Method for constructing network relationship of tourism scenic spots in combination with large model

Abstract

The invention discloses a method for constructing a text travel scenic spot network relationship in combination with a large model, and relates to the technical fields of urban planning, tourism planning, natural language processing, large models and the like, and the method comprises the following steps: firstly, constructing a scenic spot geographic entity object of a text travel scenic spot network; secondly, text comments of the travel scenic spot network are collected, and each text comment corresponds to a unique scenic spot geographic entity object; then, the names of all the scenic spot geographic entity objects are prompted to the context of the large model, and the semantic network relation between the scenic spot geographic entity objects is extracted based on the large model; and finally, based on the semantic network relationship between the extracted scenic spot geographic entity objects, constructing different types of semantic network relationship maps between the scenic spot geographic entity objects. According to the method, comments can be interpreted through a large model, and a relation network between scenic spot geographic entity objects is extracted.

Description

Technical Field

[0001] The present invention relates to the fields of urban planning, tourism planning, urban transportation and complex network modeling technology, and in particular to a method for constructing a network relationship of cultural and tourism attractions in combination with a large model. Background Art

[0002] The network relationship of cultural and tourism attractions is of great significance for applications in urban planning, tourism planning, urban transportation and other fields, and is an important reference for cultural and tourism route planning. Therefore, guiding cultural and tourism development planning and practice requires the help of the network relationship of cultural and tourism attractions. The existing technology mainly adopts the co-occurrence relationship and complex network technology between the geographical entity objects of cultural and tourism attractions, which leads to the lack of complex relationships between the geographical entity objects of cultural and tourism attractions, such as whether there is a recommended tourism relationship between the geographical entity objects of tourist attractions, whether there is a close relationship in geographical space, etc. There are still certain problems in the identification of these relationships, such as poor accuracy. To this end, it is necessary to better solve this problem through technological innovation.

[0003] The present invention considers using a large model to solve the above-mentioned problems. At present, large model technology can realize intelligent interpretation of text semantics through methods such as prompt words, and give relatively standard answers. There are many schemes for constructing these prompt words, which need to be innovatively designed according to different application fields. In general, the innovative application of large models in professional fields using prompt word technology can significantly improve the accuracy of dialogue answers and the data standardization level of results. However, large model analysis also needs to reduce the ambiguity and incompleteness of large model calculation results through a set of complete process and method innovation designs. Summary of the invention

[0004] The technical problem to be solved by the present invention is to overcome the deficiencies of the prior art and provide a method for constructing a network relationship of cultural and tourism attractions combined with a large model. The present invention can interpret comments through a large model and extract a connection relationship network between geographical entity objects of the scenic spot.

[0005] The present invention adopts the following technical solutions to solve the above technical problems:

[0006] In a first aspect, a method for constructing a network relationship of cultural and tourism attractions combined with a large model according to the present invention includes:

[0007] Step 1: Construct the geographical entity object of the cultural and tourism scenic spot network;

[0008] Step 2: Collect text reviews of the cultural and tourism scenic spots network, and each text review corresponds to a unique scenic spot geographic entity object;

[0009] Step 3, prompt the names of all scenic spot geographic entity objects to the big model;

[0010] Step 4: Extracting semantic network relationships between scenic spot geographic entity objects based on the big model;

[0011] Step 5: Based on the semantic network relationships between the extracted scenic spot geographic entity objects, construct different types of semantic network relationship maps between the scenic spot geographic entity objects.

[0012] As a further optimization scheme of the method for constructing a network relationship of cultural and tourism attractions combined with a large model described in the present invention, step 1 is specifically as follows:

[0013] Step 1.1, crawl out the geographical entity objects of tourist attractions from professional tourist attraction websites;

[0014] Step 1.2: Each scenic spot geographic entity object has a unique name.

[0015] As a further optimization scheme of the method for constructing a network relationship of cultural and tourism attractions combined with a large model described in the present invention, step 3 is specifically as follows:

[0016] Step 3.1, traverse the names of scenic spot geographic entity objects;

[0017] Step 3.2: construct a large model prompt word, and confirm the name of the traversed scenic spot geographic entity object to the large model as the name of the scenic spot geographic entity object.

[0018] As a further optimization scheme of the method for constructing a network relationship of cultural and tourism attractions combined with a large model described in the present invention, step 4 is specifically as follows:

[0019] Step 4.1, construct an entity class of the semantic network relationship to be extracted, and the entity class includes 4 attributes, namely name1, name2, rel1, and rel1a, where name1 represents the name of a scenic spot geographical entity, name2 represents the name of another scenic spot geographical entity, rel1 is a Boolean type data used to indicate whether there is an association relationship between scenic spot geographical entities, and rel1a represents the association relationship between scenic spot geographical entities;

[0020] Step 4.2, register the entity class as a parser for json format output;

[0021] Step 4.3, construct a prompt word template, and embed the output parser in step 4.2 into the prompt word template;

[0022] Step 4.4, traverse each text comment of all text comments, and use the large model to extract the name set of the scenic spot geographic entity object of each text comment; at the same time, if the name set of the extracted scenic spot geographic entity object does not include the name of the scenic spot geographic entity object to which the text comment belongs, then add the name of the scenic spot geographic entity object to which the text comment belongs to the name set of the scenic spot geographic entity object; obtain the final name set t of the scenic spot geographic entity object;

[0023] Step 4.5, further adjust the input value of the prompt word template in real time, the input value includes any two different names of scenic spot geographical entity objects in the name set t, input any two different names of scenic spot geographical entity objects in the name set t into the prompt word template, call the big model according to the prompt word template to obtain the big model return result, and extract the semantic network relationship between the scenic spot geographical entity objects from the return result.

[0024] As a further optimization scheme of the method for constructing a network relationship of cultural and tourism attractions combined with a large model described in the present invention, the value of rel1 is True or False.

[0025] As a further optimization scheme of the method for constructing a network relationship of cultural and tourism attractions combined with a large model described in the present invention, step 5 is specifically as follows:

[0026] Step 5.1, after completing the traversal processing of all text comments in step 4.4 and step 4.5, collect the json format results obtained by the large model calculation;

[0027] Step 5.2: Perform complex network modeling on all json format results in step 5.1:

[0028] First, complex network modeling uses the names of all the scenic spots’ geographical entity objects in the cultural and tourism scenic spots network as nodes of the complex network;

[0029] Secondly, when the rel1 value in each json format result is True, the name1 value and name2 value in the json format result are two network nodes of a complex network edge connection relationship, and the name1 value and name2 value form a complex network edge connection relationship; according to the complex network nodes and complex network edge connection relationships, a complex network model is formed, and the complex network model is also the semantic network relationship map between scenic spot geographic entity objects;

[0030] Finally, based on the different types of semantic network relationship maps between scenic spot geographic entity objects, the rel1a median is clustered, and different types of complex network edge link relationships are formed according to the clustering results. For each complex network edge link relationship, different types of semantic network relationship maps between scenic spot geographic entity objects are constructed.

[0031] As a further optimization scheme of the method for constructing network relationships of cultural and tourism attractions combined with a large model described in the present invention, in step 4.1, a large language model is used to construct the entity class of the semantic network relationship to be extracted, and the large language model is QWEN2.5.

[0032] As a further optimization scheme of the method for constructing a network relationship of cultural and tourism attractions combined with a large model described in the present invention, the value of name1 and the value of name2 in step 4.1 are different.

[0033] In a second aspect, an embodiment of the present invention further provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein when the processor executes the computer program, the steps of the method for constructing a network relationship of cultural and tourism attractions combined with a large model as described in the first aspect or any corresponding embodiment thereof are implemented.

[0034] In a third aspect, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for constructing a network relationship of cultural and tourism attractions combined with a large model as described in the first aspect or any corresponding embodiment thereof are implemented.

[0035] Compared with the prior art, the present invention adopts the above technical solution and has the following technical effects:

[0036] (1) The present invention provides a method for constructing a network relationship of cultural and tourism attractions in combination with a large model. By using technologies such as large model prompt words, a complex network relationship of cultural and tourism attractions can be provided in combination with a large model;

[0037] (2) The present invention combines the semantic interpretation of the big model with complex network technology, and proposes a method for constructing the network relationship of cultural and tourism attractions combined with the big model, which can deeply perform complex network modeling of the network relationship of cultural and tourism attractions from the perspective of natural language semantics. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a schematic diagram of the overall process of the present invention.

[0039] Figure 2 It is a schematic diagram of the relationship between scenic spot geographic entity objects and text reviews.

[0040] Figure 3 It is to prompt the name of the scenic spot geographical entity object to the big model.

[0041] Figure 4 It is a schematic diagram of the process of extracting semantic network relationships between scenic spot geographic entity objects in reviews using a large model.

[0042] Figure 5 It is a schematic diagram of the Python code representation of the entity class of the semantic network relationship.

[0043] Figure 6 It is a distribution diagram of the calculation results of the node weighted degree centrality of the complex network model between scenic spots.

[0044] Figure 7 It is a distribution diagram of the calculation results of the node betweenness centrality of the complex network model between scenic spots. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0046] The big language model (abbreviated as "big model") plays an important role in the current social and economic development, and has a very prominent advantage in understanding text information compared to traditional machine learning technology. Therefore, the present invention proposes a new technical method, which will use the big model to extract the connection between the geographical entity objects of many scenic spot text reviews, and finally form the results into a complex network model, thereby generating greater application value.

[0047] See attached Figure 1 The general idea of ​​the present invention is as follows: First, construct the scenic spot geographic entity objects of the cultural and tourism scenic spot network. Secondly, collect text comments of the cultural and tourism scenic spot network, and each text comment corresponds to a unique scenic spot geographic entity object. Then, prompt the names of all scenic spot geographic entity objects to the big model context, and extract the semantic network relationship between the scenic spot geographic entity objects based on the big model. Finally, based on the semantic network relationship between the extracted scenic spot geographic entity objects, construct different types of semantic network relationship maps between the scenic spot geographic entity objects.

[0048] Step 1: First, it is necessary to construct the geographical entity object of the cultural and tourism scenic spot network. The geographical entity object of the scenic spot is also the node of the cultural and tourism scenic spot network relationship. It is necessary to use a large model to calculate the connection relationship between the geographical entity objects of the scenic spot.

[0049] Step 1.1: Crawling out the geographic entity objects of tourist attractions from professional tourist attraction websites.

[0050] Step 1.2: Each scenic spot geographic entity object has a unique name. The unique name of each scenic spot geographic entity object is mainly to ensure the statistical accuracy of the final result.

[0051] Step 2, see attached Figure 2, collect text reviews of cultural and tourism scenic spots network, each text review corresponds to a unique scenic spot geographic entity object. This is determined according to the actual application situation. Generally, the generation of a text review is based on each scenic spot geographic entity object.

[0052] Step 3, prompt the names of all scenic spot geographic entity objects to the big model;

[0053] Step 3.1, traverse the names of scenic spot geographic entity objects. The purpose of this step is to enable the large model to access each scenic spot geographic entity object;

[0054] Step 3.2, see attached Figure 3 , construct a large model prompt word, and confirm the name of the traversed scenic spot geographic entity object to the large model as the name of the scenic spot geographic entity object.

[0055] This step mainly solves the problem that the large model may not be able to effectively identify all scenic spot geographic entity objects. Therefore, the names of all scenic spot geographic entity objects in the study area are prompted to the large model through the method of large model prompt words.

[0056] Step 4, see attached Figure 4 , based on the big model, the semantic network relationship between scenic spot geographic entity objects is extracted;

[0057] Step 4.1, construct an entity class of the semantic network relationship to be extracted, and the entity class includes 4 attributes, namely name1, name2, rel1, and rel1a, where name1 represents the name of a scenic spot geographical entity, name2 represents the name of another scenic spot geographical entity, rel1 is a Boolean type data used to indicate whether there is an association relationship between scenic spot geographical entities, and rel1a represents the association relationship between scenic spot geographical entities;

[0058] Attached Figure 5 This is a diagram of the Python code representation of the entity class of the semantic network relationship. The main goal of building the entity class is to make the results returned by the large model standardized. Standardized results can not only help to quickly and accurately count the results, but also make the results returned by the large model more reliable.

[0059] In step 4.1, a large language model is used to construct the entity class of the semantic network relationship to be extracted, and the large language model is QWEN2.5.

[0060] Among them, the name1 value and name2 value in step 4.1 are different.

[0061] Step 4.2: Register the entity class as a parser that outputs in json format. The present invention allows the big model to output the results of the related entity class in the common JSON format, so that a very standard big model answer can be obtained in the end.

[0062] Step 4.3: Build a prompt word template and embed the output parser in step 4.2 into the prompt word template. Only when the parser output in json format is brought into the prompt word template can the large model output the results in standard json format.

[0063] Step 4.4, traverse each text comment of all text comments, and use the large model to extract the name set of the scenic spot geographical entity objects of each text comment; at the same time, if the extracted name set of the scenic spot geographical entity objects does not include the name of the scenic spot geographical entity object to which the text comment belongs, then the name set of the scenic spot geographical entity object is added with the name of the scenic spot geographical entity object to which the text comment belongs, and the final name set t of the scenic spot geographical entity object is obtained.

[0064] This step is critical, as it ensures that all scenic spot geographic entity objects in each text review are extracted first. It should be noted that a text review may not contain the name of the scenic spot geographic entity object to which it belongs. In this case, the name of the scenic spot geographic entity object needs to be added to the name set t of the scenic spot geographic entity object during the execution process. The name set t of the scenic spot geographic entity object will play an important role in step 4.5, which is used to determine whether there is a connection relationship between any two different scenic spot geographic entity objects. This processing method also further ensures the comprehensiveness and accuracy of the results returned by the large model.

[0065] Step 4.5: Further adjust the input value of the prompt word template in real time. The input value includes any two different names of scenic spot geographic entity objects in the name set t. Input any two different names of scenic spot geographic entity objects in the name set t into the prompt word template. According to the prompt word template, call the big model to obtain the big model return result, and extract the semantic network relationship between the scenic spot geographic entity objects from the return result. The key to this step is to enable the big model to comprehensively judge and analyze any two different scenic spot entity objects in a text review, thereby ensuring the comprehensiveness of the results obtained.

[0066] Step 5: Based on the semantic network relationships between the extracted scenic spot geographic entity objects, construct different types of semantic network relationship maps between the scenic spot geographic entity objects.

[0067] Step 5.1: After completing the traversal processing of all text reviews in Step 4.4 and Step 4.5, collect the json format results calculated by the large model. This step is to summarize the results of all text reviews by the large model, and then provide a systematic and comprehensive scenic spot geographic entity object relationship modeling for the analysis of the entire study.

[0068] Step 5.2: Perform complex network modeling on all json format results in step 5.1:

[0069] First, complex network modeling uses the names of geographical entity objects of all cultural and tourism attractions in the network as nodes of the complex network.

[0070] Secondly, when the rel1 value in each json format result is true, the name1 value and the name2 value in the json format result are two network nodes in a complex network edge connection relationship, and the name1 value and the name2 value constitute a complex network edge connection relationship; according to the nodes of the complex network and the complex network edge connection relationship, a complex network model is formed, and the complex network model is also the semantic network relationship map between the scenic spot geographic entity objects.

[0071] This step is also the process of processing each JSON format result returned by the big model. First, it is necessary to determine whether there is a relationship between the two scenic spot geographic entity objects given in the given result of the big model, that is, when the rel1 value in the json format result is true, there is a relationship between the two scenic spot geographic entity objects given, otherwise there is no relationship. The name1 value and the name2 value represent two scenic spot geographic entity objects, that is, two nodes of the complex network. Obviously, if the rel1 value is true, then there is a complex network connection edge relationship between the two nodes represented by the name1 value and the name2 value.

[0072] Finally, based on the different types of semantic network relationship maps between scenic spot geographic entity objects, clustering is performed on the rel1a median, and different types of complex network edge link relationships are formed according to the clustering results. For each complex network edge link relationship, different types of semantic network relationship maps between scenic spot geographic entity objects are constructed. The rel1a median records different types of connection relationships, such as recommendation relationships between scenic spots, spatial proximity relationships, etc. According to these different types of relationships, if clustering is performed, it is obvious that complex networks under different types of relationships can be obtained.

[0073] After constructing the complex network model, the index value of the node (scenic spot geographical entity object) can be calculated according to the complex network theory. Figure 6This is the distribution diagram of the calculation results of the node weighted degree centrality of the complex network model between scenic spots. Figure 7 It is a distribution diagram of the calculation results of the node betweenness centrality of the complex network model between scenic spots. These two figures can clearly reveal the spatial distribution characteristics of the important scenic spot geographical entity objects in the study area, as well as the spatial distribution characteristics of the "hub-type" scenic spot geographical entity objects. It can be seen that the direct results of the present invention will provide richer results, which also illustrates the practical value of the present invention.

[0074] An embodiment of the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. When the processor executes the computer program, the steps of the method for constructing the network relationship of cultural and tourism attractions combined with a large model as described above are implemented.

[0075] In a third aspect, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for constructing a network relationship of cultural and tourism attractions combined with a large model as described above are implemented.

[0076] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The schemes in the embodiments of the present invention may be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0077] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0078] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0079] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0080] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0081] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for constructing a network relationship of cultural and tourism attractions combined with a large model, characterized in that: include: Step 1: Construct the geographical entity object of the cultural and tourism scenic spot network; Step 2: Collect text reviews of the cultural and tourism scenic spots network, and each text review corresponds to a unique scenic spot geographic entity object; Step 3, prompt the names of all scenic spot geographic entity objects to the big model; Step 4: Extracting semantic network relationships between scenic spot geographic entity objects based on the big model; Step 5: Based on the semantic network relationships between the extracted scenic spot geographic entity objects, construct different types of semantic network relationship maps between the scenic spot geographic entity objects.

2. According to the method for constructing a network relationship of cultural and tourism attractions combined with a large model according to claim 1, it is characterized in that: Step 1 is as follows: Step 1.1, crawl out the geographical entity objects of tourist attractions from professional tourist attraction websites; Step 1.2: Each scenic spot geographic entity object has a unique name.

3. According to the method for constructing a network relationship of cultural and tourism attractions combined with a large model according to claim 1, it is characterized in that: Step 3 is as follows: Step 3.1, traverse the names of scenic spot geographic entity objects; Step 3.2: construct a large model prompt word, and confirm the name of the traversed scenic spot geographic entity object to the large model as the name of the scenic spot geographic entity object.

4. According to the method for constructing a network relationship of cultural and tourism attractions combined with a large model according to claim 1, it is characterized in that: Step 4 is as follows: Step 4.1, construct an entity class of the semantic network relationship to be extracted, and the entity class includes 4 attributes, namely name1, name2, rel1, and rel1a, where name1 represents the name of a scenic spot geographical entity, name2 represents the name of another scenic spot geographical entity, rel1 is a Boolean type data used to indicate whether there is an association relationship between scenic spot geographical entities, and rel1a represents the association relationship between scenic spot geographical entities; Step 4.2, register the entity class as a parser for json format output; Step 4.3, construct a prompt word template, and embed the output parser in step 4.2 into the prompt word template; Step 4.4, traverse each text comment of all text comments, and use the large model to extract the name set of the scenic spot geographic entity object of each text comment; at the same time, if the name set of the extracted scenic spot geographic entity object does not include the name of the scenic spot geographic entity object to which the text comment belongs, then add the name of the scenic spot geographic entity object to which the text comment belongs to the name set of the scenic spot geographic entity object; obtain the final name set t of the scenic spot geographic entity object; Step 4.5, further adjust the input value of the prompt word template in real time, the input value includes any two different names of scenic spot geographical entity objects in the name set t, input any two different names of scenic spot geographical entity objects in the name set t into the prompt word template, call the big model according to the prompt word template to obtain the big model return result, and extract the semantic network relationship between the scenic spot geographical entity objects from the return result.

5. According to the method for constructing a network relationship of cultural and tourism attractions combined with a large model according to claim 4, it is characterized in that: The value of rel1 is True or False.

6. According to the method for constructing a network relationship of cultural and tourism attractions combined with a large model according to claim 4, it is characterized in that: Step 5 is as follows: Step 5.1, after completing the traversal processing of all text comments in step 4.4 and step 4.5, collect the json format results obtained by the large model calculation; Step 5.2: Perform complex network modeling on all json format results in step 5.1: First, complex network modeling uses the names of all the scenic spots’ geographical entity objects in the cultural and tourism scenic spots network as nodes of the complex network; Secondly, when the rel1 value in each json format result is True, the name1 value and name2 value in the json format result are two network nodes of a complex network edge connection relationship, and the name1 value and name2 value form a complex network edge connection relationship; according to the complex network nodes and complex network edge connection relationships, a complex network model is formed, and the complex network model is also the semantic network relationship map between scenic spot geographic entity objects; Finally, based on the different types of semantic network relationship maps between scenic spot geographic entity objects, the rel1a median is clustered, and different types of complex network edge link relationships are formed according to the clustering results. For each complex network edge link relationship, different types of semantic network relationship maps between scenic spot geographic entity objects are constructed.

7. According to claim 4, a method for constructing a network relationship of cultural and tourism attractions combined with a large model is characterized in that: In step 4.1, a large language model is used to construct the entity class of the semantic network relationship to be extracted, and the large language model is QWEN2.

5.

8. According to the method for constructing a network relationship of cultural and tourism attractions combined with a large model according to claim 4, it is characterized in that: In step 4.1, the value of name1 and the value of name2 are different.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that: When the processor executes the computer program, the steps of the method for constructing a network relationship of cultural and tourism attractions combined with a large model as described in any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for constructing a network relationship of cultural and tourism attractions combined with a large model as described in any one of claims 1 to 8 are implemented.

Images