Tourism resource monomer combination method, device, storage medium and computer equipment

Abstract

The present invention discloses a combination method, device, storage medium and computer equipment for tourism resource monomers, and relates to the field of tourism resource processing technology, mainly focusing on the combination accuracy and efficiency of tourism resource monomers. The method includes: obtaining travel notes data, geographic information data and tourist trajectory data in a target area; performing entity recognition on resource description data to obtain tourism resource monomers; performing classification and grading processing on tourism resource monomers to obtain resource categories and monomer levels; determining the monomer distance between each tourism resource monomer, and determining the spatial selectivity of tourism resource monomers; determining the degree of association between each tourism resource monomer, and clustering each tourism resource monomer to obtain cluster categories; constructing a knowledge graph based on resource classification, monomer level, monomer distance, spatial selectivity, association degree and cluster category, and determining the combination scheme of tourism resource monomers in the target area based on the knowledge graph.

Description

Technical Field

[0001] The present invention relates to the technical field of tourism resource processing, and in particular to a combination method, device, storage medium and computer equipment of tourism resource monomers. Background Art

[0002] With the rapid development of tourism, centralized management of tourism resources is needed. Therefore, the combination of tourism resource monomers becomes particularly important.

[0003] At present, tourism resource monomers are usually combined by artificial means. However, this combination method is greatly affected by human subjective factors, resulting in low combination accuracy of tourism resource monomers. At the same time, this artificial method also leads to low combination efficiency of tourism resource monomers. Summary of the invention

[0004] The present invention provides a method, device, storage medium and computer equipment for combining tourism resource monomers, which are mainly capable of improving the combination accuracy and efficiency of tourism resource monomers.

[0005] According to a first aspect of the present invention, there is provided a method for combining tourism resource monomers, comprising:

[0006] Obtain resource description data in the target area, wherein the resource description data includes: travel notes data, geographic information data, and tourist trajectory data;

[0007] Performing tourism resource entity recognition on the resource description data to obtain each tourism resource entity contained in the resource description data;

[0008] Classify each of the tourism resource monomers to obtain resource category information to which each of the tourism resource monomers belongs, and grade each of the tourism resource monomers to obtain monomer level information corresponding to each of the tourism resource monomers;

[0009] Based on the geographic information data, determining the monomer distance information between each of the tourism resource monomers, and determining the spatial selectivity information of each of the tourism resource monomers;

[0010] Based on the travel notes data, determining the correlation degree information between the tourism resource monomers, and performing clustering processing on the tourism resource monomers to obtain the cluster category information to which the tourism resource monomers belong;

[0011] Based on the resource category information, monomer level information, monomer distance information, spatial selectivity information, association degree information and cluster category information, a tourism resource knowledge graph corresponding to each of the tourism resource monomers is constructed, wherein each of the tourism resource monomers is used as a node in the tourism resource knowledge graph, the monomer distance information is used as an edge in the tourism resource knowledge graph, the resource category information is used as a node label of the corresponding node in the tourism resource knowledge graph, and the monomer level information, spatial selectivity information, association degree information and cluster category information are used as attribute information of the corresponding node in the tourism resource knowledge graph;

[0012] Based on the tourism resource knowledge graph, a combination scheme of tourism resource monomers in the target area is determined.

[0013] According to a second aspect of the present invention, there is provided a combination device of tourism resource monomers, comprising:

[0014] An acquisition unit, used to acquire resource description data in a target area, wherein the resource description data includes: travel notes data, geographic information data, and tourist trajectory data;

[0015] An entity recognition unit, used to perform tourism resource entity recognition on the resource description data to obtain each tourism resource entity contained in the resource description data;

[0016] A processing unit, used to classify each of the tourism resource monomers to obtain resource category information to which each of the tourism resource monomers belongs, and to grade each of the tourism resource monomers to obtain monomer level information corresponding to each of the tourism resource monomers;

[0017] A first determining unit is used to determine the monomer distance information between each of the tourism resource monomers and the spatial selectivity information of each of the tourism resource monomers based on the geographic information data;

[0018] A second determining unit is used to determine the association degree information between the tourism resource monomers based on the travel record data, and cluster the tourism resource monomers to obtain the cluster category information to which the tourism resource monomers belong;

[0019] A construction unit, for constructing a tourism resource knowledge graph corresponding to each of the tourism resource monomers based on the resource category information, monomer level information, monomer distance information, spatial selectivity information, association degree information and cluster category information, wherein each of the tourism resource monomers is used as a node in the tourism resource knowledge graph, the monomer distance information is used as an edge in the tourism resource knowledge graph, the resource category information is used as a node label of the corresponding node in the tourism resource knowledge graph, and the monomer level information, spatial selectivity information, association degree information and cluster category information are used as attribute information of the corresponding node in the tourism resource knowledge graph;

[0020] The third determining unit is used to determine a combination scheme of tourism resource monomers in the target area based on the tourism resource knowledge graph.

[0021] According to a third aspect of the present invention, there is provided a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above method for combining tourism resource monomers.

[0022] According to a fourth aspect of the present invention, there is provided a computer device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for combining tourism resource monomers when executing the program.

[0023] According to a method, device, storage medium and computer equipment for combining tourism resource monomers provided by the present invention, compared with the current way of combining tourism resource monomers by manual means, the present invention obtains resource description data in a target area, wherein the resource description data includes: travel notes data, geographic information data, and tourist trajectory data; and performs tourism resource entity recognition on the resource description data to obtain each tourism resource monomer contained in the resource description data; at the same time, each tourism resource monomer is classified to obtain resource category information to which each tourism resource monomer belongs, and each tourism resource monomer is graded to obtain monomer level information corresponding to each tourism resource monomer; and based on the geographic information data, the monomer distance information between each tourism resource monomer is determined, as well as the spatial selectivity information of each tourism resource monomer is determined; at the same time, based on the The method comprises the following steps: according to the travel data, determining the association degree information between each tourism resource monomer, and clustering each tourism resource monomer to obtain the cluster category information to which each tourism resource monomer belongs; then, based on the resource category information, monomer level information, monomer distance information, spatial selectivity information, association degree information and cluster category information, constructing a tourism resource knowledge graph corresponding to each tourism resource monomer, wherein each tourism resource monomer serves as a node in the tourism resource knowledge graph, the monomer distance information serves as an edge in the tourism resource knowledge graph, the resource category information serves as a node label of the corresponding node in the tourism resource knowledge graph, and the monomer level information, spatial selectivity information, association degree information and cluster category information serve as attribute information of the corresponding node in the tourism resource knowledge graph; finally, based on the tourism resource knowledge graph, determining the combination scheme of tourism resource monomers in the target area. Therefore, by identifying each tourism resource monomer contained in the resource description data, and determining the resource category information, monomer level information, monomer distance information, spatial selectivity information, correlation degree information and clustering category information of each tourism resource monomer, a tourism resource knowledge graph is constructed according to the above-mentioned information, and finally the tourism resource monomers in the target area are combined according to the tourism resource knowledge graph, that is, by determining various attribute information of tourism resource monomers and constructing a knowledge graph based on various attribute information, monomers are combined through a knowledge graph containing various attribute information, which can improve the combination efficiency and combination accuracy of tourism resource monomers. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0025] Figure 1A flow chart of a method for combining tourism resource monomers provided by an embodiment of the present invention is shown;

[0026] Figure 2 A flow chart of another method for combining tourism resource monomers provided by an embodiment of the present invention is shown;

[0027] Figure 3 A schematic diagram showing the structure of a combination device of a tourism resource monomer provided by an embodiment of the present invention is shown;

[0028] Figure 4 A schematic structural diagram of another combination device of tourism resource monomers provided by an embodiment of the present invention is shown;

[0029] Figure 5 A schematic diagram of the physical structure of a computer device provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0030] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0031] At present, the method of combining tourism resource monomers through artificial means leads to low combination accuracy of tourism resource monomers. At the same time, this artificial method also leads to low combination efficiency of tourism resource monomers.

[0032] In order to solve the above problems, the embodiment of the present invention provides a method for combining tourism resource monomers, such as Figure 1 As shown, the method includes:

[0033] 101. Obtain resource description data in a target area, wherein the resource description data includes: travel notes data, geographic information data, and tourist trajectory data.

[0034] For the embodiment of the present invention, travel data specifically refers to tourists' comment data. Various travel websites are visited to identify the URL (uniform resource locator) pattern and web page structure of the tourists' comment data. By writing a Python (a computer programming language) program, the tourists' comment data in the target area are crawled and stored in the relational database access, including user ID, comment content, timestamp field and other information. At the same time, the tourists' trajectory data in the target area is obtained through the API, including the latitude and longitude of the trajectory points, timestamp field and other information, and the Overpass API tool is used to download the latest road network and attractions in the target area, that is, geographic information data.

[0035] Furthermore, after obtaining the travel notes data, geographic information data, and tourist trajectory data of the target area, the above data needs to be preprocessed. For example, for the travel notes data of tourists, delete the data with less than 5 Chinese characters, delete duplicate data, delete advertisements and other irrelevant comments; for the tourist trajectory data, delete the data with a length of less than 1km and the data with a duration of less than 1h; for the geographic information data, merge the data with different names of the same place, and supplement the names that may appear in all scenic spots. Then, according to the preprocessed data, the tourism resource monomers in the target area are combined, thereby avoiding the problem of low combination efficiency and low combination accuracy caused by artificial combination of tourism resource monomers based on subjective experience, so that the embodiment of the present invention can improve the combination efficiency and combination accuracy of tourism resource monomers. The embodiment of the present invention is mainly applicable to the scene of combining tourism resource monomers. The execution subject of the embodiment of the present invention is a device or equipment that can combine tourism resource monomers, which can be specifically set on the server side.

[0036] 102. Perform tourism resource entity recognition on the resource description data to obtain each tourism resource entity contained in the resource description data.

[0037] Among them, tourism resource monomer refers to a single individual of the basic type of tourism resource that can be used as an independent viewing object or utilization. Tourism resource monomers include independent tourism resource monomers and collective tourism resource monomers. Independent tourism resource monomers refer to tourism resource individuals that can exist independently, such as a historical site, a natural landscape, etc. Collective tourism resource monomers refer to a collection formed by the combination of multiple independent monomers of the same type, such as a group of cultural relics, a natural scenic area, etc. For the embodiment of the present invention, the tourist description data may include different tourism resource monomers. Therefore, in order to extract the tourism resource monomers in the target area, it is necessary to perform entity recognition on the tourist description data in the target area, so as to identify the various tourism resource monomers contained in the target area. In order to ensure that each tourism resource monomer in the target area is identified, remote sensing image data in the target area may also be obtained, and the remote sensing image data may be analyzed and processed to obtain each tourism resource monomer in the target area from the remote sensing image data. Then, the category, level, monomer distance, spatial selectivity information, correlation degree information, etc. of each tourism resource monomer are determined, and a knowledge graph is constructed based on the above data. Finally, based on the knowledge graph, the tourism resource monomers in the target area are combined, thereby analyzing various attribute information of the tourism resource monomers, and constructing a knowledge graph based on the analysis results. Finally, the tourism resource monomers are combined according to the knowledge graph, which can improve the combination efficiency and combination accuracy of the tourism resource monomers.

[0038] 103. Classify each tourism resource monomer to obtain resource category information to which each tourism resource monomer belongs, and grade each tourism resource monomer to obtain monomer level information corresponding to each tourism resource monomer.

[0039] The categories to which the tourism resource monomers belong include: geographical landscape, water landscape, biological landscape, astronomical and climate landscape, buildings and facilities, historical relics, tourist shopping, cultural activities, etc. For the embodiment of the present invention, firstly, the tourism resource monomers are classified according to the characteristic information of each tourism resource monomer to obtain the category information to which each tourism resource monomer belongs, and then, according to the information such as travel notes data, each tourism resource monomer is graded to obtain the monomer level information corresponding to each tourism resource monomer, wherein the monomer level information is affected by the importance of the monomer in the minds of tourists, the satisfaction degree of tourists with the monomer, etc. For example, the greater the importance of the monomer in the minds of tourists, the higher the satisfaction degree of the monomer, and the higher the level corresponding to the monomer.

[0040] 104. Based on geographic information data, determine the distance information between each tourism resource unit, and determine the spatial selectivity information of each tourism resource unit.

[0041] Among them, the spatial selectivity information refers to the number of times a tourist passes through a certain monomer when playing in the target area, which is called the spatial selectivity information of the monomer. For the embodiment of the present invention, the location information of each tourism resource monomer is recorded in the geographic information data. Through the geographic information data, the distance between each tourism resource monomer, that is, the monomer distance information, can be obtained. At the same time, the spatial selectivity information of each tourism resource monomer is determined by the number of times the tourist passes through the tourism resource monomer when playing in the target area. Then, the tourism resource knowledge map is constructed based on the monomer distance information and the spatial selectivity information, which can ensure the comprehensiveness of the information of the knowledge map. Then, the tourism resource monomers are combined according to the knowledge map, which can improve the combination accuracy of the tourism resource monomers.

[0042] 105. Based on the travel notes data, determine the association degree information between the tourism resource monomers, and perform clustering processing on the tourism resource monomers to obtain the cluster category information to which the tourism resource monomers belong.

[0043] The correlation degree information refers to the possibility of tourists visiting monomer B after visiting monomer A. For the embodiment of the present invention, the correlation degree information between various tourism resource monomers can be summarized through travel notes data. At the same time, clustering processing is performed on each tourism resource monomer to obtain the clustering category between each tourism resource monomer. Then, according to the resource category information, monomer level information, monomer distance information, spatial selectivity information, correlation degree information and clustering category information of each tourism resource monomer, a tourism resource knowledge graph corresponding to each tourism resource monomer is constructed. Finally, according to the tourism resource knowledge graph, each tourism resource monomer is combined, so that the tourism resource monomers are combined through the attribute information of each tourism resource monomer clearly displayed in the knowledge graph, which can improve the combination efficiency and combination accuracy of tourism resource monomers.

[0044] 106. Based on resource category information, monomer level information, monomer distance information, spatial selectivity information, association degree information and cluster category information, a tourism resource knowledge graph corresponding to each tourism resource monomer is constructed.

[0045] For the embodiment of the present invention, after obtaining the resource category information, monomer level information, monomer distance information, spatial selectivity information, association degree information and cluster category information of each tourism resource monomer, each tourism resource monomer is used as a node, the monomer distance information between each tourism resource monomer is used as an edge, the resource category information is used as the node label of the corresponding node, the monomer level information, spatial selectivity information, association degree information, cluster category information is used as the attribute information of the corresponding node, and a tourism resource knowledge graph corresponding to each tourism resource monomer is constructed. Finally, the tourism resource monomers are combined according to the resource knowledge graph containing various information, which can improve the combination efficiency and combination accuracy of the tourism resource monomers.

[0046] 107. Based on the tourism resource knowledge graph, determine the combination plan of tourism resource units in the target area.

[0047] For the embodiments of the present invention, a knowledge graph can be constructed specifically through Neo4j (a high-performance graph database). After the tourism resource knowledge graph is constructed, the tourism resource knowledge graph is stored in a search system. The user inputs demand information in the search system, and the search system performs semantic analysis on the user's demand information. Based on the analysis results, the attribute relationship of each monomer is checked in the tourism resource knowledge graph, and the tourism resource monomers are combined according to the information recorded in the knowledge graph. Therefore, by determining various attribute information of each tourism resource monomer and constructing a knowledge graph based on various comprehensive attribute information, the monomers are finally combined according to the knowledge graph, which can improve the combination efficiency and combination accuracy of tourism resource monomers.

[0048] According to a method for combining tourism resource monomers provided by the present invention, compared with the current method of combining tourism resource monomers by manual means, the present invention obtains resource description data in a target area, wherein the resource description data includes: travel notes data, geographic information data, and tourist trajectory data; and performs tourism resource entity recognition on the resource description data to obtain each tourism resource monomer contained in the resource description data; at the same time, classifies each tourism resource monomer to obtain resource category information to which each tourism resource monomer belongs, and performs graded processing on each tourism resource monomer to obtain monomer level information corresponding to each tourism resource monomer; and based on the geographic information, classifies each tourism resource monomer to obtain the information of the tourism resource category to which each tourism resource monomer belongs. The method comprises the following steps: according to the travel information data, determining the monomer distance information between each of the tourism resource monomers, and determining the spatial selectivity information of each of the tourism resource monomers; at the same time, based on the travel notes data, determining the correlation degree information between each of the tourism resource monomers, and clustering each of the tourism resource monomers to obtain the cluster category information to which each of the tourism resource monomers belongs; then, based on the resource category information, monomer level information, monomer distance information, spatial selectivity information, correlation degree information and cluster category information, constructing a tourism resource knowledge graph corresponding to each of the tourism resource monomers; and finally, determining a combination scheme of tourism resource monomers in the target area based on the tourism resource knowledge graph. Therefore, by identifying each tourism resource monomer contained in the resource description data, and determining the resource category information, monomer level information, monomer distance information, spatial selectivity information, correlation degree information and clustering category information of each tourism resource monomer, a tourism resource knowledge graph is constructed according to the above-mentioned information, and finally the tourism resource monomers in the target area are combined according to the tourism resource knowledge graph, that is, by determining various attribute information of tourism resource monomers and constructing a knowledge graph based on various attribute information, monomers are combined through a knowledge graph containing various attribute information, which can improve the combination efficiency and combination accuracy of tourism resource monomers.

[0049] Further, in order to better illustrate the above process of combining tourism resource monomers, as a refinement and extension of the above embodiment, the embodiment of the present invention provides another method for combining tourism resource monomers, such as Figure 2 As shown, the method includes:

[0050] 201. Obtain resource description data in a target area, wherein the resource description data includes: travel notes data, geographic information data, and tourist trajectory data.

[0051] Specifically, crawler technology can be used to obtain various travel notes published by tourists on travel websites, and at the same time obtain geographic information data and tourist trajectory data in the target area. It should be noted that the above data are authorized and agreed to be obtained, and are not tourist privacy data.

[0052] 202. Perform word segmentation processing on the resource description data to obtain each word segment corresponding to the resource description data.

[0053] 203. Input each word segment into a preset entity recognition model for entity recognition to obtain each tourism resource entity included in the resource description data.

[0054] Specifically, the LUA algorithm in the TexSmart natural language processing tool can be used to perform named entity recognition on the resource description data. At the same time, a model including a first recursive neural network and a second recursive neural network can also be used to perform entity recognition on the resource description data. The specific recognition process is: perform word segmentation processing on the resource description data to obtain each word segmentation corresponding to the resource description data, input each word segmentation into the first recursive neural network according to its order in the resource description data for feature extraction, and obtain the first feature vector corresponding to each word segmentation; input each word segmentation into the second recursive neural network in reverse order according to its order in the resource description data for feature extraction, and obtain the second feature vector corresponding to each word segmentation; merge the first feature vector and the second feature vector to obtain the merged feature vector, and determine the entity corresponding to each word segmentation in the resource description data according to the merged feature vector, that is, the tourism resource monomer. The present invention adopts advanced natural language processing and geographic information system technology, and can automatically identify tourism resource monomers from text and image data, reduce the dependence on manual investigation, and improve the speed and accuracy of tourism resource monomer recognition. At the same time, the database used for storing resource description data of the present invention is updated in real time and dynamically, and can adapt to real-time changes in the market, thereby improving the combination accuracy of tourism resource monomers and the practicality of the combined ones.

[0055] 204. Classify each tourism resource monomer to obtain resource category information to which each tourism resource monomer belongs, and grade each tourism resource monomer to obtain monomer level information corresponding to each tourism resource monomer.

[0056] For the embodiment of the present invention, after determining that the resource description data contains each tourism resource monomer, it is also necessary to classify each tourism resource monomer. Based on this, step 204 specifically includes: constructing multiple preset initial classification models, and obtaining sample tourism resource monomers and their corresponding actual category information; based on the number of models of the preset initial classification model, dividing the sample tourism resource monomers and their corresponding actual category information into multiple groups of training data and multiple groups of test data; using each group of training data to train the corresponding preset initial classification model to obtain the trained preset initial classification model, wherein the sample tourism resource monomer is used as input data during the training process, and the actual category information is used as input data. To output data; use each group of test data to test the corresponding trained preset initial classification model to obtain the classification accuracy corresponding to each of the trained preset initial classification models; determine the maximum classification accuracy among the classification accuracies, and determine the preset initial classification model corresponding to the maximum classification accuracy as the preset classification model, wherein the preset classification model includes a preset random forest model; input each of the tourism resource monomers into each classification tree in the preset random forest model for classification, and obtain the classification results of each classification tree for the tourism resource monomer; based on the classification results corresponding to each classification tree, determine the resource category information to which each of the tourism resource monomers belongs.

[0057] Specifically, in the process of classification using a preset classification model, in order to improve the classification accuracy of the classification model, the classification model needs to be trained in advance. Based on this, the method includes: pre-constructing a preset initial classification model, the number of models can be multiple (wherein, the structures of multiple preset initial classification models can be the same or different), and then dividing the sample data (including sample tourism resource monomers and their corresponding actual category information) into training data and verification data. In the training data, the training data is divided into multiple groups of training sub-data according to the number of models, and the corresponding preset initial classification models are trained using the training sub-data. During the training process, the sample tourism resource monomer is used as input data and the actual category information is used as output data, thereby obtaining each trained preset initial classification model. Then, according to the number of models of the trained preset initial classification model, the verification data is divided into multiple groups of verification sub-data, and the corresponding trained preset initial classification models are verified using multiple groups of verification sub-data to obtain the classification accuracy corresponding to each trained preset initial classification model, and finally the model with the highest classification accuracy is selected as the preset classification model. At the same time, in the process of training the preset initial classification model using training data, the loss function corresponding to the model can also be constructed through the predicted classification information predicted by the preset initial classification model and the corresponding actual classification information. Finally, the preset initial classification model is continuously iteratively trained according to the loss function to obtain a preset classification model with higher classification accuracy.

[0058] Furthermore, the preset classification model can specifically be a random forest classification model, then at least one model structure in the preset initial classification model should be a random forest model. Before the training, the number of decision trees (classification trees) in the preset initial classification model, the decision tree depth, and the number of features used to split nodes in the decision tree are set, such as the number of decision trees is 100, the maximum depth is 5, and the number of features is 8. After the training is completed, the tourism resource monomers are input into each classification tree in the preset random forest model for classification, and the classification results of each classification tree for the tourism resource monomer are obtained. Then, based on the classification results corresponding to each classification tree, it is determined which classification result has the largest number of predictions, and finally the classification result with the largest number of predictions is determined as the resource category information to which the tourism resource monomer belongs.

[0059] Further, it is also necessary to determine the monomer level information corresponding to each tourism resource monomer. Based on this, step 204 specifically includes: determining the frequency of occurrence of each tourism resource monomer in the travel notes data, and using the frequency as the importance corresponding to each tourism resource monomer; performing semantic analysis on the travel notes data to which each tourism resource monomer belongs to obtain the tourists' satisfaction with each tourism resource monomer; determining the word frequency of each tourism resource monomer in each network, and based on the word frequency, determining the influence of each tourism resource monomer; determining the weight coefficients corresponding to the importance, satisfaction, and influence respectively; based on the weight coefficients, adding the importance, satisfaction, and influence corresponding to each tourism resource monomer to obtain the monomer level information corresponding to each tourism resource monomer; wherein, the method for determining the weight coefficients corresponding to the importance, satisfaction, and influence respectively includes: determining the importance standard deviation and the importance average value corresponding to the importance of each tourism resource monomer, and based on the importance standard deviation and the importance average value, standardizing the importance to obtain the standardized importance; determining the importance of each tourism resource monomer The satisfaction standard deviation and the satisfaction average of the satisfaction level corresponding to each tourism resource monomer are determined, and based on the satisfaction standard deviation and the satisfaction average, the satisfaction level is standardized to obtain the standardized satisfaction level; the influence standard deviation and the influence average of the influence level corresponding to each tourism resource monomer are determined, and based on the influence standard deviation and the influence average, the influence level is standardized to obtain the standardized influence level; based on the standardized importance, standardized satisfaction and standardized influence level corresponding to each tourism resource monomer, a resource classification decision matrix is ​​constructed, wherein each tourism resource monomer constitutes a row in the resource classification decision matrix, and the standardized importance, standardized satisfaction and standardized influence level constitute a column in the resource classification decision matrix; based on each element in the resource classification decision matrix, the entropy values ​​corresponding to the importance, satisfaction and influence level are determined respectively; based on the entropy values, the difference coefficients corresponding to the importance, satisfaction and influence level are determined respectively; based on the difference coefficients, the weight coefficients corresponding to the importance, satisfaction and influence level are determined respectively.

[0060] Specifically, determine the number of times each tourism resource monomer appears in the travel notes data, and determine the number as the importance of the corresponding tourism resource monomer. For example, if monomer A appears 60 times in the travel notes data, 60 is determined as the importance of monomer A. Then, perform semantic analysis on the travel notes data to which the tourism resource monomer belongs. Specifically, use the SnowNLP model to perform sentiment analysis on the travel notes data, judge the sentiment tendency of the comments containing the tourism resource monomer, and use the judgment result as the tourists' satisfaction with the tourism resource monomer. The score is between 0 and 1 to reflect the image perception of the tourism resource monomer in the minds of tourists. At the same time, measure the influence of the tourism resource monomer on the Internet through the popularity of the whole network search keywords. For example, crawl Baidu search engine to retrieve web pages covering the tourism resource monomer, determine the word frequency of the tourism resource monomer in each web page, and then calculate the influence of the tourism resource monomer according to the following formula:

[0061] influence=log 10 frequency

[0062] Among them, influence is the influence degree of tourism resources, and frequency is the word frequency corresponding to the tourism resource monomer. Therefore, the influence degree corresponding to each tourism resource monomer can be calculated through the above formula.

[0063] Furthermore, after calculating the importance, satisfaction and influence of each tourism resource unit, it is also necessary to determine the weight coefficients corresponding to the importance index, satisfaction index and influence index respectively. In order to improve the calculation accuracy of the weight coefficient, it is first necessary to standardize the data under each indicator. Specifically, the standardization can be performed according to the following formula:

[0064]

[0065] Among them, x i ′ represents the standardized data, i represents any one of the importance, satisfaction and influence, ui represents the mean of the data, for example, the mean of importance, the mean of satisfaction, the mean of influence, σ iRepresents the data standard deviation, for example, the standard deviation of importance, the standard deviation of satisfaction, and the standard deviation of influence, so that the standardized importance, standardized satisfaction, and standardized influence can be obtained according to the above formula. Further, the standardized data (standardized importance, standardized satisfaction, and standardized influence) are constructed into a resource classification decision matrix R, in which each row represents an evaluation object (tourism resource monomer), and each column represents an evaluation index (importance index, satisfaction index, and influence index). The standardized satisfaction, standardized importance, and standardized influence corresponding to each tourism resource are used as the elements in the resource classification decision matrix R, and then the entropy value of each index is determined according to the following formula:

[0066]

[0067] Among them, e j represents the entropy value of the jth indicator, n is the number of evaluation objects (tourism resource monomers), p ij is the element in the i-th row and j-th column of the decision matrix. According to the above formula, the entropy value of each evaluation index (satisfaction, importance, and influence) can be calculated. Furthermore, according to the entropy value of each evaluation index, the difference coefficient of each index is calculated according to the following formula:

[0068] g i =1-e j

[0069] Among them, g i It represents the difference coefficient of the jth indicator, that is, the difference coefficient corresponding to any one of the data of importance, satisfaction and influence. Further, according to the difference coefficient corresponding to each indicator, the weight coefficient corresponding to each indicator is determined according to the following formula:

[0070]

[0071] Among them, w j is the weight of the jth indicator, and m is the total number of indicators, i.e. 3.

[0072] Further, after respectively calculating the weight coefficients corresponding to the importance index, satisfaction index and influence index, the importance, satisfaction and influence are respectively multiplied by the corresponding weight coefficients, and the multiplication results are added, and the addition result is determined as the score of each tourism resource monomer, and the monomer level information corresponding to each tourism resource monomer is determined according to the score, for example, according to the addition result, the preset level configuration table is queried, wherein the preset level configuration table records the monomer level corresponding to each score interval, for example, if the score corresponding to monomer A is 89 points, by querying the preset level configuration table, it can be known that the monomer level corresponding to monomer A is B. Thus, through big data analysis and sentiment analysis technology, the embodiment of the present invention can deeply explore the behavior patterns and market trends of tourists, and make a more accurate and comprehensive evaluation of the value of tourism resource monomers.

[0073] 205. Based on geographic information data, determine the distance information between each tourism resource unit, and determine the spatial selectivity information of each tourism resource unit.

[0074] Corresponding to the embodiment of the present invention, in order to combine tourism resource monomers, it is also necessary to obtain the monomer distance information between each tourism resource monomer. Specifically, the monomer distance information between each tourism resource monomer can be determined through the location information of each tourism resource monomer displayed in the geographic information data.

[0075] Furthermore, in order to combine tourism resource monomers, it is also necessary to determine the spatial selectivity information of each tourism resource monomer. Based on this, step 205 specifically includes: based on the tourist trajectory data, determining the various driving routes of tourists in the target area; determining the target driving route that includes any tourism resource monomer in each of the tourism resource monomers among the driving routes, and determining the number of the target driving routes as the spatial selectivity information of any tourism resource monomer.

[0076] Specifically, for example, tourists form a total of 5000 driving routes in the target area, of which 2000 driving routes pass through resource monomer A, then 2000 is determined as the spatial selectivity information of resource monomer A.

[0077] 206. Based on the travel notes data, determine the correlation degree information between each tourism resource monomer, and perform clustering processing on each tourism resource monomer to obtain the cluster category information to which each tourism resource monomer belongs.

[0078] For the embodiment of the present invention, in order to combine tourism resource monomers, it is also necessary to determine the correlation degree information between each tourism resource monomer. Based on this, step 206 specifically includes: determining target travel record data containing at least two tourism resource monomers in each travel record data; determining a frequent item set for each tourism resource monomer appearing in the target travel record data, wherein each frequent item set contains a tourism resource monomer; determining the first frequency of the corresponding tourism resource monomer in each of the frequent item sets in each of the target travel record data, and determining the first frequency as a frequent support information of the frequent item set; determining a target frequent item set greater than a preset support threshold in the frequent support information corresponding to each frequent item set; determining two frequent item sets based on each of the target frequent item sets, wherein each of the two frequent item sets contains two tourism resource monomers; determining the second frequency of the corresponding two tourism resource monomers in each of the two frequent item sets appearing together in each of the target travel record data, and determining the second frequency as the two frequent item sets. The second frequent support information of each of the two frequent item sets is determined; in the second frequent support information corresponding to each of the two frequent item sets, a target two frequent item set greater than a preset support threshold is determined, and in the second frequent support information corresponding to each of the two frequent item sets, a removal two frequent item set less than or equal to the preset support threshold is determined, wherein each of the removed two frequent item sets contains a removal tourism resource monomer combination consisting of two tourism resource monomers; based on each of the target two frequent item sets, a tri-frequent item set is determined, wherein each of the tri-frequent item sets contains three tourism resource monomers, and each of the tri-frequent item sets does not include a removal tourism resource monomer combination; a third frequency of the corresponding three tourism resource monomers in each of the three frequent item sets appearing in each of the target travel diary data is determined, and the third frequency is determined as the tri-frequent support information of the tri-frequent item sets; based on the one frequent item set and its corresponding one frequent support information, the two frequent item sets and their corresponding two frequent support information, and the tri-frequent item sets and their corresponding three frequent support information, the association degree information between each of the tourism resource monomers is determined.

[0079] Specifically, the target travel diary data containing two or more tourism resource monomers are determined in each travel diary data, and then a frequent item set is determined for each tourism resource monomer appearing in the target travel diary data. For example, if each tourism resource monomer appearing in the target travel diary data includes A, B, C, D, E, and F, the target travel diary 1 contains four monomers ABCD, the target travel diary 2 contains four monomers ABCE, the target travel diary 3 contains four monomers BDEF, the target travel diary 4 contains four monomers BCDE, the target travel diary 5 contains four monomers ACDF, and the target travel diary 6 contains three monomers ABC, then the target travel diary 7 contains three monomers ABE, then a frequent item set is [A], [B], [C], [D], [E], [F]. The frequency of monomer A in the 7 target travel notes is 5 / 7 (i.e., a frequent support information), the frequency of monomer B in the 7 target travel notes is 6 / 7, the frequency of monomer C in the 7 target travel notes is 5 / 7, the frequency of monomer D in the 7 target travel notes is 4 / 7, the frequency of monomer E in the 7 target travel notes is 3 / 7, and the frequency of monomer F in the 7 target travel notes is 2 / 7. If the preset support threshold is 2.1 / 7, then a frequent item set less than the preset support threshold is The set is [F], and the target frequent item set greater than the preset support threshold is [A], [B], [C], [D], [E], then according to the target frequent item set, the second frequent item set is constructed as [AB], [AC], [AD], [AE], [BC], [BD], [BE], [CD], [CE], [DE]. The second frequency of monomer AB appearing in each target travel record data at the same time is 4 / 7 (i.e., the second frequent support information), the second frequency of monomer AC appearing in each target travel record data at the same time is 4 / 7, the second frequency of monomer AD appearing in each target travel record data at the same time is 2 / 7, and the second frequency of monomer The second frequency of AE appearing in each target travel record data at the same time is 2 / 7, the second frequency of BC appearing in each target travel record data at the same time is 4 / 7, the second frequency of BD appearing in each target travel record data at the same time is 3 / 7, the second frequency of BE appearing in each target travel record data at the same time is 3 / 7, the second frequency of CD appearing in each target travel record data at the same time is 3 / 7, the second frequency of CE appearing in each target travel record data at the same time is 2 / 7, and the second frequency of DE appearing in each target travel record data at the same time is 1 / 7. It can be seen that it is less than the preset support threshold 2.1 / 7 of the removed two frequent item sets are [AD], [AE], [CE], [DE]. Then, three frequent item sets are constructed from the target two frequent item sets that are greater than the preset support threshold. Among them, the removed two frequent item sets are not included in the three frequent item sets. The three frequent item sets obtained are: [ABC], [BCD]. The frequency of each tourism resource monomer in the three frequent item set [ABC] appearing in the target tourist data together is calculated, that is, the three frequent support information is 3 / 7. The frequency of each tourism resource monomer in the three frequent item set [BCD] appearing in the target tourist data together, that is, the three frequent support information is 2 / 7. The three frequent item sets greater than the preset support threshold are selected as [ABC]. In the example, the true subsets of [ABC] are [A], [B], [C], [AC], [AB], and [BC]. When A monomer appears, the probability that BC monomer also appears at the same time is the support information of [ABC] divided by the support information of [A], and the result is 3 / 5, that is, the association degree information between A monomer and BC monomer is 3 / 5. Similarly, the association degree information between BC monomer and A monomer is 3 / 4, the association degree information between B monomer and AC monomer is 1 / 2, the association degree information between AC monomer and B monomer is 3 / 4, the association degree information between C monomer and AB monomer is 3 / 5, and the association degree between AB monomer and C monomer is 3 / 4. .

[0080] Furthermore, it is necessary to perform clustering processing on each tourism resource monomer. Based on this, step 206 specifically includes: determining any tourism resource monomer among the tourism resource monomers, and counting the data density in the field corresponding to the any tourism resource monomer; if the data density is greater than a preset density threshold, determining the any tourism resource monomer as a cluster center; based on the cluster center, clustering the tourism resource monomers to obtain the cluster category information to which each tourism resource monomer belongs, wherein the tourism resource monomer that does not belong to any cluster category information is determined as an abnormal monomer.

[0081] Among them, the preset density threshold is set according to actual needs. Specifically, several definitions are first determined: core point (cluster center): a point containing more than MinPts number of tourism resource monomers within radius r; data density is the number of tourism resource monomers within radius r; boundary point (boundary tourism resource monomer): a point where the number of tourism resource monomers within radius r is less than MinPts, but falls within the neighborhood of the core point; noise point (abnormal tourism resource monomer): a point that is neither a core point nor a boundary point. The specific clustering method is: for each tourism resource monomer, the set of data in its neighborhood (within radius r) is calculated. When the number of data in the set exceeds the preset density threshold, the tourism resource monomer is determined as the core point (cluster center), and then the remaining points are checked to see if they are in the neighborhood of the core point. The remaining points in the neighborhood of the core point are classified into one category with the cluster center, and the remaining points not in the neighborhood of the core point are determined as abnormal monomers. In this way, each tourism resource monomer can be clustered through the above method to obtain the cluster category information to which each tourism resource monomer belongs.

[0082] 207. Based on resource category information, monomer level information, monomer distance information, spatial selectivity information, association degree information and cluster category information, a tourism resource knowledge graph corresponding to each tourism resource monomer is constructed.

[0083] Specifically, after obtaining the resource category information, monomer level information, monomer distance information, spatial selectivity information, association degree information and cluster category information of each tourism resource monomer, each tourism resource monomer is taken as a node, the monomer distance information between each tourism resource monomer is taken as an edge, the resource category information is taken as the node label of the corresponding node, and the monomer level information, spatial selectivity information, association degree information, cluster category information is taken as the attribute information of the corresponding node, to construct a tourism resource knowledge graph corresponding to each tourism resource monomer.

[0084] 208. Based on the tourism resource knowledge graph, determine the combination plan of tourism resource units in the target area.

[0085] In the above manner, a knowledge graph of tourism resources in different regions can be constructed, and the knowledge graph of tourism resources can be stored in a search system. When a user wants to combine tourism resource monomers in a certain region, the user inputs the demand information in the search system. In order to deal with the uncertainty and ambiguity in the user input, the embodiment of the present invention can adopt a fuzzy matching algorithm, which can identify and parse approximate query conditions, and convert the user's demand information into precise parameters that can be used to search and filter the tourism resource database. The fuzzy matching module includes a semantic analyzer, which uses a predefined tourism resource ontology library and association rules to understand the user's intention, and associates it with the corresponding entities and attributes in the knowledge graph, thereby realizing the combination processing of tourism resource monomers. At the same time, the user can require the system to output the calculated classification, evaluation, clustering, selectivity, and association rules respectively, or query the attribute relationship of tourism resource monomers through the knowledge graph constructed by Neo4j, which can be directly used for regional resource combination. Alternatively, you can also use the system's built-in multi-objective optimization algorithm to input constraints (e.g., highly rated tourism resource monomers, rich types, strong spatial agglomeration, high concentration and dispersion, strong association rules correspond to a set of monomers with high scores, rich monomer categories, few clustering labels, and high association confidence) to obtain multiple solutions with different trade-offs. Then, the solution set of the tourism resource monomer combination that balances the constraints is displayed on a real map, and the relationships and attributes between them are represented by lines and graphs. In this way, the combination of tourism resource monomers through a comprehensive knowledge graph can improve the combination efficiency and combination accuracy of tourism resource monomers.

[0086] According to another method for combining tourism resource monomers provided by the present invention, compared with the current method of combining tourism resource monomers by manual means, the present invention obtains resource description data in a target area, wherein the resource description data includes: travel notes data, geographic information data, and tourist trajectory data; and performs word segmentation processing on the resource description data to obtain each word segmentation corresponding to the resource description data; and inputs each word segmentation into a preset entity recognition model for entity recognition to obtain each tourism resource monomer contained in the resource description data; at the same time, classifies each tourism resource monomer to obtain resource category information to which each tourism resource monomer belongs, and performs graded processing on each tourism resource monomer to obtain each tourism resource monomer. The monomer level information corresponding to the tourism resource monomer; and based on the geographic information data, determine the monomer distance information between each of the tourism resource monomers, and determine the spatial selectivity information of each of the tourism resource monomers; at the same time, based on the travel data, determine the correlation degree information between each of the tourism resource monomers, and cluster each of the tourism resource monomers to obtain the cluster category information to which each of the tourism resource monomers belongs; then, based on the resource category information, monomer level information, monomer distance information, spatial selectivity information, correlation degree information and cluster category information, construct a tourism resource knowledge graph corresponding to each of the tourism resource monomers; finally, based on the tourism resource knowledge graph, determine the combination plan of tourism resource monomers in the target area. Therefore, by identifying each tourism resource monomer contained in the resource description data, and determining the resource category information, monomer level information, monomer distance information, spatial selectivity information, correlation degree information and clustering category information of each tourism resource monomer, a tourism resource knowledge graph is constructed according to the above-mentioned information, and finally the tourism resource monomers in the target area are combined according to the tourism resource knowledge graph, that is, by determining various attribute information of tourism resource monomers and constructing a knowledge graph based on various attribute information, monomers are combined through a knowledge graph containing various attribute information, which can improve the combination efficiency and combination accuracy of tourism resource monomers.

[0087] Further, as Figure 1 The specific implementation of the present invention provides a combination device of tourism resource monomers, such as Figure 3 As shown, the device includes: an acquisition unit 31, an entity identification unit 32, a processing unit 33, a first determination unit 34, a second determination unit 35, a construction unit 36 ​​and a third determination unit 37.

[0088] The acquisition unit 31 may be used to acquire resource description data in a target area, wherein the resource description data includes travel notes data, geographic information data, and tourist trajectory data.

[0089] The entity identification unit 32 may be used to perform tourism resource entity identification on the resource description data to obtain each tourism resource entity contained in the resource description data.

[0090] The processing unit 33 can be used to classify each of the tourism resource monomers to obtain resource category information to which each of the tourism resource monomers belongs, and to grade each of the tourism resource monomers to obtain monomer level information corresponding to each of the tourism resource monomers.

[0091] The first determination unit 34 may be used to determine the monomer distance information between the tourism resource monomers and the spatial selectivity information of the tourism resource monomers based on the geographic information data.

[0092] The second determination unit 35 may be used to determine the association degree information between the tourism resource monomers based on the travel note data, and perform clustering processing on the tourism resource monomers to obtain the cluster category information to which the tourism resource monomers belong.

[0093] The construction unit 36 ​​can be used to construct a tourism resource knowledge graph corresponding to each of the tourism resource monomers based on the resource category information, monomer level information, monomer distance information, spatial selectivity information, association degree information and cluster category information, wherein each of the tourism resource monomers serves as a node in the tourism resource knowledge graph, the monomer distance information serves as an edge in the tourism resource knowledge graph, the resource category information serves as a node label of the corresponding node in the tourism resource knowledge graph, and the monomer level information, spatial selectivity information, association degree information and cluster category information serve as attribute information of the corresponding node in the tourism resource knowledge graph.

[0094] The third determination unit 37 may be configured to determine a combination scheme of tourism resource monomers in the target area based on the tourism resource knowledge graph.

[0095] In specific application scenarios, such as Figure 4 As shown, in order to determine each tourism resource entity included in the resource description data, the entity recognition unit 32 includes a word segmentation processing module 321 and an entity recognition module 322.

[0096] The word segmentation processing module 321 can be used to perform word segmentation processing on the resource description data to obtain each word segment corresponding to the resource description data.

[0097] The entity recognition module 322 can be used to input each word segment into a preset entity recognition model for entity recognition, and obtain each tourism resource monomer included in the resource description data.

[0098] In a specific application scenario, in order to determine the resource category information to which a tourism resource unit belongs, the processing unit 33 includes a construction module 331, a division module 332, a training module 333, a testing module 334, a first determination module 335, and a classification module 336.

[0099] The construction module 331 can be used to construct multiple preset initial classification models and obtain sample tourism resource monomers and their corresponding actual category information.

[0100] The division module 332 may be used to divide the sample tourism resource monomers and their corresponding actual category information into multiple groups of training data and multiple groups of test data based on the model number of the preset initial classification model.

[0101] The training module 333 can be used to train the corresponding preset initial classification model using each group of training data to obtain the trained preset initial classification model, wherein the sample tourism resource monomer is used as input data and the actual category information is used as output data during the training process.

[0102] The testing module 334 can be used to test the corresponding trained preset initial classification model using each group of test data to obtain the classification accuracy corresponding to each trained preset initial classification model.

[0103] The first determination module 335 can be used to determine the maximum classification accuracy among the classification accuracies, and determine the preset initial classification model corresponding to the maximum classification accuracy as the preset classification model, wherein the preset classification model includes a preset random forest model.

[0104] The classification module 336 can be used to input each of the tourism resource monomers into each classification tree in the preset random forest model for classification, and obtain the classification results of each classification tree for the tourism resource monomer.

[0105] The first determination module 335 may be used to determine the resource category information to which each tourism resource monomer belongs based on the classification results corresponding to each classification tree.

[0106] In a specific application scenario, in order to perform hierarchical processing on each tourism resource entity, the processing unit 33 further includes a semantic analysis module 337 and an addition module 338 .

[0107] The first determination module 335 may also be used to determine the frequency of occurrence of each tourism resource monomer in the travel diary data, and use the frequency as the importance corresponding to each tourism resource monomer.

[0108] The semantic analysis module 337 can also be used to perform semantic analysis on the travel notes data belonging to each of the tourism resource entities to obtain the tourists' satisfaction with each of the tourism resource entities.

[0109] The first determination module 335 may also be used to determine the word frequency of each tourism resource monomer in each network, and determine the influence degree of each tourism resource monomer based on the word frequency.

[0110] The first determination module 335 may also be used to determine weight coefficients corresponding to the importance, satisfaction, and influence, respectively.

[0111] The adding module 338 may be used to add the importance, satisfaction and influence corresponding to each of the tourism resource monomers based on the weight coefficient to obtain the monomer level information corresponding to each of the tourism resource monomers.

[0112] Among them, in order to determine the weight coefficient, the first determination module 335 can be specifically used to determine the importance standard deviation and the importance average value corresponding to each of the tourism resource monomers, and based on the importance standard deviation and the importance average value, standardize the importance to obtain the standardized importance; determine the satisfaction standard deviation and the satisfaction average value of the satisfaction level corresponding to each of the tourism resource monomers, and based on the satisfaction standard deviation and the satisfaction average value, standardize the satisfaction to obtain the standardized satisfaction; determine the influence standard deviation and the influence average value of the influence level corresponding to each of the tourism resource monomers, and based on the influence standard deviation and the influence average value, standardize the influence level. The degree of influence is standardized to obtain the standardized degree of influence; based on the standardized importance, standardized satisfaction and standardized degree of influence corresponding to each tourism resource monomer, a resource classification decision matrix is ​​constructed, wherein each tourism resource monomer constitutes a row in the resource classification decision matrix, and the standardized importance, standardized satisfaction and standardized degree of influence constitute a column in the resource classification decision matrix; based on each element in the resource classification decision matrix, the entropy values ​​corresponding to the importance, satisfaction and influence are determined respectively; based on the entropy values, the difference coefficients corresponding to the importance, satisfaction and influence are determined respectively; based on the difference coefficients, the weight coefficients corresponding to the importance, satisfaction and influence are determined respectively.

[0113] In a specific application scenario, in order to determine the spatial selectivity information of each tourism resource unit, the first determination unit 34 can be specifically used to determine the various driving routes of tourists in the target area based on the tourist trajectory data; determine the target driving route that includes any tourism resource unit in each tourism resource unit among the various driving routes, and determine the number of the target driving routes as the spatial selectivity information of any tourism resource unit.

[0114] In a specific application scenario, in order to determine the association degree information between each tourism resource monomer, the second determination unit 35 can be specifically used to determine the target travel record data containing at least two tourism resource monomers in each travel record data; determine a frequent item set for each tourism resource monomer appearing in the target travel record data, wherein each frequent item set contains a tourism resource monomer; determine the first frequency of the corresponding tourism resource monomer in each of the frequent item sets in each of the target travel record data, and determine the first frequency as the frequent support information of the frequent item set; determine a target frequent item set greater than a preset support threshold in the frequent support information corresponding to each frequent item set; based on each of the target frequent item sets, determine two frequent item sets, wherein each of the two frequent item sets contains two tourism resource monomers; determine the second frequency of the corresponding two tourism resource monomers in each of the two frequent item sets appearing together in each of the target travel record data, and determine the second frequency as the two frequent support information of the two frequent item sets. degree information; determining target two frequent item sets greater than a preset support threshold in the two frequent support information corresponding to each of the two frequent item sets, and determining eliminated two frequent item sets less than or equal to the preset support threshold in the two frequent support information corresponding to each of the two frequent item sets, wherein each eliminated two frequent item set includes an eliminated tourism resource monomer combination consisting of two tourism resource monomers; based on each of the target two frequent item sets, determining three frequent item sets, wherein each three frequent item set includes three tourism resource monomers, and each three frequent item set does not include an eliminated tourism resource monomer combination; determining the third frequency of the corresponding three tourism resource monomers in each of the three frequent item sets appearing in each of the target travel diary data, and determining the third frequency as the three frequent support information of the three frequent item sets; based on the one frequent item set and its corresponding one frequent support information, the two frequent item sets and their corresponding two frequent support information, and the three frequent item sets and their corresponding three frequent support information, determining the association degree information between each of the tourism resource monomers.

[0115] In a specific application scenario, in order to perform clustering processing on each tourism resource monomer, the second determination unit 35 includes a second determination module 351 and a clustering module 352 .

[0116] The second determination module 351 may be used to determine any tourism resource monomer among the tourism resource monomers, and to count the data density in the field corresponding to the any tourism resource monomer.

[0117] The second determination module 351 may also be used to determine any tourism resource monomer as a cluster center if the data density is greater than a preset density threshold.

[0118] The clustering module 352 can be used to cluster the tourism resource monomers based on the cluster centers to obtain cluster category information to which the tourism resource monomers belong, wherein tourism resource monomers that do not belong to any cluster category information are determined as abnormal monomers.

[0119] It should be noted that for other corresponding descriptions of the functional modules involved in the combined device of a tourism resource monomer provided in the embodiment of the present invention, reference can be made to Figure 1 The corresponding description of the method shown will not be repeated here.

[0120] Based on the above Figure 1 The method shown, accordingly, the embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the following steps are implemented: obtaining resource description data in the target area, wherein the resource description data includes: travel notes data, geographic information data, and tourist trajectory data; performing tourism resource entity recognition on the resource description data to obtain each tourism resource monomer contained in the resource description data; classifying each tourism resource monomer to obtain resource category information to which each tourism resource monomer belongs, and grading each tourism resource monomer to obtain monomer level information corresponding to each tourism resource monomer; based on the geographic information data, determining the monomer distance information between each tourism resource monomer, and determining the spatial selectivity information of each tourism resource monomer; based on the travel notes data, determining The information on the degree of association between each of the tourism resource monomers is obtained, and each of the tourism resource monomers is clustered to obtain the cluster category information to which each of the tourism resource monomers belongs; based on the resource category information, monomer level information, monomer distance information, spatial selectivity information, association degree information and cluster category information, a tourism resource knowledge graph corresponding to each of the tourism resource monomers is constructed, wherein each of the tourism resource monomers is used as a node in the tourism resource knowledge graph, the monomer distance information is used as an edge in the tourism resource knowledge graph, the resource category information is used as a node label of the corresponding node in the tourism resource knowledge graph, and the monomer level information, spatial selectivity information, association degree information and cluster category information are used as attribute information of the corresponding node in the tourism resource knowledge graph; based on the tourism resource knowledge graph, a combination scheme of tourism resource monomers in the target area is determined.

[0121] Based on the above Figure 1 The method shown and Figure 3 The embodiment of the device shown in the figure, the embodiment of the present invention also provides a physical structure diagram of a computer device, such as Figure 5 As shown, the computer device includes: a processor 41, a memory 42, and a computer program stored in the memory 42 and executable on the processor, wherein the memory 42 and the processor 41 are both arranged on a bus 43, and the processor 41 implements the following steps when executing the program: obtaining resource description data within the target area, wherein the resource description data includes: travel notes data, geographic information data, and tourist trajectory data; performing tourism resource entity recognition on the resource description data to obtain each tourism resource monomer contained in the resource description data; classifying each tourism resource monomer to obtain resource category information to which each tourism resource monomer belongs, and grading each tourism resource monomer to obtain monomer level information corresponding to each tourism resource monomer; determining monomer distance information between each tourism resource monomer based on the geographic information data, and determining the spatial selectivity information; based on the travel data, determine the correlation degree information between each of the tourism resource monomers, and cluster each of the tourism resource monomers to obtain the cluster category information to which each of the tourism resource monomers belongs; based on the resource category information, monomer level information, monomer distance information, spatial selectivity information, correlation degree information and cluster category information, construct a tourism resource knowledge graph corresponding to each of the tourism resource monomers, wherein each of the tourism resource monomers is used as a node in the tourism resource knowledge graph, the monomer distance information is used as an edge in the tourism resource knowledge graph, the resource category information is used as a node label of the corresponding node in the tourism resource knowledge graph, and the monomer level information, spatial selectivity information, correlation degree information, and cluster category information are used as attribute information of the corresponding node in the tourism resource knowledge graph; based on the tourism resource knowledge graph, determine the combination scheme of tourism resource monomers in the target area.

[0122] Through the technical solution of the present invention, the present invention obtains resource description data in the target area, wherein the resource description data includes: travel notes data, geographic information data, and tourist trajectory data; and performs tourism resource entity recognition on the resource description data to obtain each tourism resource monomer contained in the resource description data; at the same time, each tourism resource monomer is classified to obtain resource category information to which each tourism resource monomer belongs, and each tourism resource monomer is graded to obtain monomer level information corresponding to each tourism resource monomer; and based on the geographic information data, the monomer distance information between each tourism resource monomer is determined, as well as the spatial selectivity information of each tourism resource monomer is determined; at the same time, based on the travel notes data, the correlation degree information between each tourism resource monomer is determined , and cluster the tourism resource monomers to obtain the cluster category information to which the tourism resource monomers belong; then, based on the resource category information, monomer level information, monomer distance information, spatial selectivity information, association degree information and cluster category information, a tourism resource knowledge graph corresponding to each of the tourism resource monomers is constructed, wherein each of the tourism resource monomers is used as a node in the tourism resource knowledge graph, the monomer distance information is used as an edge in the tourism resource knowledge graph, the resource category information is used as a node label of the corresponding node in the tourism resource knowledge graph, and the monomer level information, spatial selectivity information, association degree information and cluster category information are used as attribute information of the corresponding node in the tourism resource knowledge graph; finally, based on the tourism resource knowledge graph, a combination scheme of tourism resource monomers in the target area is determined. Therefore, by identifying each tourism resource monomer contained in the resource description data, and determining the resource category information, monomer level information, monomer distance information, spatial selectivity information, correlation degree information and clustering category information of each tourism resource monomer, a tourism resource knowledge graph is constructed according to the above-mentioned information, and finally the tourism resource monomers in the target area are combined according to the tourism resource knowledge graph, that is, by determining various attribute information of tourism resource monomers and constructing a knowledge graph based on various attribute information, monomers are combined through a knowledge graph containing various attribute information, which can improve the combination efficiency and combination accuracy of tourism resource monomers.

[0123] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0124] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for combining tourism resource monomers, It is characterized in that include: Obtain resource description data in the target area, wherein the resource description data includes: travel notes data, geographic information data, and tourist trajectory data; Performing tourism resource entity recognition on the resource description data to obtain each tourism resource entity contained in the resource description data; Classify each of the tourism resource monomers to obtain resource category information to which each of the tourism resource monomers belongs, and grade each of the tourism resource monomers to obtain monomer level information corresponding to each of the tourism resource monomers; Based on the geographic information data, determining the monomer distance information between each of the tourism resource monomers, and determining the spatial selectivity information of each of the tourism resource monomers; Determine target travel record data including at least two tourism resource monomers in each travel record data; For each tourism resource monomer appearing in the target travel diary data, a frequent item set is determined, wherein each frequent item set contains a tourism resource monomer; Determine a first frequency of occurrence of a corresponding tourism resource monomer in each of the frequent item sets in each of the target travel diary data, and determine the first frequency as a frequent support information of the frequent item set; Determine a target frequent item set greater than a preset support threshold value in the frequent support information corresponding to each frequent item set; Based on each of the target frequent item sets, two frequent item sets are determined, wherein each of the two frequent item sets contains two tourism resource monomers; Determine the second frequency of the two tourism resource monomers in each of the two frequent item sets appearing together in each of the target travel diary data, and determine the second frequency as the second frequent support information of the two frequent item sets; Determine the target two frequent item sets greater than the preset support threshold in the two frequent support information corresponding to each of the two frequent item sets, and determine the eliminated two frequent item sets less than or equal to the preset support threshold in the two frequent support information corresponding to each of the two frequent item sets, wherein each eliminated two frequent item set contains an eliminated tourism resource monomer combination consisting of two tourism resource monomers; Based on each of the target two frequent item sets, three frequent item sets are determined, wherein each of the three frequent item sets contains three tourism resource monomers, and each of the three frequent item sets does not include a combination of tourism resource monomers that are eliminated; Determine the third frequency of the three tourism resource monomers in each of the three frequent item sets appearing in each of the target travel diary data, and determine the third frequency as the three frequent support information of the three frequent item sets; Determine the association degree information between the tourism resource monomers based on the one frequent item set and its corresponding one frequent support information, the two frequent item sets and its corresponding two frequent support information, and the three frequent item sets and its corresponding three frequent support information; Performing clustering processing on each of the tourism resource monomers to obtain cluster category information to which each of the tourism resource monomers belongs; Based on the resource category information, monomer level information, monomer distance information, spatial selectivity information, association degree information and cluster category information, a tourism resource knowledge graph corresponding to each of the tourism resource monomers is constructed, wherein each of the tourism resource monomers is used as a node in the tourism resource knowledge graph, the monomer distance information is used as an edge in the tourism resource knowledge graph, the resource category information is used as a node label of the corresponding node in the tourism resource knowledge graph, and the monomer level information, spatial selectivity information, association degree information and cluster category information are used as attribute information of the corresponding node in the tourism resource knowledge graph; Based on the tourism resource knowledge graph, a combination scheme of tourism resource monomers in the target area is determined.

2. The method according to claim 1, It is characterized in that The performing tourism resource entity identification on the resource description data to obtain each tourism resource entity contained in the resource description data includes: Performing word segmentation processing on the resource description data to obtain each word segmentation corresponding to the resource description data; The individual word segments are input into a preset entity recognition model for entity recognition, so as to obtain the individual tourism resource entities contained in the resource description data.

3. The method according to claim 1, It is characterized in that The classification process of each tourism resource monomer to obtain resource category information to which each tourism resource monomer belongs includes: Construct multiple preset initial classification models and obtain sample tourism resource monomers and their corresponding actual category information; Based on the model quantity of the preset initial classification model, the sample tourism resource monomers and their corresponding actual category information are divided into multiple groups of training data and multiple groups of test data; Using each group of training data to train the corresponding preset initial classification model, a trained preset initial classification model is obtained, wherein during the training process, the sample tourism resource monomer is used as input data, and the actual category information is used as output data; Using each group of test data to test the corresponding trained preset initial classification model, to obtain the classification accuracy corresponding to each trained preset initial classification model; Determining a maximum classification accuracy among the classification accuracies, and determining a preset initial classification model corresponding to the maximum classification accuracy as a preset classification model, wherein the preset classification model includes a preset random forest model; Input each of the tourism resource monomers into each classification tree in the preset random forest model for classification, and obtain the classification results of each classification tree for the tourism resource monomer; Based on the classification results corresponding to each classification tree, the resource category information to which each tourism resource monomer belongs is determined.

4. The method according to claim 1, It is characterized in that The step of performing classification processing on each of the tourism resource monomers to obtain monomer level information corresponding to each of the tourism resource monomers includes: Determine the frequency of occurrence of each of the tourism resource monomers in the travel diary data, and use the frequency as the importance corresponding to each of the tourism resource monomers; Performing semantic analysis on the travel notes data belonging to each of the tourism resource monomers to obtain the tourists' satisfaction with each of the tourism resource monomers; Determine the word frequency of each tourism resource monomer in each network, and determine the influence degree of each tourism resource monomer based on the word frequency; Determine the weight coefficients corresponding to the importance, satisfaction, and influence respectively; Based on the weight coefficient, the importance, satisfaction and influence corresponding to each tourism resource monomer are added together to obtain monomer level information corresponding to each tourism resource monomer; The determination of the weight coefficients corresponding to the importance, satisfaction and influence respectively includes: Determine the importance standard deviation and the importance average of the importance corresponding to each tourism resource monomer, and based on the importance standard deviation and the importance average, perform standardization on the importance to obtain a standardized importance; Determine the satisfaction degree standard deviation and the satisfaction degree average value of the satisfaction degree corresponding to each of the tourism resource monomers, and standardize the satisfaction degree based on the satisfaction degree standard deviation and the satisfaction degree average value to obtain the standardized satisfaction degree; Determine the standard deviation of the impact degree and the average impact degree of the impact degree corresponding to each of the tourism resource monomers, and based on the standard deviation of the impact degree and the average impact degree, perform standardization processing on the impact degree to obtain a standardized impact degree; Based on the standardized importance, standardized satisfaction and standardized influence corresponding to each tourism resource monomer, a resource classification decision matrix is ​​constructed, wherein each tourism resource monomer constitutes a row in the resource classification decision matrix, and the standardized importance, standardized satisfaction and standardized influence constitute a column in the resource classification decision matrix; Based on each element in the resource classification decision matrix, the entropy values ​​corresponding to the importance, satisfaction and influence are determined respectively; Based on the entropy value, the difference coefficients corresponding to the importance, satisfaction and influence are determined respectively; Based on the difference coefficient, the weight coefficients corresponding to the importance, satisfaction and influence are determined respectively.

5. The method according to claim 1, It is characterized in that The determining of the spatial selectivity information of each tourism resource monomer includes: Based on the tourist trajectory data, determining various travel routes of tourists in the target area; A target driving route including any tourism resource monomer among the tourism resource monomers is determined in each driving route, and the number of the target driving routes is determined as the spatial selectivity information of any tourism resource monomer.

6. The method according to claim 1, It is characterized in that The clustering process is performed on each of the tourism resource monomers to obtain the cluster category information to which each of the tourism resource monomers belongs, including: Determine any tourism resource monomer among the tourism resource monomers, and count the data density in the field corresponding to the any tourism resource monomer; If the data density is greater than a preset density threshold, any tourism resource monomer is determined as a cluster center; Based on the cluster center, the tourism resource monomers are clustered to obtain the cluster category information to which the tourism resource monomers belong, wherein the tourism resource monomers that do not belong to any cluster category information are determined as abnormal monomers.

7. A combination device of tourism resource monomers, It is characterized in that include: An acquisition unit, used to acquire resource description data in a target area, wherein the resource description data includes: travel notes data, geographic information data, and tourist trajectory data; An entity recognition unit, used to perform tourism resource entity recognition on the resource description data to obtain each tourism resource entity contained in the resource description data; A processing unit, used to classify each of the tourism resource monomers to obtain resource category information to which each of the tourism resource monomers belongs, and to grade each of the tourism resource monomers to obtain monomer level information corresponding to each of the tourism resource monomers; A first determining unit is used to determine the monomer distance information between each of the tourism resource monomers and the spatial selectivity information of each of the tourism resource monomers based on the geographic information data; The second determination unit is used to determine target travel diary data containing at least two tourism resource monomers in each travel diary data; determine a frequent item set for each tourism resource monomer appearing in the target travel diary data, wherein each frequent item set contains a tourism resource monomer; determine the first frequency of the corresponding tourism resource monomer in each frequent item set appearing in each target travel diary data, and determine the first frequency as a frequent support information of the frequent item set; determine a target frequent item set greater than a preset support threshold in the frequent support information corresponding to each frequent item set; determine two frequent item sets based on each target frequent item set, wherein each two frequent item sets contains two tourism resource monomers; determine the second frequency of the corresponding two tourism resource monomers in each two frequent item sets appearing in each target travel diary data, and determine the second frequency as the two frequent support information of the two frequent item sets; determine a target frequent item set greater than a preset support threshold in the two frequent support information corresponding to each two frequent item sets Two frequent item sets, and determine the two frequent item sets that are less than or equal to the preset support threshold in the two frequent support information corresponding to each of the two frequent item sets, wherein each of the two frequent item sets contains a combination of tourism resource monomers consisting of two tourism resource monomers; based on each of the target two frequent item sets, determine three frequent item sets, wherein each of the three frequent item sets contains three tourism resource monomers, and each of the three frequent item sets does not include the combination of tourism resource monomers; determine the third frequency of the corresponding three tourism resource monomers in each of the three frequent item sets appearing in each of the target travel notes data, and determine the third frequency as the three frequent support information of the three frequent item sets; based on the one frequent item set and its corresponding one frequent support information, the two frequent item sets and their corresponding two frequent support information, and the three frequent item sets and their corresponding three frequent support information, determine the correlation degree information between each of the tourism resource monomers; and perform clustering processing on each of the tourism resource monomers to obtain the cluster category information to which each of the tourism resource monomers belongs; A construction unit, for constructing a tourism resource knowledge graph corresponding to each of the tourism resource monomers based on the resource category information, monomer level information, monomer distance information, spatial selectivity information, association degree information and cluster category information, wherein each of the tourism resource monomers is used as a node in the tourism resource knowledge graph, the monomer distance information is used as an edge in the tourism resource knowledge graph, the resource category information is used as a node label of the corresponding node in the tourism resource knowledge graph, and the monomer level information, spatial selectivity information, association degree information and cluster category information are used as attribute information of the corresponding node in the tourism resource knowledge graph; The third determining unit is used to determine a combination scheme of tourism resource monomers in the target area based on the tourism resource knowledge graph.

8. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images