File travel resource investigation system and method based on multi-source data fusion and artificial intelligence

Abstract

The invention discloses a document travel resource investigation system and method based on multi-source data fusion and artificial intelligence, and relates to the technical field of document travel resource investigation, and the document travel resource method based on multi-source data fusion and artificial intelligence specifically comprises the following steps: step 1, establishing a distributed document travel resource data acquisition system, the method comprises the following steps of 1, collecting text and travel resource data of each region through multiple channels, 2, analyzing data features of the multi-source text and travel resource data, and comprehensively defining feature values of the text and travel resource data, and 3, setting resource evaluation indexes according to the feature values of the text and travel resource data, and completing investigation and evaluation of text and travel resources of the corresponding region according to the resource evaluation indexes. According to the method, the problems of data islands and evaluation subjectivity in a traditional method are solved, multi-source data fusion, dynamic quantitative evaluation and intelligent decision support are realized, and the development success rate of text travel resources can be improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of cultural and tourism resource investigation, and specifically relates to a cultural and tourism resource investigation system and method based on multi-source data fusion and artificial intelligence. Background Art

[0002] In recent years, with the rapid development of the tourism industry, the importance of cultural and tourism resource investigation has become increasingly prominent. Cultural and tourism resources are an important foundation for carrying out cultural and tourism activities and developing the cultural and tourism industry. Through cultural and tourism resource investigation, the status of cultural and tourism resources within a region can be comprehensively understood and mastered, providing an accurate basis for scientific planning and reasonable development. However, traditional cultural and tourism resource investigation mainly relies on manual on-site investigation. This method not only takes a lot of time and effort, but also the data dimensions that can be collected are relatively limited, usually only limited to basic information such as geographical location, area, and architectural style, and it is difficult to cover multi-dimensional information such as tourist behavior, emotional feedback, and environmental changes. Due to relying on manual on-site investigation, the data update cycle is long, and it is difficult to reflect the latest status of cultural and tourism resources and market changes in a timely manner, resulting in poor timeliness of the investigation results. Although existing technologies have proposed methods for analyzing cultural and tourism big data, when dealing with multi-modal data such as text, images, and videos, they face the problem of semantic alignment.

[0003] Therefore, there is an urgent need for a cultural and tourism resource investigation method based on multi-source data fusion and artificial intelligence to break through the above-mentioned technical bottlenecks and limitations of existing technologies. Summary of the Invention

[0004] The purpose of the present invention is to provide a cultural and tourism resource investigation system and method based on multi-source data fusion and artificial intelligence to solve the technical problems in the prior art that it is difficult to cover multi-dimensional information and it is difficult to process multi-modal data.

[0005] To achieve the above purpose, the present invention adopts the following technical solutions: A cultural and tourism resource investigation method based on multi-source data fusion and artificial intelligence, including: Step 1, establish a distributed cultural and tourism resource data collection system, and collect cultural and tourism resource data of each region through various channels; Step 2, analyze the data characteristics of multi-source cultural and tourism resource data, and comprehensively define the characteristic values of cultural and tourism resource data; Step 3, set resource evaluation indicators based on the characteristic values of cultural and tourism resource data, and complete the investigation and evaluation of cultural and tourism resources in the corresponding region according to the resource evaluation indicators.

[0006] Further, the method for establishing a distributed cultural and tourism resource data collection system is as follows: Determine the area range of cultural and tourism resources to be investigated, denoted as the investigation area, and determine the scenic spots owned by the investigation area. In this embodiment, a place or area that provides tourist services and has clear management boundaries is defined as a scenic spot. Build a distributed cultural and tourism resource data collection system based on the heterogeneous data integration framework of Apache NiFi. Divide time periods by T duration, and use the distributed data crawling system to collect scenic spot IoT data, social media data, and scenic spot business data in real time for each time period. The scenic spot IoT data, social media data, and scenic spot business data are collectively referred to as cultural and tourism resource data.

[0007] Furthermore, analyze the data characteristics of multi-source cultural and tourism resource data, and comprehensively define the characteristic values of cultural and tourism resource data. The specific method is as follows: Conduct time series analysis on the tourist volume data, extract the change trend and peak time of tourist flow, and use statistical analysis methods to quantify the fluctuation of tourist flow by calculating the average value and standard deviation of the daily tourist volume within a time period, so as to obtain the tourist flow characteristics. Normalize the environmental data to eliminate the influence of different dimensions on data analysis, extract the environmental quality characteristics, and comprehensively consider the tourist flow characteristics and environmental quality characteristics to determine the IoT data characteristic values of the scenic spot; Preprocess the social media data, design a cross-modal mapping model based on the CLIP architecture, align the latent space characteristics of text, images, and sensor data, decompose the video data into continuous image data at a preset frame rate, extract the audio content through video editing software, use the Whisper model to transcribe the audio content part into text content, use OCR technology to extract the text content in the image data, preset a cultural and tourism vocabulary database composed of tendency words, extract the tendency words of the social media review text in the cultural and tourism resource data through the DeBERTa-v3 model, and determine the emotional tendency characteristic values of the scenic spot by analyzing the tendency words appearing in the text content of the social media data; Determine the accuracy of the scenic spot business data, and comprehensively determine the attraction characteristic values based on the total tourist flow data of the scenic spot within a time period and the geographical distribution variance of different tourist source areas.

[0008] Furthermore, comprehensively consider the tourist flow characteristics and environmental quality characteristics to determine the IoT data characteristic values of the scenic spot. The specific method is as follows: Use the formula to represent the IoT data characteristic values of the scenic spot, where t represents the t-th time period, x represents the x-th scenic spot, represents the IoT data characteristic value of scenic spot x in the t-th time period, a represents the a-th day, and h represents a total of h days, represents the tourist flow of scenic spot x on the a-th day in the t-th time period, represents the average daily tourist flow of scenic spot x in the t-th time period, Represents the maximum number of tourists in scenic area x within one day during the t time period. Represents the minimum number of tourists in scenic area x within one day during the t time period. Represents the total number of tourists in scenic area x during the t time period. n represents n types of environmental data, and i represents the i-th type of environmental data. Represents the actual measured average value of environmental data i in scenic area x during the t-th time period. Represents the standard threshold of environmental data i. Represents the weight coefficient of environmental data i.

[0009] Furthermore, by analyzing the tendency words appearing in the text content of social media data, determine the emotional tendency characteristic value of the scenic area. The specific method is as follows: Using the formula Represents the emotional tendency characteristic value, where t represents the t-th time period, and x represents the x-th scenic area. Represents the emotional tendency characteristic value of scenic area x during the t-th time period. m represents m social media data, and j represents the j-th social media data. Represents the number of times the tendency words appear in the text content of the j-th social media data about scenic area x during the t time period. Represents the tendency degree value of the j-th social media data about scenic area x during the t time period. Represents the frequency weight coefficient. Represents the degree weight coefficient.

[0010] Furthermore, based on the total tourist flow data of the scenic area within a time period and the geographical distribution variance of different tourist sources, determine the attraction characteristic value by integrating the commercial data of the scenic area. The specific method is as follows: Using the formula Represents the attraction characteristic value, where t represents the t-th time period, and x represents the x-th scenic area. Represents the attraction characteristic value of scenic area x during the t-th time period. Represents the total number of tourists in scenic area x during the t-th time period. Represents the average stay duration of tourists in scenic area x during the t time period. Represents the average consumption of tourists in scenic area x during the t time period. Represents the geographical distribution variance of the tourist sources of scenic area x during the t-th time period.

[0011] Furthermore, set the resource evaluation index by integrating the characteristic values of the cultural and tourism resource data. The specific method is as follows: Integrate the characteristic values of the Internet of Things data, emotional tendency characteristic values, and attraction characteristic values of the scenic area within N time periods, and use the formula Represents the resource evaluation index, where Represents the resource evaluation index of scenic area x. Represents the average value of the IoT data characteristic values of scenic area x within N time periods. Represents the maximum value of the IoT data characteristic values of scenic area x within N time periods. Represents the average value of the ratio of the IoT data characteristic values of scenic area x within N time periods. Represents the average value of the emotional tendency characteristic values of scenic area x within N time periods. Represents the average value of the attraction characteristic values of scenic area x within N time periods.

[0012] Furthermore, based on the resource evaluation index, the investigation and evaluation of the cultural and tourism resources in the corresponding area are completed. The specific method is as follows: Based on the resource evaluation index, an evaluation interval is set, and the evaluation interval is divided by the threshold J. When is less than the threshold J, it means that the evaluation of scenic area x is unqualified and belongs to the cultural and tourism resources restricted from development. When is greater than or equal to the threshold J and less than J + A (A is a constant value), it means that the evaluation of scenic area x is qualified, but there are problems to be optimized. For example, noise reduction projects need to be implemented. When is greater than or equal to J + A, it means that the evaluation of scenic area x is excellent and belongs to the cultural and tourism resources for priority development.

[0013] The present invention also provides a cultural and tourism resource investigation system based on multi-source data fusion and artificial intelligence, which is applied to the method for investigating cultural and tourism resources based on multi-source data fusion and artificial intelligence, including: A cultural and tourism resource data collection module, used to establish a distributed cultural and tourism resource data acquisition system and collect the cultural and tourism resource data of each region through multiple channels. A cultural and tourism resource data characteristic value definition module, used to analyze the data characteristics of multi-source cultural and tourism resource data and comprehensively define the characteristic values of cultural and tourism resource data. A cultural and tourism resource investigation and evaluation module, used to set resource evaluation indicators based on the characteristic values of cultural and tourism resource data and complete the investigation and evaluation of the cultural and tourism resources in the corresponding area according to the resource evaluation indicators. In summary, due to the adoption of the above technical solutions, the beneficial effects of the present invention are as follows: 1. By establishing a distributed cultural and tourism resource data acquisition system, the present invention can efficiently integrate data from multiple channels. Using the heterogeneous data integration framework of Apache NiFi, the system can adapt to different data sources and data formats in real time and dynamically. Through comprehensive, efficient, and accurate data collection and processing, it provides strong data support and technical guarantee for the intelligent management and optimization of cultural and tourism resources, and helps to promote the sustainable development and innovative upgrading of the cultural and tourism industry. 2. By comprehensively considering the tourist flow characteristics and environmental quality characteristics, the present invention determines the characteristic values of the Internet of Things data in the scenic area, providing comprehensive data support for the intelligent management and decision-making of the scenic area. By means of the preset cultural and tourism vocabulary database and the degree value of tendency vocabulary, the emotional tendency of the social media data is judged, providing an important basis for the reputation management of the scenic area and the evaluation of tourist satisfaction. By comprehensively considering the total tourist flow, tourist stay duration, average consumption, and the geographical distribution variance of tourist sources, reliable data support is provided for the marketing and strategic planning of the scenic area; 3. The present invention comprehensively considers multiple dimensions such as the characteristic values of the Internet of Things data, emotional tendency characteristic values, and attraction characteristic values in the scenic area, ensuring the comprehensiveness and accuracy of resource evaluation. According to these indicators, the investigation and evaluation of cultural and tourism resources in the corresponding areas are completed, providing strong support and guarantee for the reasonable development, optimized management, and sustainable development of cultural and tourism resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0015] Figure 1 Shows the method step diagram of the cultural and tourism resource investigation method based on multi-source data fusion and artificial intelligence; Figure 2 Shows the method step diagram of setting resource evaluation indicators based on the characteristic values of comprehensive cultural and tourism resource data; Figure 3 Shows the module diagram of the cultural and tourism resource investigation system based on multi-source data fusion and artificial intelligence. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0017] Embodiment 1. As Figure 1 、 Figure 2 shown, the cultural and tourism resource investigation method based on multi-source data fusion and artificial intelligence specifically includes the following steps: Step 1. Establish a distributed cultural and tourism resource data acquisition system, and collect cultural and tourism resource data in each region through various channels.

[0018] Determine the area range of cultural and tourism resources to be investigated, denoted as the investigation area, and determine the scenic spots owned by the investigation area. In this embodiment, a place or area that provides tourist services and has clear management boundaries is defined as a scenic spot. Build a distributed cultural and tourism resource data collection system based on the heterogeneous data integration framework of Apache NiFi. Divide time periods by T duration, and use the distributed data crawling system to collect scenic spot IoT data, social media data, and scenic spot business data in real time for each time period. The scenic spot IoT data, social media data, and scenic spot business data are collectively referred to as cultural and tourism resource data; Determine the number and types of scenic spots owned by the investigation area. By presetting the LoRaWAN low-power wide-area network in different scenic spots, integrating various sensor devices in the scenic spots, and combining binocular vision and infrared thermal imaging technology, the number of tourists in the scenic spot is monitored in real time. The environmental data of the scenic spot is monitored in real time through environmental monitoring sensor devices. The environmental data refers to data related to the environmental quality of the scenic spot, such as the average noise frequency, water turbidity, and air quality during the business hours of the scenic spot. The number of tourists data and environmental data are recorded as scenic spot IoT data; Build a distributed crawler cluster containing z nodes, and collect data on the scenic spot social platforms in parallel at different social platform nodes. For text data, use the BERT model to extract keywords from the text. For image or video data, use the ResNet152 feature retrieval technology to extract high-level features in the image, and judge whether they are similar by comparing the feature vectors of different images or videos, so as to remove duplicate content. The text data, images, or video data published by the same user at the same time are recorded as one piece of social media data; Connect to the scenic spot OTA platform through the OpenAPI gateway, and use OAuth 2.0+JWT to securely obtain the scenic spot OTA reservation data. Determine the cancellation volume, per capita consumption, and tourist stay duration of the existing scenic spots in the investigation area, and record them as scenic spot business data.

[0019] Step 2: Analyze the data characteristics of multi-source cultural and tourism resource data, and comprehensively define the characteristic values of cultural and tourism resource data.

[0020] Conduct time series analysis on the number of tourists data, extract the change trend and peak period of tourist flow, and use statistical analysis methods to quantify the fluctuation of tourist flow by calculating the average value and standard deviation of the number of tourists per day within a time period, and obtain the tourist flow characteristics. Normalize the environmental data to eliminate the influence of different dimensions on data analysis, and extract the environmental quality characteristics. Combine the tourist flow characteristics and environmental quality characteristics to determine the characteristic values of the scenic spot IoT data. The specific formula is as follows: ; where t represents the t-th time period, and x represents the x-th scenic spot, Denote the IoT data characteristic value of scenic area x in the t-th time period. a represents the a-th day, h represents a total of h days. Denote the tourist flow of scenic area x on the a-th day in the t-th time period. Denote the average value of the daily tourist flow of scenic area x in the t-th time period. Denote the maximum value of the tourist flow within a day of scenic area x in the t-th time period. Denote the minimum value of the tourist flow within a day of scenic area x in the t-th time period. Denote the total tourist flow of scenic area x in the t-th time period. n represents n types of environmental data, and i represents the i-th type of environmental data. Denote the actual measured average value of environmental data i of scenic area x in the t-th time period. Denote the standard threshold of environmental data i. Denote the weight coefficient of environmental data i. In this embodiment, the standard thresholds of environmental data are set with reference to the "Tourist Scenic Area Quality Grade Assessment Standard". Preprocess the social media data, remove the word segmentation, stop words and repeated words contained in the text of the social media data, and adjust the image data and video data to a preset format. Design a cross-modal mapping model based on the CLIP architecture to align the latent space features of text, images, and sensor data. Decompose the video data into continuous image data at a preset frame rate, extract the audio content through video editing software, transcribe the audio content part into text content using the Whisper model, extract the text content in the image data using OCR technology, preset a cultural and tourism vocabulary database for the scenic area, which is composed of tendency words used to judge the overall emotional tendency of the text. Each tendency word corresponds to a degree value. When the degree value is positive, it indicates that the tendency word has a positive tendency, and when the degree value is negative, it indicates that the tendency word has a negative tendency. Judge the emotional tendency of the text content by calculating the sum value of the degree values of all tendency words in the text content. Extract the tendency words of the social media review text in the cultural and tourism resource data through the DeBERTa-v3 model. Calculate the sum of the degree values of the tendency words in the text content of the social media data to obtain the tendency degree value of the social media data. Determine the tendency degree values of all social media data within a time period to obtain the emotional tendency characteristic value of the scenic area within that time period. The specific calculation formula is as follows: ; Among them, t represents the t-th time period, and x represents the x-th scenic area. Denote the emotional tendency characteristic value of scenic area x in the t-th time period. m represents m social media data, and j represents the j-th social media data. Denote the number of occurrences of tendency words in the text content of the j-th social media data about scenic area x in the time period t. represents the inclination degree value of the j-th social media data regarding scenic area x in time period t. represents the frequency weight coefficient. represents the degree weight coefficient.

[0021] Determine the accuracy of the commercial data of the scenic area. Based on the total tourist flow data of the scenic area and the geographical distribution variance of different tourist source areas within a time period, comprehensively determine the attractiveness eigenvalue of the scenic area according to the commercial data of the scenic area. The specific formula is as follows: ; where t represents the t-th time period, and x represents the x-th scenic area. represents the attractiveness eigenvalue of scenic area x in the t-th time period. represents the total tourist flow of scenic area x in the t-th time period. represents the average staying duration of tourists in scenic area x during time period t. represents the average consumption of tourists in scenic area x during time period t. represents the geographical distribution variance of the tourist source areas of scenic area x in the t-th time period. The larger the variance, the more dispersed the tourist sources; the smaller the variance, the more concentrated the tourist sources.

[0022] Step 3: Set resource evaluation indicators based on the eigenvalues of the comprehensive cultural and tourism resource data, and complete the investigation and evaluation of the cultural and tourism resources in the corresponding area according to the resource evaluation indicators.

[0023] Set resource evaluation indicators by comprehensively considering the eigenvalues of the Internet of Things data, emotional inclination eigenvalues, and attractiveness eigenvalues of the scenic area within N time periods. The specific formula is as follows: ; where represents the resource evaluation indicator of scenic area x. represents the average value of the eigenvalues of the Internet of Things data of scenic area x within N time periods. represents the maximum value of the eigenvalues of the Internet of Things data of scenic area x within N time periods. represents the average value of the ratio of the eigenvalues of the Internet of Things data of scenic area x within N time periods. represents the average value of the emotional inclination eigenvalues of scenic area x within N time periods. represents the average value of the attractiveness eigenvalues of scenic area x within N time periods.

[0024] Furthermore, The specific formula of ; where t represents the t-th time period, and x represents the x-th scenic area. Denote the IoT data characteristic value of scenic area x in the t-th time period, and N represents a total of N time periods.

[0025] Based on the resource evaluation index, set the evaluation interval, and divide the evaluation interval with the threshold J. When is less than the threshold J, it means that the evaluation of scenic area x is unqualified, and it belongs to the cultural and tourism resources restricted from development. When is greater than or equal to the threshold J and less than J+A, where A is a constant value, it means that the evaluation of scenic area x is qualified, but there are problems to be optimized, such as the need to implement a noise reduction project. When is greater than or equal to J+A, it means that the evaluation of scenic area x is excellent, and it belongs to the cultural and tourism resources given priority for development.

[0026] Embodiment 2. As Figure 3 shown, the cultural and tourism resource investigation system based on multi-source data fusion and artificial intelligence specifically includes: The cultural and tourism resource data collection module is used to determine the area range to be investigated for cultural and tourism resources, denoted as the investigation area, determine the scenic areas owned by the investigation area. In this embodiment, a place or area that provides tourist services and has clear management boundaries is defined as a scenic area. A distributed cultural and tourism resource data acquisition system is constructed based on the heterogeneous data integration framework of Apache NiFi. Divide the time period by T duration, and real-time collect the scenic area IoT data, social media data and scenic area business data of each time period through the distributed data crawling system. The scenic area IoT data, social media data and scenic area business data are collectively referred to as cultural and tourism resource data; Determine the number and type of scenic areas owned by the investigation area. By pre-setting the LoRaWAN low-power wide-area network in different scenic areas, integrating various sensor devices in the scenic areas, and combining binocular vision and infrared thermal imaging technology, real-time monitor the number of tourists in the scenic area. Real-time monitor the environmental data of the scenic area through environmental monitoring sensor devices. The environmental data refers to the data related to the environmental quality of the scenic area, such as the average noise frequency, water turbidity and air quality during the business hours of the scenic area. Record the number of tourists data and environmental data as scenic area IoT data; Construct a distributed crawler cluster containing z nodes, and collect data on the scenic area social platforms in parallel at different social platform nodes. For text data, extract keywords from the text through the BERT model. For image or video data, use the ResNet152 feature retrieval technology to extract the high-level features in the image, and judge whether they are similar by comparing the feature vectors of different images or videos, so as to remove duplicate content. Record the text data, images or video data published by the same user at the same time as a social media data; Connect to the scenic area OTA platform through the OpenAPI gateway, and use OAuth 2.0+JWT to securely obtain the reservation data of the scenic area OTA. Determine the cancellation volume, per capita consumption, and tourist stay duration of the existing scenic areas in the investigation area, which are recorded as the commercial data of the scenic area.

[0027] The cultural and tourism resource data eigenvalue definition module performs time series analysis on the tourist volume data, extracts the change trend and peak period of the tourist flow, and uses statistical analysis methods to quantify the fluctuation of the tourist flow by calculating the average value and standard deviation of the daily tourist volume within a period of time, obtaining the tourist flow characteristics. Normalize the environmental data to eliminate the influence of different dimensions on data analysis, extract the environmental quality characteristics, and combine the tourist flow characteristics and environmental quality characteristics to determine the eigenvalue of the Internet of Things data of the scenic area. The specific formula is as follows: ; Among them, t represents the t-th time period, x represents the x-th scenic area, represents the eigenvalue of the Internet of Things data of scenic area x in the t-th time period, a represents the a-th day, h represents a total of h days, represents the tourist flow of scenic area x on the a-th day in the t-th time period, represents the average value of the daily tourist flow of scenic area x in the t-th time period, represents the maximum value of the daily tourist flow of scenic area x in the t-th time period, represents the minimum value of the daily tourist flow of scenic area x in the t-th time period, represents the total tourist flow of scenic area x in the t-th time period, n represents n types of environmental data, and i represents the i-th type of environmental data, represents the actual measured average value of environmental data i of scenic area x in the t-th time period, represents the standard threshold of environmental data i, represents the weight coefficient of environmental data i. In this embodiment, the standard threshold of environmental data is set with reference to the "Tourist Scenic Area Quality Grade Assessment Standard"; Preprocess the social media data to remove the word segmentation, stop words, and duplicate words in the text of the social media data, and adjust the image data and video data to a preset format. Design a cross-modal mapping model based on the CLIP architecture to align the latent space features of text, images, and sensor data. Decompose the video data into continuous image data at a preset frame rate, extract the audio content through video editing software, use the Whisper model to transcribe the audio content part into text content, use OCR technology to extract the text content in the image data, and preset a cultural and tourism vocabulary database for the scenic area. This database consists of tendency words, which are used to judge the overall emotional tendency of the text. Each tendency word corresponds to a degree value. When the degree value is positive, it indicates that the tendency word has a positive tendency, and when the degree value is negative, it indicates that the tendency word has a negative tendency. By calculating the sum of the degree values of all tendency words in the text content, judge the emotional tendency of the text content. Extract the tendency words of the social media review text in the cultural and tourism resource data through the DeBERTa-v3 model. By analyzing the tendency words appearing in the text content of the social media data, calculate the sum of the degree values of the tendency words in the text to obtain the tendency degree value of the social media data. By determining the tendency degree values of all social media data within a time period, obtain the emotional tendency characteristic value of the scenic area within that time period. The specific calculation formula is as follows: ; Among them, \(t\) represents the \(t\)th time period, \(x\) represents the \(x\)th scenic area, represents the IoT data characteristic value of scenic area \(x\) in the \(t\)th time period, \(a\) represents the \(a\)th day, and \(h\) represents a total of \(h\) days, represents the tourist flow of scenic area \(x\) on the \(a\)th day in the \(t\)th time period, represents the average daily tourist flow of scenic area \(x\) in the \(t\)th time period, represents the maximum daily tourist flow of scenic area \(x\) in the \(t\)th time period, represents the minimum daily tourist flow of scenic area \(x\) in the \(t\)th time period, represents the total tourist flow of scenic area \(x\) in the \(t\)th time period, \(n\) represents \(n\) types of environmental data, and \(i\) represents the \(i\)th type of environmental data, represents the actual measured average value of environmental data \(i\) of scenic area \(x\) in the \(t\)th time period, represents the standard threshold of environmental data \(i\), represents the weight coefficient of environmental data \(i\).

[0028] Determine the accuracy of the scenic area business data. Based on the total tourist flow data of the scenic area within a time period and the geographical distribution variance of different tourist source areas, comprehensively determine the attraction characteristic value according to the scenic area business data. The specific formula is as follows: ; Among them, \(t\) represents the \(t\)-th time period, and \(x\) represents the \(x\)-th scenic area. represents the attractiveness eigenvalue of scenic area \(x\) in the \(t\)-th time period. represents the total tourist flow of scenic area \(x\) in the \(t\)-th time period. represents the average stay duration of tourists in scenic area \(x\) during the \(t\)-th time period. represents the average consumption of tourists in scenic area \(x\) during the \(t\)-th time period. represents the variance of the geographical distribution of tourist sources in scenic area \(x\) in the \(t\)-th time period. The larger the variance, the more dispersed the tourist sources; the smaller the variance, the more concentrated the tourist sources.

[0029] The cultural and tourism resource investigation and evaluation module sets resource evaluation indicators by comprehensively considering the IoT data eigenvalues, sentiment tendency eigenvalues, and attractiveness eigenvalues of scenic areas within \(N\) time periods. The specific formula is as follows: ; Among them, represents the resource evaluation indicator of scenic area \(x\). represents the average value of the IoT data eigenvalues of scenic area \(x\) within \(N\) time periods. represents the maximum value of the IoT data eigenvalues of scenic area \(x\) within \(N\) time periods. represents the average value of the ratio of the IoT data eigenvalues of scenic area \(x\) within \(N\) time periods. represents the average value of the sentiment tendency eigenvalues of scenic area \(x\) within \(N\) time periods. represents the average value of the attractiveness eigenvalues of scenic area \(x\) within \(N\) time periods.

[0030] Furthermore, The specific formula of is as follows: ; Among them, \(t\) represents the \(t\)-th time period, and \(x\) represents the \(x\)-th scenic area. represents the IoT data eigenvalue of scenic area \(x\) in the \(t\)-th time period, and \(N\) represents a total of \(N\) time periods.

[0031] Based on the resource evaluation indicator, an evaluation interval is set, and the evaluation interval is divided by the threshold \(J\). When is less than the threshold \(J\), it means that the evaluation of scenic area \(x\) is unqualified, and it belongs to the cultural and tourism resources restricted from development. When is greater than or equal to the threshold \(J\) and less than \(J + A\) (\(A\) is a constant value), it means that the evaluation of scenic area \(x\) is qualified, but there are problems to be optimized, such as the need to implement noise reduction projects. When is greater than or equal to \(J + A\), it means that the evaluation of scenic area \(x\) is excellent, and it belongs to the cultural and tourism resources given priority for development.

[0032] The above are only the preferred specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, making equivalent substitutions or changes, shall be covered by the protection scope of the present invention.

[0033] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the present invention to only the specific embodiments. Obviously, according to the content of this specification, many modifications and changes can be made. This specification selects and specifically describes these embodiments in order to better explain the principle and practical application of the present invention, so that those skilled in the art can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A cultural and tourism resource survey method based on multi-source data fusion and artificial intelligence, characterized in that: include: Step 1: Establish a distributed cultural and tourism resource data collection system to collect cultural and tourism resource data in various regions through multiple channels; Step 2: Analyze the data characteristics of multi-source cultural and tourism resource data and comprehensively define the characteristic values ​​of cultural and tourism resource data; Step 3: Set resource evaluation indicators based on the characteristic values ​​of comprehensive cultural and tourism resource data, and complete the investigation and evaluation of cultural and tourism resources in the corresponding area according to the resource evaluation indicators.

2. The cultural tourism resource investigation method based on multi-source data fusion and artificial intelligence according to claim 1 is characterized in that: Establish a distributed cultural and tourism resource data collection system. The specific methods are as follows: Determine the area scope of cultural and tourism resources to be investigated, recorded as the investigation area, and determine the scenic spots owned by the investigation area. In this embodiment, places or areas that provide sightseeing services and have clear management boundaries are defined as scenic spots. A distributed cultural and tourism resource data collection system is built based on the heterogeneous data integration framework of Apache NiFi. The time period is divided into T time periods. The scenic spot Internet of Things data, social media data and scenic spot business data of each time period are collected in real time through the distributed data crawling system. The scenic spot Internet of Things data, social media data and scenic spot business data are collectively referred to as cultural and tourism resource data.

3. The cultural tourism resource investigation method based on multi-source data fusion and artificial intelligence according to claim 1 is characterized in that: Analyze the data characteristics of multi-source cultural and tourism resource data and comprehensively define the characteristic values ​​of cultural and tourism resource data. The specific method is as follows: Conduct time series analysis on the number of tourists data, extract the changing trend and peak period of tourist flow, use statistical analysis methods to quantify the fluctuation of tourist flow by calculating the average and standard deviation of the number of tourists per day in a period of time, obtain the characteristics of tourist flow, normalize the environmental data, eliminate the influence of different dimensions on data analysis, extract environmental quality characteristics, integrate tourist flow characteristics and environmental quality characteristics, and determine the characteristic values ​​of the scenic area's IoT data; Preprocess the social media data, design a cross-modal mapping model based on the CLIP architecture, align the latent spatial features of text, image, and sensor data, decompose the video data into continuous image data at a preset frame rate, extract the audio content through video editing software, use the Whisper model to partially transcribe the audio content into text content, use OCR technology to extract the text content in the image data, preset a cultural and tourism vocabulary database composed of preference words, extract the preference words of social media comment texts in cultural and tourism resource data through the DeBERTa-v3 model, and determine the sentiment preference feature value of the scenic spot by analyzing the preference words appearing in the text content of social media data; Determine the accuracy of the scenic spot's commercial data, based on the total tourist flow data of the scenic spot within a period of time and the geographical distribution variance of different tourist sources, and determine the attraction characteristic value by integrating the scenic spot's commercial data.

4. The method for investigating cultural and tourism resources based on multi-source data fusion and artificial intelligence according to claim 1 is characterized in that: The characteristics of tourist flow and environmental quality are combined to determine the characteristic values ​​of the IoT data of the scenic spot. The specific method is as follows: Using the formula represents the IoT data feature value of the scenic spot, where t represents the t-th time period, x represents the x-th scenic spot, represents the IoT data feature value of scenic spot x in the tth time period, a represents the ath day, h represents a total of h days, represents the tourist flow of scenic spot x on day a during time period t, represents the average daily tourist flow of scenic spot x in time period t, It represents the maximum tourist flow of scenic spot x in one day during time period t. It represents the minimum tourist flow of scenic spot x in one day during time period t. represents the total tourist flow of scenic spot x in time period t, n represents n types of environmental data, i represents the i-th type of environmental data, represents the actual measured average value of environmental data i in scenic spot x in the tth time period, represents the standard threshold of environmental data i, Represents the weight coefficient of environmental data i.

5. The method for investigating cultural and tourism resources based on multi-source data fusion and artificial intelligence according to claim 1 is characterized in that: By analyzing the trend words appearing in the text content of social media data, the sentiment trend feature value of the scenic spot is determined. The specific method is as follows: Using the formula represents the characteristic value of sentiment tendency, where t represents the t-th time period, x represents the x-th scenic spot, represents the sentiment characteristic value of scenic spot x in the tth time period, m represents m social media data, j represents the jth social media data, represents the number of occurrences of the preferred words in the j-th social media data text content about scenic spot x in time period t, represents the inclination value of the j-th social media data about scenic spot x in time period t, represents the frequency weight coefficient, Represents the degree weight coefficient.

6. The method for investigating cultural and tourism resources based on multi-source data fusion and artificial intelligence according to claim 1 is characterized in that: Based on the total tourist flow data of the scenic spot within a period of time and the geographical distribution variance of different tourist sources, the attraction characteristic value is determined by integrating the commercial data of the scenic spot. The specific method is as follows: Using the formula represents the attraction characteristic value, where t represents the t-th time period, x represents the x-th scenic spot, represents the attractiveness characteristic value of scenic spot x in the tth time period, represents the total tourist flow of scenic spot x in the tth time period, represents the average length of stay of tourists in scenic spot x during time period t, represents the average consumption of tourists in scenic spot x during time period t, It represents the variance of the geographical distribution of tourists’ origins in scenic spot x in the tth time period.

7. The method for investigating cultural and tourism resources based on multi-source data fusion and artificial intelligence according to claim 1 is characterized in that: The characteristic values ​​of comprehensive cultural and tourism resource data are used to set resource evaluation indicators. The specific method is as follows: Comprehensively consider the IoT data feature values, emotional tendency feature values, and attractiveness feature values ​​of the scenic spot in N time periods, and use the formula Represents resource evaluation indicators, where represents the resource evaluation index of scenic spot x, It represents the average value of the IoT data characteristic value of scenic spot x in N time periods. It represents the maximum value of the IoT data characteristic value of scenic spot x in N time periods. It represents the average value of the IoT data characteristic value ratio of scenic spot x in N time periods. represents the average value of the sentiment tendency characteristic value of scenic spot x in N time periods, It represents the average value of the attraction characteristic value of scenic spot x in N time periods.

8. The method for investigating cultural and tourism resources based on multi-source data fusion and artificial intelligence according to claim 1 is characterized in that: Complete the investigation and evaluation of the cultural and tourism resources in the corresponding area according to the resource evaluation indicators. The specific methods are as follows: Based on the resource evaluation index, the evaluation interval is set and the evaluation interval is divided by threshold J. When it is less than the threshold value J, it means that the scenic spot x is not qualified and belongs to the cultural tourism resources that are restricted from development. When it is greater than or equal to the threshold value J and less than J+A, A is a constant value, which means that the scenic spot x is qualified, but there are problems to be optimized, such as the need to implement noise reduction projects. When it is greater than or equal to J+A, it means that scenic spot x has been evaluated as excellent and is a cultural and tourism resource that is given priority for development.

9. A cultural tourism resource survey system based on multi-source data fusion and artificial intelligence, applied to a cultural tourism resource survey method based on multi-source data fusion and artificial intelligence as claimed in any one of claims 1 to 8, characterized in that: include: The cultural and tourism resource data collection module is used to establish a distributed cultural and tourism resource data collection system to collect cultural and tourism resource data in various regions through multiple channels; The cultural and tourism resource data characteristic value definition module is used to analyze the data characteristics of multi-source cultural and tourism resource data and comprehensively define the characteristic values ​​of cultural and tourism resource data; The cultural and tourism resource survey and evaluation module is used to set resource evaluation indicators based on the characteristic values ​​of cultural and tourism resource data, and complete the survey and evaluation of cultural and tourism resources in the corresponding area according to the resource evaluation indicators.

Images