourists at a preset attraction, which time point is most likely to become a peak crowd period, which types of tourists are more inclined to visit the attraction, etc.

[0019] The provided method has the following beneficial effects: 1. Improve forecast accuracy: By integrating multiple types of tourism information (emotions, preferences, historical visit records, etc.), this method can more accurately predict future market demand, and the forecast results are more reliable than traditional methods that only rely on historical statistical data and trend analysis.

[0020] 2. Increase the timeliness of predictions: Combining real-time sentiment analysis and travel preference analysis can allow tourism managers to promptly understand changes in market demand, providing strong support for rapid response and adjustment of strategies.

[0021] 3. Optimize resource allocation: Tourism management parties can make arrangements for crowd control, security measures, and transportation at scenic spots in advance based on the predicted number of tourists and group characteristics, thereby improving the quality of tourist experience and avoiding tourist dissatisfaction or safety hazards caused by improper resource allocation.

[0022] 4. Enhance market insight: This method not only focuses on the number of tourists and the time of visit, but also digs deeply into tourists’ interests, preferences and emotional feedback, which helps tourism managers to understand the market more deeply, discover potential market demands and problem points, and thus promote product and service innovation.

[0023] 5. Promote personalized services: By analyzing the travel preferences of different groups, tourism service providers can provide more personalized services and suggestions, such as recommending safe and educational activities for family tours, and suggesting the best viewing time and routes for tourists who love natural scenery, etc., to improve tourist satisfaction.

[0024] In summary, the tourism market demand forecasting method based on deep learning not only overcomes the limitations of traditional methods, but also brings more positive impacts, helping the tourism industry to better adapt to market changes and improve service quality.

[0025] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0027] Figure 1 This is a schematic flow chart of the steps of a method for predicting market demand for tourism management based on deep learning provided by an embodiment of the present application; Figure 2 It is a structural diagram of a tourism management market demand forecasting system based on deep learning provided in one embodiment of the present application; Figure 3 It is a schematic block diagram of the structure of a computer device provided in one embodiment of the present application.

[0028] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. DETAILED DESCRIPTION

[0029] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0030] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0031] It should be understood that, in order to facilitate the clear description of the technical solutions of the embodiments of the present invention, in the embodiments of the present invention, the words "first", "second", etc. are used to distinguish the same items or similar items with substantially the same functions and effects. Those skilled in the art can understand that the words "first", "second", etc. do not limit the quantity and execution order, and the words "first", "second", etc. do not necessarily limit the difference.

[0032] It should be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0033] It should also be understood that the term “and / or” used in the specification and appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0034] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0035] With the rapid development of the digital economy, business management decisions are facing a complex and changing market environment. Traditional decision support systems mainly rely on historical data and expert experience, and it is difficult to make accurate and timely predictions for rapidly changing market trends and consumer behaviors. In addition, the constant changes in the internal and external environment of the industry have also increased the difficulty of enterprise operation and management.

[0036] Therefore, there is an urgent need for a tourism management market demand forecasting method based on deep learning to solve at least one of the above problems.

[0037] To solve the above problems, please refer to Figure 1 , Figure 1This is a schematic flow chart of a tourism management market demand forecasting method based on deep learning provided by an embodiment of the present application. The tourism management market demand forecasting method based on deep learning can be implemented by a computer device, which can be deployed on a single server or a server cluster. It can also be deployed on a handheld terminal, a laptop computer, a wearable device or a robot, etc.

[0038] It should be noted that the acquisition of any information corresponding to this application complies with the requirements of relevant laws and regulations, and is implemented with the user's consent and without infringing on relevant privacy.

[0039] To solve the above problems, please refer to Figure 1 Specifically, Figure 1 As shown, the provided tourism management market demand forecasting method based on deep learning includes steps S101 to S105. The details are as follows: Step S101. Acquire tourism-related information corresponding to a preset tourism management platform; the tourism management platform is used to manage a plurality of preset scenic spots.

[0040] Specifically, various forms of tourism-related information are obtained from the tourism management platform, including text data (such as tourist comments, social media posts, forum discussions), image data (such as photos uploaded by tourists), and structured data (such as booking records, weather forecasts, holiday information, etc.). These data need to be cleaned and preprocessed to remove invalid or low-quality information to ensure the accuracy and availability of the data. Get tourists' comments and ratings on various attractions from a certain travel booking platform. Capture posts and comments about tourist attractions from social media platforms (such as Weibo, WeChat, Facebook). Get real-time traffic data, weather forecasts, holiday arrangements, etc. of attractions from the tourism management platform. It can comprehensively cover various types of tourism information and provide rich and multi-dimensional data support for subsequent analysis. Through data cleaning and preprocessing, the quality of the data is improved, thereby improving the accuracy of subsequent analysis.

[0041] Step S102: Perform sentiment analysis on the tourism-related information based on natural language processing technology to obtain positive sentiment information, negative sentiment information and information of interest to tourists.

[0042] Specifically, the obtained text data is subjected to sentiment analysis using natural language processing (NLP) technology to identify positive and negative sentiments in the text and information that tourists are interested in. Common sentiment analysis methods include dictionary-based methods, machine learning-based methods, and deep learning-based methods. The text is preprocessed by word segmentation, part-of-speech tagging, named entity recognition, and other preprocessing operations, and then sentiment annotation is performed through sentiment classification models (such as BERT, Transformer).

[0043] For example, analyzing tourist comments: "The service at this hotel is very considerate and the rooms are very clean." Identify positive sentiment information. Analyze social media posts: "This trip is so disappointing. The attractions are poorly maintained." Identify negative sentiment information. Analyze user feedback: "I especially like the food and night market in this city." Identify information that tourists are interested in.

[0044] Deep learning models can more accurately capture the sentiment in text and improve the accuracy of sentiment analysis. Through sentiment analysis, tourism management can timely understand tourists' real feelings and opinions, so as to improve and optimize services. Identifying information that tourists are interested in helps recommend personalized attractions and services.

[0045] Step S103: Acquire travel information corresponding to the tourism-related information; the travel information includes a plurality of travel preference information and a travel group type corresponding to each of the travel preference information.

[0046] Specifically, tourists' travel information is obtained from tourism management platforms and external data sources (such as traffic data and travel platform data). Travel information is analyzed to extract travel preferences (such as travel mode, preferred attraction type, travel time, etc.) and travel group types (such as individual travelers, families, couples, elderly tourists, etc.). Travel information is classified and summarized using data mining techniques (such as clustering algorithms). For example, tourists' travel modes (self-driving, public transportation, walking, etc.) and travel times (weekends, weekdays, holidays, etc.) are obtained from travel booking platforms. Social media analysis shows that the proportion of family tourists at a certain attraction is relatively high. The time distribution of popular travel routes and attractions is obtained from travel platforms. By analyzing travel information, tourism management can better understand tourists' travel habits and preferences and provide more personalized services for tourists. It helps to optimize the allocation of tourism resources, such as adding children's entertainment facilities to attractions with more family tourists. It provides a scientific basis for traffic arrangements and crowd control at attractions.

[0047] Step S104: Acquire the scenic spot visit information of each preset scenic spot corresponding to the tourism related information.

[0048] Specifically, the visit information of each preset scenic spot is obtained from the tourism management platform and the management system of each scenic spot, including the historical number of tourists, stay time, visit frequency, peak time, etc. The data is cleaned and normalized to ensure the consistency and comparability of the data. Time series analysis techniques (such as ARIMA, LSTM) are used to model historical data and predict future tourist visit trends. The number of tourists in the past year was obtained from the management platform of an ancient town, and it was found that the number of tourists on holidays and weekends was large. The stay time data of tourists was obtained from the management system of a beach attraction, and it was found that tourists stayed longer in summer. The daily number of visitors and peak time period were obtained from the entrance gate data of a museum.

[0049] By analyzing historical visit data, tourism managers can understand the passenger flow patterns of each scenic spot and provide a basis for future resource allocation and management. Time series analysis technology can help predict future trends in the number of tourists, reduce resource waste and improve the tourist experience.

[0050] Step S105. Generate market demand information in a preset time period based on the positive emotion information, negative emotion information, tourist interest information, travel preference information, travel group type and scenic spot visit information; the market demand information at least includes the tourist quantity information, popular time information and tourist preference information of the preset scenic spot.

[0051] Specifically, various information obtained in steps S101 to S104 are integrated together, and a deep learning model (such as a neural network, a convolutional neural network, a recursive neural network, etc.) is used for comprehensive analysis and prediction. The model input includes sentiment information, travel preferences, group types, and scenic spot visit data, and the output is market demand information within a preset time period, such as the number of tourists, popular times, tourist preferences, etc. The model is trained and tuned to ensure the accuracy and reliability of the prediction results.

[0052] The data input into the model include: positive sentiment information (number of positive comments), negative sentiment information (number of negative comments), travel preference information (proportion of self-driving tours), travel group type (proportion of family tourists), visit data (historical number of tourists, length of stay), etc. of a certain scenic spot in the past week. The model predicts the number of tourists, popular times (such as weekends and holidays), and tourist preferences (such as natural scenery, history and culture, etc.) of the scenic spot in the next month.

[0053] Deep learning models can process multi-dimensional and complex data, improving the accuracy and timeliness of forecasts. Through comprehensive analysis, tourism managers can fully understand market demand, adjust strategies and services in a timely manner, and improve tourist satisfaction. It helps to optimize the allocation of tourism resources, such as adding staff and adjusting ticketing policies during predicted popular time periods.

[0054] Through the above steps, the tourism management market demand forecasting method based on deep learning can effectively utilize multi-source information, improve the accuracy and timeliness of forecasting, help tourism management parties better cope with the complex and changing market environment, optimize resource allocation, enhance tourist experience, and promote the healthy development of the tourism industry.

[0055] In some embodiments, the sentiment analysis of the tourism-related information based on natural language processing technology to obtain positive sentiment information, negative sentiment information and information of interest to tourists includes: extracting text information, audio information and image information of the tourism-related information; obtaining text sentiment information corresponding to the text information; obtaining expression information corresponding to the image information, and obtaining image sentiment information based on the expression information; obtaining tone information corresponding to the audio information, and obtaining audio sentiment information based on the tone information; obtaining sentiment themes corresponding to the text sentiment information, image sentiment information and audio sentiment information; and generating the positive sentiment information, negative sentiment information and information of interest to tourists based on the text sentiment information, image sentiment information, audio sentiment information and sentiment themes.

[0056] Text information, such as extracting comments and posts from tourism-related text data. Audio information, such as extracting voice recordings and voice comments from tourism-related audio data. Image information, such as extracting photos and video screenshots uploaded by tourists from tourism-related image data. Use NLP technology to perform sentiment analysis on text data and identify positive sentiment, negative sentiment and information of interest. Common sentiment analysis models include BERT, Transformer, etc. Analyze the expressions in the image and extract the sentiment information in the image. Common image sentiment analysis methods include convolutional neural network (CNN) models. Analyze the tone in the audio and extract the sentiment information in the audio. Common audio sentiment analysis methods include Mel-frequency cepstral coefficients (MFCC) and recurrent neural network (RNN) models. Combine text sentiment information, image sentiment information and audio sentiment information to identify specific sentiment themes, such as "service attitude", "attraction maintenance", "food experience", etc. Generate comprehensive positive sentiment information, negative sentiment information and information of interest to tourists based on sentiment themes and other sentiment information.

[0057] Exemplarily, before generating the positive emotion information, negative emotion information and information of interest to tourists based on the text emotion information, image emotion information and audio emotion information, it also includes: inputting the text emotion information, image emotion information and audio emotion information into a preset emotion intensity scoring model respectively, and obtaining emotion scoring information corresponding to the text emotion information, image emotion information and audio emotion information respectively; obtaining social network interaction information of the preset tourists from the tourism-related information; constructing a social relationship map of the preset tourists based on the social network interaction information; used to analyze the interaction of the preset tourists at each of the preset attractions; generating the positive emotion information and negative emotion information based on the text emotion information, image emotion information, audio emotion information and the corresponding emotion scoring information; and generating information of interest to tourists based on the social relationship map and the emotion scoring information.

[0058] By inputting text emotion information, image emotion information and audio emotion information into the emotion intensity scoring model respectively, the corresponding emotion scoring information is obtained. The emotion intensity scoring model can be a regression model based on deep learning, such as LSTM or CNN. The social network interaction information of the preset tourists is obtained from the tourism-related information, and the social relationship map of the preset tourists is constructed. The social network interaction information may include likes, comments, reposts, etc. The interaction of the preset tourists at each preset scenic spot is analyzed to understand the connection and common interests between tourists. According to the text emotion information, image emotion information, audio emotion information and their corresponding emotion scoring information, comprehensive positive emotion information and negative emotion information are generated. Combined with the social relationship map and emotion scoring information, information of interest to tourists is generated.

[0059] Through multimodal (text, image, audio) sentiment analysis, we can more comprehensively capture tourists' real feelings and opinions and improve the accuracy of sentiment analysis. The sentiment intensity scoring model can quantify the intensity of sentiment information, allowing tourism managers to more intuitively understand tourists' sentiment tendencies. Building a social relationship map helps to discover interactions and common interests between tourists and provide support for personalized services. By comprehensively analyzing various types of sentiment information and social relationships, tourism managers can better understand tourists' needs and provide personalized recommendations and services.

[0060] In some embodiments, the obtaining of travel information corresponding to the tourism-related information; the travel information including multiple travel preference information and travel group types corresponding to each of the travel preference information, includes: performing named entity recognition on the tourism-related information, obtaining the tourist personal information, destination, transportation method and travel time corresponding to the destination of the preset tourist; obtaining the click information and browsing information of the preset tourist on the tourism management platform; inputting the tourist personal information into a preset machine learning model, and outputting the travel group type; generating the travel preference information based on the click information, browsing information and the transportation method and travel time corresponding to the destination.

[0061] By performing named entity recognition on tourism-related information, the personal information of tourists (such as age, gender), destination, mode of transportation (such as self-driving, public transportation) and travel time (such as weekends, holidays) of preset tourists are extracted.

[0062] Obtain the click and browse information of preset tourists on the tourism management platform to understand the tourists' interest in different attractions.

[0063] Input tourists’ personal information into preset machine learning models (such as decision trees, random forests, neural networks) for classification, and output travel group types, such as families, couples, individual travelers, etc.

[0064] Generate specific travel preference information based on click information, browsing information, transportation mode corresponding to the destination, and travel time.

[0065] Through named entity recognition and machine learning models, user portraits can be constructed more accurately to understand tourists' personal information and travel preferences. Tourism management can provide personalized travel suggestions and services for different groups of tourists based on the generated travel preference information. Understanding the travel preferences and time distribution of different groups will help optimize the flow of people at scenic spots and the allocation of service resources.

[0066] In some embodiments, the obtaining of the scenic spot visit information of each preset scenic spot corresponding to the tourism-related information includes: obtaining multiple movement trajectory information corresponding to the tourism-related information; inputting the multiple movement trajectory information into a preset clustering model, and the clustering model outputs the unpopular area information and the popular area information of each preset scenic spot; and generating the scenic spot visit information of each preset scenic spot based on the unpopular area information and the popular area information.

[0067] By obtaining tourists' movement trajectory information from tourism-related data, such as location records of travel applications, WiFi connection records in scenic spots, etc. Input multiple movement trajectory information into a preset clustering model (such as K-means, DBSCAN), and the model outputs the unpopular area information and popular area information of each preset scenic spot. Obtain image information and its upload timestamp from tourism-related information. Location information and tourist expression information: Input the image information into a preset convolutional neural network (CNN) model, and output the location information and tourist expression information corresponding to the image information. Input the tourist expression information, location information and multiple movement trajectory information into the preset clustering model to generate the unpopular area information and popular area information of each preset scenic spot.

[0068] Exemplarily, before inputting the multiple movement trajectory information into the preset clustering model, it also includes: obtaining image information and image upload timestamp corresponding to the tourism-related information; inputting the image information into a preset convolutional neural network model, and outputting location information corresponding to the image information and tourist expression information in the image information; and inputting the tourist expression information, location information and multiple movement trajectory information into the preset clustering model, and outputting unpopular area information and popular area information of each of the preset attractions.

[0069] By inputting image information (photos uploaded by tourists) into the convolutional neural network (CNN) model, the location information and tourist expression information in the image are output. The location information can be identified by geographic tags or background features in the image, and the expression information can be extracted through facial recognition technology. The tourist expression information, location information and multiple movement trajectory information are input into the preset clustering model, and the unpopular area information and popular area information of each preset attraction are output.

[0070] By analyzing the movement trajectory information and image location information through clustering models, the unpopular and popular areas of each scenic spot can be more finely identified, providing a scientific basis for scenic spot management. Combined with the information of tourists' expressions, it is possible to understand the emotional changes of tourists in different areas, and timely discover and solve the problems of tourists' dissatisfaction. By analyzing the unpopular and popular areas, tourism management can adjust the service and facility layout of the scenic spot to improve the overall experience of tourists. Comprehensive analysis using multi-source data (movement trajectory, images, expressions) helps to achieve data-driven tourism management decisions.

[0071] In some embodiments, the market demand information in a preset time period is generated based on the positive emotion information, negative emotion information, information of interest to tourists, travel preference information, travel group type and attraction visit information, including: obtaining historical travel information and corresponding historical market demand information corresponding to each of the preset attractions; the historical travel information includes historical positive emotion information, historical negative emotion information, historical information of interest, historical preference information, historical travel group type and historical attraction visit information; constructing a gated recurrent neural network model based on the historical market demand information and historical travel information; inputting the positive emotion information, negative emotion information, information of interest to tourists, travel preference information, travel group type and attraction visit information into the gated recurrent neural network model, and the gated recurrent neural network model outputs the market demand information.

[0072] Among them, the above information is respectively: Historical positive sentiment information: positive comments of tourists on scenic spots in the past period of time. Historical negative sentiment information: negative comments of tourists on scenic spots in the past period of time. Historical interest information: special points of interest of tourists in scenic spots or activities in the past period of time. Historical preference information: travel preferences of tourists in the past period of time, such as transportation and travel time. Historical travel group type: type of tourist group that visited scenic spots in the past period of time, such as families, couples, individual travelers, etc. Historical scenic spot visit information: visit data of scenic spots in the past period of time, such as the number of tourists, stay time, peak hours, etc.

[0073] Obtain the actual market demand information of each preset attraction in the past period of time, such as the actual number of tourists, popular times, and tourist preferences. Use historical market demand information and historical travel information to build a gated recurrent neural network (GRU, LSTM) model. The gated recurrent neural network model can capture long-term dependencies in time series data and is suitable for handling dynamic changes in tourism market demand.

[0074] The currently acquired positive emotion information, negative emotion information, tourist interest information, travel preference information, travel group type and scenic spot visit information are input into the gated recurrent neural network model. The model outputs the market demand information within the preset time period, including the number of tourists, popular time and tourist preferences.

[0075] The gated recurrent neural network model can capture long-term dependencies in time series data and improve the ability to predict dynamic changes in tourism market demand. The model is constructed by historical travel information and historical market demand information, making full use of past experience data to improve the accuracy and reliability of predictions. Multi-dimensional information such as emotional information, preference information, group type and access data are comprehensively considered to make the prediction results more comprehensive and detailed. Based on the prediction results, tourism management can allocate resources and adjust strategies more scientifically to improve tourist experience and service quality.

[0076] In some embodiments, before generating the market demand information in a preset time period based on the positive emotion information, negative emotion information, information of interest to tourists, travel preference information, travel group type and scenic spot visit information, it also includes: obtaining weather information and holiday information corresponding to the scenic spot visit information; and generating the market demand information in a preset time period based on the positive emotion information, negative emotion information, information of interest to tourists, travel preference information, travel group type, scenic spot visit information, weather information and holiday information.

[0077] Obtain weather information for each preset scenic spot within a preset time period from the weather forecast platform, such as sunny days, rainy days, temperature, etc. Obtain holiday information within a preset time period from the tourism management platform or official holiday information source, such as Spring Festival, National Day, weekends, etc. Input positive emotion information, negative emotion information, tourist interest information, travel preference information, travel group type, scenic spot visit information, weather information, and holiday information into the prediction model. The model outputs market demand information within a preset time period, including the number of tourists, popular times, and tourist preferences.

[0078] Weather information and holiday information have a significant impact on tourism demand. Adding these factors can greatly improve the accuracy and reliability of forecasts. Through multi-dimensional information input, the forecast model can more comprehensively reflect market demand and reduce the impact of a single factor on the forecast results. Tourism managers can adjust resources and services in a timely manner according to changes in weather and holidays, such as adding indoor activities on rainy days and adding staff on holidays. Through more accurate market demand forecasts, tourism managers can better meet the needs of tourists and enhance the overall experience of tourists.

[0079] In some embodiments, after generating the market demand information in a preset time period based on the positive emotion information, negative emotion information, tourist interest information, travel preference information, travel group type and attraction visit information, it also includes: generating resource configuration information corresponding to each of the preset attractions based on the market demand information; the resource configuration information includes at least transportation capacity adjustment information, optimized accommodation and catering service information and travel route design information; and completing resource configuration for each of the preset attractions based on the resource configuration information.

[0080] The market demand information generated by step S105 includes the number of tourists, popular times, and tourist preferences within a preset time period. For example, according to the predicted number of tourists and popular times, the capacity of transportation vehicles can be adjusted, such as increasing bus schedules, setting up temporary parking lots, etc. According to the number of tourists and preferences, accommodation and catering services can be optimized, such as increasing room supply during popular time periods, launching seasonal special dishes, etc. According to the preferences of tourists and information on scenic spot visits, reasonable travel routes can be designed, such as recommending the best tour sequence, setting up guide services, etc. According to the generated resource allocation information, resource allocation for each preset scenic spot can be completed, such as adjusting staff, optimizing service facilities, etc.

[0081] If it is predicted that the number of tourists to a certain scenic spot will increase significantly during holidays, increase the number of bus trips in advance and set up temporary parking lots. If it is predicted that a certain scenic spot will have a large number of tourists during the summer vacation and that tourists prefer history and culture, launch rooms with historical and cultural themes and special restaurants. If it is predicted that a certain scenic spot will have a large number of tourists on weekends and that tourists prefer outdoor activities, design a reasonable tourist route that includes outdoor activities and provide guided tour services.

[0082] The resource allocation information generated by demand forecasting can scientifically adjust resources such as transportation, accommodation and catering to improve resource utilization efficiency. Targeted service optimization and adjustment, such as increasing the frequency of transportation during popular time periods and launching special accommodation and catering, can enhance the service experience of tourists. Through reasonable adjustment of transportation capacity and design of tourist routes, the congestion of scenic spots can be effectively reduced and the satisfaction of tourists can be improved. Tourism management can adjust service strategies in a timely manner according to market demand information, better respond to market changes, and enhance market responsiveness.

[0083] Exemplarily, the expression corresponding to the traffic capacity adjustment information includes: ; in, for the transportation capacity adjustment information, The baseline capacity is an initial value set based on past average capacity demand. To predict the number of tourists, it is the number of tourists predicted in the preset time period. is the popular time coefficient, which is the influence coefficient of popular time periods on the number of tourists. is the weather impact coefficient, which is the impact coefficient of different weather types on the number of tourists. The holiday coefficient is the impact coefficient of holidays on the number of tourists, which is usually greater than 1. and The first adjustment coefficient and the second adjustment coefficient are preset respectively, and are used to balance the weights of various factors.

[0084] Exemplary optimized accommodation and catering service information includes accommodation optimization information and catering optimization information.

[0085] Expressions for property optimization information include: ; in, to optimize information for said accommodation, This is the base number of rooms, an initial value set based on average past demand. To predict the number of tourists, it is the number of tourists predicted in the preset time period. is the tourist preference coefficient, which is the influence coefficient of different tourist preferences on accommodation demand. is the weather impact coefficient, which is the impact coefficient of different weather types on the number of tourists. The holiday coefficient is the impact coefficient of holidays on the number of tourists, which is usually greater than 1. and They are respectively the preset third adjustment coefficient and the fourth adjustment coefficient, which are used to balance the weights of various factors.

[0086] The expressions of catering optimization information include: ; in, optimizing information for said catering, This is the base number of rooms, an initial value set based on average past demand. To predict the number of tourists, it is the number of tourists predicted in the preset time period. is the tourist preference coefficient, which is the influence coefficient of different tourist preferences on accommodation demand. is the weather impact coefficient, which is the impact coefficient of different weather types on the number of tourists. The holiday coefficient is the impact coefficient of holidays on the number of tourists, which is usually greater than 1. and They are respectively the preset fifth adjustment coefficient and the sixth adjustment coefficient, which are used to balance the weights of various factors.

[0087] Exemplarily, the expression of the travel route design information includes: ; in, designing information for said tourist route, For the baseline route, the initial value is set based on the average demand in the past. To predict the number of tourists, it is the number of tourists predicted in the preset time period. is the popular time coefficient, which is the influence coefficient of popular time periods on the number of tourists. is the tourist preference coefficient, which is the influence coefficient of different tourist preferences on accommodation demand. is the weather impact coefficient, which is the impact coefficient of different weather types on the number of tourists. , and They are the preset seventh adjustment coefficient, eighth adjustment coefficient and ninth adjustment coefficient, which are used to balance the weights of various factors. is the frequency of scenic spot visits, which is the historical frequency of visits to each scenic spot.

[0088] The formula can flexibly respond to the dynamic changes of different factors by adjusting coefficients and parameters, thus enhancing the adaptability of the model. The formula design provides a more scientific basis for decision-making, helping tourism management parties to more effectively allocate resources and adjust strategies. Through more sophisticated and dynamic resource allocation, it can better meet the needs of tourists and improve their overall experience and service satisfaction.

[0089] See also Figure 2 As shown, Figure 2 : is a schematic diagram of the structure of the tourism management market demand forecasting system 200 based on deep learning provided in an embodiment of the present application. The tourism management market demand forecasting system 200 based on deep learning is used to execute the steps of the tourism management market demand forecasting method based on deep learning shown in the above embodiments. The tourism management market demand forecasting system 200 based on deep learning can be a single server or a server cluster, or the tourism management market demand forecasting system 200 based on deep learning can be a terminal, which can be a handheld terminal, a laptop computer, a wearable device or a robot, etc.

[0090] like Figure 2 As shown, the tourism management market demand forecasting system 200 based on deep learning includes: An information acquisition unit 201 is used to acquire tourism-related information corresponding to a preset tourism management platform; the tourism management platform is used to manage a plurality of preset scenic spots; The sentiment analysis unit 202 is used to perform sentiment analysis on the tourism-related information based on natural language processing technology to obtain positive sentiment information, negative sentiment information and information of interest to tourists; A travel acquisition unit 203 is used to acquire travel information corresponding to the tourism related information; the travel information includes a plurality of travel preference information and a travel group type corresponding to each of the travel preference information; The access acquisition unit 204 is used to acquire the scenic spot access information of each preset scenic spot corresponding to the tourism related information; The demand generation unit 205 is used to generate market demand information in a preset time period based on the positive emotion information, negative emotion information, tourist interest information, travel preference information, travel group type and attraction visit information; the market demand information at least includes the tourist quantity information, popular time information and tourist preference information of the preset attraction.

[0091] It should be noted that technical personnel in the relevant field can clearly understand that, for the convenience and conciseness of description, the specific working processes of the tourism management market demand forecasting system based on deep learning and each module described above can refer to the corresponding processes in the embodiments of the tourism management market demand forecasting method based on deep learning described in the above embodiments, and will not be repeated here.

[0092] The above-mentioned tourism management market demand forecasting method based on deep learning can be implemented in the form of a computer program. The computer program can be used in Figure 2 Run on the system shown.

[0093] See also Figure 3 , Figure 3 1 is a schematic block diagram of the structure of a computer device provided in an embodiment of the present application. The computer device includes a processor, a memory and a network interface connected via a device bus, wherein the memory may include a storage medium and an internal memory.

[0094] The storage medium can store an operating device and a computer program. The computer program includes program instructions, and when the program instructions are executed, the processor can execute any one of the tourism management market demand forecasting methods based on deep learning.

[0095] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

[0096] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any tourism management market demand forecasting method based on deep learning.

[0097] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will understand that Figure 3The structure shown in the figure is only a block diagram of a part of the structure related to the scheme of the present application, and does not constitute a limitation on the terminal to which the scheme of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0098] It should be understood that the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0099] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps: Obtaining tourism-related information corresponding to a preset tourism management platform; the tourism management platform is used to manage multiple preset scenic spots; Perform sentiment analysis on the tourism-related information based on natural language processing technology to obtain positive sentiment information, negative sentiment information and information of interest to tourists; Acquire travel information corresponding to the tourism-related information; the travel information includes a plurality of travel preference information and a travel group type corresponding to each of the travel preference information; Acquire scenic spot visit information of each preset scenic spot corresponding to the tourism related information; Market demand information in a preset time period is generated based on the positive emotion information, negative emotion information, tourist interest information, travel preference information, travel group type and attraction visit information; the market demand information at least includes the tourist quantity information, popular time information and tourist preference information of the preset attractions.

[0100] In some embodiments, the sentiment analysis of the tourism-related information based on natural language processing technology to obtain positive sentiment information, negative sentiment information and information of interest to tourists includes: extracting text information, audio information and image information of the tourism-related information; obtaining text sentiment information corresponding to the text information; obtaining expression information corresponding to the image information, and obtaining image sentiment information based on the expression information; obtaining tone information corresponding to the audio information, and obtaining audio sentiment information based on the tone information; obtaining sentiment themes corresponding to the text sentiment information, image sentiment information and audio sentiment information; and generating the positive sentiment information, negative sentiment information and information of interest to tourists based on the text sentiment information, image sentiment information, audio sentiment information and sentiment themes.

[0101] Exemplarily, before generating the positive emotion information, negative emotion information and information of interest to tourists based on the text emotion information, image emotion information and audio emotion information, it also includes: inputting the text emotion information, image emotion information and audio emotion information into a preset emotion intensity scoring model respectively, and obtaining emotion scoring information corresponding to the text emotion information, image emotion information and audio emotion information respectively; obtaining social network interaction information of the preset tourists from the tourism-related information; constructing a social relationship map of the preset tourists based on the social network interaction information; used to analyze the interaction of the preset tourists at each of the preset attractions; generating the positive emotion information and negative emotion information based on the text emotion information, image emotion information, audio emotion information and the corresponding emotion scoring information; and generating information of interest to tourists based on the social relationship map and the emotion scoring information.

[0102] In some embodiments, the obtaining of travel information corresponding to the tourism-related information; the travel information including multiple travel preference information and travel group types corresponding to each of the travel preference information, includes: performing named entity recognition on the tourism-related information, obtaining the tourist personal information, destination, transportation method and travel time corresponding to the destination of the preset tourist; obtaining the click information and browsing information of the preset tourist on the tourism management platform; inputting the tourist personal information into a preset machine learning model, and outputting the travel group type; generating the travel preference information based on the click information, browsing information and the transportation method and travel time corresponding to the destination.

[0103] In some embodiments, the obtaining of the scenic spot visit information of each preset scenic spot corresponding to the tourism-related information includes: obtaining multiple movement trajectory information corresponding to the tourism-related information; inputting the multiple movement trajectory information into a preset clustering model, and the clustering model outputs the unpopular area information and the popular area information of each preset scenic spot; and generating the scenic spot visit information of each preset scenic spot based on the unpopular area information and the popular area information.

[0104] Exemplarily, before inputting the multiple movement trajectory information into the preset clustering model, it also includes: obtaining image information and image upload timestamp corresponding to the tourism-related information; inputting the image information into a preset convolutional neural network model, and outputting location information corresponding to the image information and tourist expression information in the image information; and inputting the tourist expression information, location information and multiple movement trajectory information into the preset clustering model, and outputting unpopular area information and popular area information of each of the preset attractions.

[0105] In some embodiments, the market demand information in a preset time period is generated based on the positive emotion information, negative emotion information, information of interest to tourists, travel preference information, travel group type and attraction visit information, including: obtaining historical travel information and corresponding historical market demand information corresponding to each of the preset attractions; the historical travel information includes historical positive emotion information, historical negative emotion information, historical information of interest, historical preference information, historical travel group type and historical attraction visit information; constructing a gated recurrent neural network model based on the historical market demand information and historical travel information; inputting the positive emotion information, negative emotion information, information of interest to tourists, travel preference information, travel group type and attraction visit information into the gated recurrent neural network model, and the gated recurrent neural network model outputs the market demand information.

[0106] In some embodiments, before generating the market demand information in a preset time period based on the positive emotion information, negative emotion information, information of interest to tourists, travel preference information, travel group type and scenic spot visit information, it also includes: obtaining weather information and holiday information corresponding to the scenic spot visit information; and generating the market demand information in a preset time period based on the positive emotion information, negative emotion information, information of interest to tourists, travel preference information, travel group type, scenic spot visit information, weather information and holiday information.

[0107] In some embodiments, after generating the market demand information in a preset time period based on the positive emotion information, negative emotion information, tourist interest information, travel preference information, travel group type and attraction visit information, it also includes: generating resource configuration information corresponding to each of the preset attractions based on the market demand information; the resource configuration information includes at least transportation capacity adjustment information, optimized accommodation and catering service information and travel route design information; and completing resource configuration for each of the preset attractions based on the resource configuration information.

[0108] The present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the processor implements the steps of the tourism management market demand forecasting method based on deep learning as described in the first aspect above.

[0109] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc., equipped on the computer device.

[0110] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A tourism management market demand forecasting method based on deep learning, characterized in that: The method comprises: Obtaining tourism-related information corresponding to a preset tourism management platform; the tourism management platform is used to manage multiple preset scenic spots; Based on natural language processing technology, sentiment analysis is performed on the tourism-related information to obtain positive sentiment information, negative sentiment information and information of interest to tourists; Acquire travel information corresponding to the tourism-related information; the travel information includes a plurality of travel preference information and a travel group type corresponding to each of the travel preference information; Acquire scenic spot visit information of each preset scenic spot corresponding to the tourism related information; Market demand information in a preset time period is generated based on the positive emotion information, negative emotion information, tourist interest information, travel preference information, travel group type and attraction visit information; the market demand information at least includes the tourist quantity information, popular time information and tourist preference information of the preset attractions.

2. The method according to claim 1, characterized in that The sentiment analysis of the tourism-related information based on natural language processing technology to obtain positive sentiment information, negative sentiment information and information of interest to tourists includes: Extracting text information, audio information and image information of the tourism related information; Obtaining text sentiment information corresponding to the text information; Acquire expression information corresponding to the image information, and acquire image emotion information according to the expression information; Acquire tone information corresponding to the audio information, and acquire audio emotion information according to the tone information; Obtaining emotional themes corresponding to the text emotional information, image emotional information, and audio emotional information; The positive emotion information, negative emotion information and tourist interest information are generated according to the text emotion information, image emotion information, audio emotion information and emotion theme.

3. The method according to claim 2, characterized in that Before generating the positive emotion information, negative emotion information and tourist interest information according to the text emotion information, the image emotion information and the audio emotion information, the method further includes: Inputting the text emotion information, image emotion information and audio emotion information into a preset emotion intensity scoring model respectively, and obtaining emotion scoring information corresponding to the text emotion information, image emotion information and audio emotion information respectively; Obtaining social network interaction information of preset tourists from the tourism related information; A social relationship map of the preset tourists is constructed based on the social network interaction information; used to analyze the interaction of the preset tourists at each of the preset attractions; the positive emotion information and negative emotion information are generated based on the text emotion information, image emotion information, audio emotion information and corresponding emotion score information; and tourist interest information is generated based on the social relationship map and the emotion score information.

4. The method according to claim 1, characterized in that The obtaining of travel information corresponding to the tourism-related information; the travel information including a plurality of travel preference information and a travel group type corresponding to each of the travel preference information, including: Perform named entity recognition on the tourism-related information to obtain the personal information of the preset tourists, the destination, the transportation method and travel time corresponding to the destination; Obtaining the click information and browsing information of preset tourists on the tourism management platform; Inputting the tourist personal information into a preset machine learning model and outputting the travel group type; The travel preference information is generated according to the click information, the browsing information, and the transportation mode and travel time corresponding to the destination.

5. The method according to claim 1, characterized in that The obtaining of the scenic spot visit information of each preset scenic spot corresponding to the tourism related information includes: Acquire multiple movement trajectory information corresponding to the tourism related information; Inputting the plurality of movement trajectory information into a preset clustering model, the clustering model outputting unpopular area information and popular area information of each preset scenic spot; The scenic spot visiting information of each preset scenic spot is generated according to the unpopular area information and the popular area information.

6. The method according to claim 5, characterized in that Before inputting the plurality of movement trajectory information into a preset clustering model, the method further includes: Obtain image information and image upload timestamp corresponding to the tourism related information; The image information is input into a preset convolutional neural network model, and the location information corresponding to the image information and the tourist expression information in the image information are output; the tourist expression information, location information and multiple movement trajectory information are input into a preset clustering model, and the unpopular area information and popular area information of each preset scenic spot are output.

7. The method according to claim 1, characterized in that The generating of market demand information in a preset time period according to the positive emotion information, negative emotion information, tourist interest information, travel preference information, travel group type and scenic spot visit information includes: Obtaining historical travel information and corresponding historical market demand information corresponding to each of the preset attractions; the historical travel information includes historical positive emotion information, historical negative emotion information, historical interest information, historical preference information, historical travel group type and historical attraction visit information; Constructing a gated recurrent neural network model based on the historical market demand information and the historical travel information; The positive emotion information, negative emotion information, tourist interest information, travel preference information, travel group type and scenic spot visit information are input into the gated recurrent neural network model, and the gated recurrent neural network model outputs the market demand information.

8. The method according to claim 1, characterized in that Before generating the market demand information in a preset time period according to the positive emotion information, the negative emotion information, the tourist interest information, the travel preference information, the travel group type and the scenic spot visit information, the method further includes: Obtain weather information and holiday information corresponding to the scenic spot visit information; and generate market demand information in a preset time period based on the positive emotion information, negative emotion information, tourist interest information, travel preference information, travel group type, scenic spot visit information, weather information and holiday information.

9. The method according to claim 1, characterized in that: After generating the market demand information in a preset time period according to the positive emotion information, the negative emotion information, the tourist interest information, the travel preference information, the travel group type and the scenic spot visit information, the method further includes: Generate resource allocation information corresponding to each of the preset scenic spots according to the market demand information; the resource allocation information at least includes transportation capacity adjustment information, optimized accommodation and catering service information and travel route design information; The resource configuration for each of the preset scenic spots is completed according to the resource configuration information.

10. A tourism management market demand forecasting system based on deep learning, characterized in that: The system comprises: An information acquisition unit, used to acquire tourism-related information corresponding to a preset tourism management platform; The tourism management platform is used to manage multiple preset scenic spots; A sentiment analysis unit, used to perform sentiment analysis on the tourism-related information based on natural language processing technology to obtain positive sentiment information, negative sentiment information and information of interest to tourists; A travel acquisition unit, configured to acquire travel information corresponding to the tourism related information; the travel information includes a plurality of travel preference information and a travel group type corresponding to each of the travel preference information; A visit acquisition unit, used to acquire scenic spot visit information of each preset scenic spot corresponding to the tourism related information; A demand generation unit is used to generate market demand information in a preset time period based on the positive emotion information, negative emotion information, tourist interest information, travel preference information, travel group type and scenic spot visit information; the market demand information at least includes the tourist quantity information, popular time information and tourist preference information of the preset scenic spot.

Images