Intelligent exhibition hall explanation control method and sharing system based on large language model

Abstract

The present application relates to the field of exhibition hall intelligent management technology, and in particular to an exhibition hall intelligent explanation control method and sharing system based on a large language model, the method comprising obtaining an exhibition image, determining a visitor's tour route based on the exhibition image; identifying the group of exhibits to be visited corresponding to the tour route and the visiting order of each exhibit to be visited in the group of exhibits to be visited, the number of exhibits to be visited contained in the group of exhibits to be visited is not less than the preset number of exhibits; determining the visitor's initial feedback explanation content according to the group of exhibits to be visited and the corresponding visiting order; when the visitor's visit needs are obtained, identifying the interest demand features contained in the visit needs; optimizing the initial feedback explanation content based on the interest demand features, obtaining the target feedback explanation content, and feeding back the target feedback explanation content to the explanation receiving device carried by the visitor. The present application facilitates improving the accuracy of visitors receiving the explanation content.

Description

Technical Field

[0001] The present application relates to the technical field of intelligent management of exhibition halls, and in particular to an intelligent exhibition hall explanation control method and sharing system based on a large language model. Background Art

[0002] With the advancement of technology and the evolution of consumer demand, exhibition halls are still an important carrier of information dissemination, education and entertainment. Visitors can learn the background information of the exhibits and deeply understand the connotation and value of the exhibits by giving explanations while visiting the exhibition hall.

[0003] In the related technology, generally multiple explanation areas are set up in the exhibition hall, and explanation broadcasting equipment and positioning devices are set up in each explanation area. After the visitor wears the explanation receiving device and enters the explanation area, he can be identified and found by the positioning device set up in the explanation area. Then the explanation broadcasting equipment can transmit the pre-recorded explanation content to the explanation receiving device carried by the visitor. However, since the explanation area is fixed and limited, when the visitor is in a mobile state, the visitor may not be able to receive the explanation content due to inaccurate positioning or the received explanation content may be incorrect, which may affect the visitor's exhibition experience. Summary of the invention

[0004] In order to improve the accuracy of visitors receiving explanations, the present application provides an exhibition hall intelligent explanation control method and sharing system based on a large language model.

[0005] In the first aspect, the present application provides an exhibition hall intelligent explanation control method based on a large language model, which adopts the following technical solutions:

[0006] An intelligent exhibition hall explanation control method based on a large language model, comprising:

[0007] Acquire exhibition images, and determine the visitor's visiting route based on the exhibition images;

[0008] Identifying a group of exhibits to be visited corresponding to the tour route and a visiting order of each exhibit to be visited in the group of exhibits to be visited, wherein the number of exhibits to be visited in the group of exhibits to be visited is not less than a preset number of exhibits;

[0009] Determining the initial feedback explanation content of the visitor according to the exhibit groups to be visited and the corresponding visiting order;

[0010] When the visitor's visiting needs are obtained, identifying the interesting demand features contained in the visiting needs;

[0011] The initial feedback explanation content is optimized based on the interested demand characteristics to obtain target feedback explanation content, and the target feedback explanation content is fed back to the explanation receiving device carried by the visitor.

[0012] By adopting the above technical solution, by identifying the visitor's visiting route and the exhibit group to be visited, it is possible to provide personalized initial feedback explanation content according to the specific behavior of the visitor, and then optimize the initial feedback explanation content according to the received visiting needs, so as to improve the visitor's satisfaction and experience while enriching the feedback content, rather than fixedly feedback the pre-entered explanation content. In addition, after the explanation content is generated in real time according to the visitor's visiting route, the generated content can be fed back to the visitor in real time, without the need for the visitor to enter the fixed explanation area before the corresponding content can be fed back, thereby reducing the probability of visitors receiving incorrect content due to inaccurate positioning, and further improving the accuracy of visitors receiving the explanation content.

[0013] In a possible implementation, optimizing the initial feedback explanation content based on the interested demand feature to obtain target feedback explanation content includes:

[0014] Determine, based on the interested demand feature and a first mapping relationship, a linkage type corresponding to the interested demand feature, wherein the first mapping relationship is a correspondence between the interested demand feature and the linkage type, and the linkage type includes an origin linkage and a material linkage;

[0015] According to the linkage type, identifying linkage content included in the initial feedback explanation content;

[0016] The linkage content is determined as clue content, and the initial feedback explanation content is optimized based on the clue content to obtain the target feedback explanation content.

[0017] By adopting the above technical solution, by considering and analyzing the linkage types corresponding to the characteristics of interest needs in the process of forming the target feedback explanation content, it is convenient to improve the richness of the target feedback explanation content, thereby improving the visitor's experience. In addition, the linkage types that visitors are interested in are determined through the characteristics of their interest needs, and these linkage contents are used as clues to optimize the initial feedback explanation content. This can not only provide visitors with richer and more interesting information, but also improve the adaptability between the visitors' visiting needs and the received explanation content, thereby improving the visitor's satisfaction.

[0018] In a possible implementation, when the visitor's visiting demand is not obtained, the method further includes:

[0019] Determining the visitor's visit characteristics according to the exhibition image;

[0020] Determining similar persons to the visitor according to the visit characteristics, wherein the matching degree between the visit characteristics of the similar persons and the visit characteristics of the visitor is higher than a preset matching degree;

[0021] The similar interest demand characteristics corresponding to the similar persons are obtained, and the initial feedback explanation content is optimized based on the similar interest demand characteristics.

[0022] By adopting the above technical solution, through identifying and analyzing the visitor's visiting characteristics, it is convenient to determine the similar persons of the visitor. Since the similar persons may have similar visiting needs as the visitor, when the visitor has not issued a visiting demand, the visitor's initial feedback explanation content is optimized according to the visiting needs of the similar persons, so as to improve the richness of the initial feedback explanation content, so as to improve the effectiveness of the target feedback content.

[0023] In a possible implementation, after feeding back the target feedback explanation content to the explanation receiving device carried by the visitor, the method further includes:

[0024] Identify the exhibit identifiers contained in the target feedback explanation content, and divide the target feedback explanation content based on each exhibit identifier to obtain the exhibit feedback explanation content corresponding to each exhibit identifier;

[0025] According to the visiting order, each exhibit feedback explanation content is written into a temporary circular buffer pool, wherein the temporary circular buffer pool includes a plurality of data buffers, and each exhibit feedback explanation content corresponds to a data buffer;

[0026] When a review instruction is received, the review exhibit identifier contained in the review instruction is identified, and based on the review exhibit identifier, the exhibit feedback explanation content in the corresponding data buffer is fed back to the explanation receiving device carried by the visitor.

[0027] By adopting the above technical solution, after the target feedback explanation content is generated and fed back, the generation record and the generation result are not directly deleted, but the generated target feedback explanation content is temporarily stored to meet the visitor's review needs. In addition, since there may be linkage between the explanation contents corresponding to different exhibit identifications in the target feedback explanation content, the target feedback explanation content can be divided before writing it into the temporary buffer pool. By storing the exhibit feedback explanation content corresponding to each exhibit identification separately, it is convenient to improve the convenience of data management or data call process.

[0028] In one possible implementation, the method further includes:

[0029] Obtaining a visit demand log corresponding to the target feedback explanation content;

[0030] Determine the number of visit demands corresponding to each exhibit identifier according to the visit demand log, and determine the retention duration corresponding to each exhibit identifier based on the number of visit demands and the retention mapping relationship, wherein the retention mapping relationship is the corresponding relationship between the number of visit demands and the retention duration;

[0031] According to each exhibit identification and the retention time corresponding to each exhibit identification, the clearing time of each data buffer is determined.

[0032] By adopting the above technical solution, the generated target feedback explanation content is stored for the convenience of subsequent retrieval and viewing by visitors. However, since the number of visitors may be large, the amount of information that needs to be stored will also increase. By analyzing the visit demand log, the frequency of calling the feedback explanation content of different exhibits can be determined. Finally, based on the frequency of calling the feedback explanation content of each exhibit, the corresponding retention time is determined, so as to ensure that the exhibit feedback explanation content can meet the calling demand while reducing the occupancy rate of storage space.

[0033] In one possible implementation, the method further includes:

[0034] Determine the current visiting position of the visitor according to the exhibition image;

[0035] Acquire a historical visit demand log generated in a first preset time period for a current visit area, and identify a target interest demand feature corresponding to the historical visit demand log;

[0036] If the visitor is not detected to have issued a target visiting demand within the second preset time period, prompt information is generated according to the target interest demand characteristics, and the prompt information is fed back to the explanation receiving device carried by the visitor. The target visiting demand is a visiting demand containing the target interest demand characteristics.

[0037] By adopting the above technical solution, by analyzing the historical visiting needs of other visitors in the current visiting area where the visitor is located, the general points of interest in the current visiting area can be determined, and when the system detects that the visitor has not issued a target visiting need containing general points of interest within the second preset time period, prompt information is generated based on these characteristics and fed back to the visitor, so as to reduce the probability of the visitor missing important information. In addition, by tracking the visitor's location and visiting needs in real time, the explanation content and prompt information are adjusted in time, so as to provide personalized services to the visitors, and to enhance the visitor's visiting experience by enriching the feedback content.

[0038] In a second aspect, the present application provides a sharing system, which adopts the following technical solution:

[0039] A sharing system, comprising:

[0040] at least one processor;

[0041] Memory;

[0042] At least one application, wherein the at least one application is stored in a memory and configured to be executed by at least one processor, and the at least one application is configured to: execute the above-mentioned exhibition hall intelligent explanation control method based on a large language model.

[0043] In a third aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution:

[0044] A computer-readable storage medium includes: a computer program that can be loaded by a processor and execute the above-mentioned exhibition hall intelligent explanation control method based on a large language model.

[0045] In a fourth aspect, the present application provides a computer program product, which adopts the following technical solution:

[0046] A computer program product includes a computer program, which, when executed by a processor, implements the above-mentioned exhibition hall intelligent explanation control method based on a large language model.

[0047] In summary, the present application includes at least one of the following beneficial technical effects:

[0048] By identifying the visitor's tour route and the exhibit group to be visited, it is possible to provide personalized initial feedback explanation content based on the visitor's specific behavior, and then optimize the initial feedback explanation content based on the received visiting needs, so as to improve the visitor's satisfaction and experience while enriching the feedback content, rather than fixedly feedback the pre-entered explanation content. In addition, after the explanation content is generated in real time according to the visitor's tour route, the generated content can be fed back to the visitor in real time, without the need for the visitor to enter the fixed explanation area before the corresponding content can be fed back, thereby reducing the probability of visitors receiving incorrect content due to inaccurate positioning, and thus facilitating improving the accuracy of visitors receiving the explanation content.

[0049] After the target feedback explanation content is generated and fed back, the generation record and generation result are not directly deleted, but the generated target feedback explanation content is temporarily stored to meet the visitor's review needs. In addition, since there may be linkage between the explanation contents corresponding to different exhibit identifications in the target feedback explanation content, the target feedback explanation content can be divided before writing it into the temporary buffer pool. By storing the exhibit feedback explanation content corresponding to each exhibit identification separately, it is convenient to improve the convenience of data management or data call process. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a flow chart of an exhibition hall intelligent explanation control method based on a large language model in an embodiment of the present application;

[0051] Figure 2 This is a schematic diagram of a data storage process in an embodiment of the present application;

[0052] Figure 3 It is a structural diagram of a sharing system in an embodiment of the present application. DETAILED DESCRIPTION

[0053] The following is combined with Figure 1-3 This application is described in further detail.

[0054] After reading this specification, those skilled in the art may make non-creative modifications to this embodiment as needed, but such modifications are protected by patent law as long as they are within the scope of the claims of this application.

[0055] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0056] It should be noted that in the optional embodiments of the present application, the object information and other related data involved, when the embodiments in the present application are applied to specific products or technologies, need to obtain the permission or consent of the object, and the collection, use and processing of the relevant data need to comply with the relevant laws, regulations and standards of the relevant countries and regions. In other words, if the embodiments of the present application involve data related to the object, it needs to be obtained with the authorization and consent of the object, the authorization and consent of the relevant departments, and in compliance with the relevant laws, regulations and standards of the country and region. If personal information is involved in the embodiments, the acquisition of all personal information needs to obtain the consent of the individual. If sensitive information is involved, the separate consent of the information subject needs to be obtained. The embodiments also need to be implemented with the authorization and consent of the object.

[0057] Specifically, the embodiment of the present application provides an exhibition hall intelligent explanation control method based on a large language model, which is executed by a shared system, and the shared system can be a server or a terminal device, wherein the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc., but is not limited thereto. The terminal device and the server can be directly or indirectly connected via wired or wireless communication, and the embodiment of the present application does not limit this.

[0058] refer to Figure 1 , Figure 1 1 is a flow chart of an intelligent exhibition hall explanation control method based on a large language model in an embodiment of the present application, the method comprising steps S110 to S150, wherein:

[0059] Step S110: Acquire exhibition images, and determine the visitor's visiting route based on the exhibition images.

[0060] Specifically, the exhibition images can be collected by the image acquisition device installed in the exhibition hall and then uploaded to the sharing system. The exhibition images may contain one visitor or multiple visitors, and different visitors have different tour routes. When determining the visitor's tour route from the exhibition images, the mobile personnel target can be first identified from the exhibition images. Different personnel identifiers can be used for different mobile personnel targets. The movement route of each personnel identifier in the exhibition image can be determined by observing the position change of each personnel identifier. The movement route can be used as the movement trend of each visitor, and the tour route corresponding to each visitor is predicted based on the movement trend.

[0061] There is a mapping relationship between movement trends and visiting routes. The visiting route corresponding to the movement trend can be determined based on the visitor's movement trend and the mapping relationship. The mapping relationship between the movement trend and the visiting route includes the visiting route corresponding to each movement trend, which is determined by relevant staff based on historical visiting data and uploaded to the sharing system. The specific content is not limited in the embodiments of this application.

[0062] Step S120: identifying the group of exhibits to be visited corresponding to the visiting route and the visiting order of each exhibit to be visited in the group of exhibits to be visited, and the number of exhibits to be visited included in the group of exhibits to be visited is not less than a preset number of exhibits.

[0063] Specifically, each tour route corresponds to different exhibits to be visited, and different routes correspond to different numbers of exhibits to be visited. In order to improve the effectiveness of the tour route, in the embodiment of the present application, the number of exhibits to be visited corresponding to each tour route is not less than the preset number of exhibits, wherein the preset number of exhibits can be 2 or 3. The exhibits to be visited corresponding to each tour route can be identified from the exhibition image by feature recognition, and the tour route with the number of exhibits to be visited less than the preset number of exhibits can be determined as an abnormal tour route. After the sharing system detects the generation of an abnormal tour route, the tour route can be regenerated according to the movement trend of the visitor corresponding to the abnormal tour route.

[0064] The exhibits to be visited corresponding to the tour route may be determined as a group of exhibits to be visited, and the visiting order of the group of exhibits to be visited is the order in which each exhibit to be visited appears in the corresponding tour route.

[0065] Step S130: determining the initial feedback explanation content of the visitor according to the exhibit groups to be visited and the corresponding visiting sequence.

[0066] Specifically, the initial feedback explanation content includes the basic explanation content of each exhibit to be visited. Each basic explanation content can be entered into the sharing system in advance by relevant staff. The order of each basic explanation content in the initial feedback explanation content is consistent with the visiting order. The speech speed of the initial feedback explanation content can be automatically adjusted according to the visitor's moving position, so that the visitor can refer to the exhibits to be visited when listening to the initial feedback explanation content. For example, when the visitor's moving route includes 3 exhibits to be visited, namely exhibits a, exhibits b and exhibits c, the playback speed of the initial feedback explanation content can be adjusted according to the visitor's moving speed between the display location of exhibit a and the display location of exhibit b, so as to ensure that the visitor can listen to the basic explanation content about exhibit a before arriving at exhibit b.

[0067] Step S140: When the visitor's visiting needs are obtained, the interested demand features included in the visiting needs are identified.

[0068] Specifically, the visitor carries a device for receiving explanations, which can communicate with the sharing system, receive explanation content fed back by the sharing system, and upload the collected visiting requirements to the sharing system, wherein the visiting requirements of the visitor can be uploaded to the sharing system in text format, and when the explanation receiving device has a recording function, it can also be uploaded to the sharing system in audio format. When the visiting requirements are obtained, semantic recognition can be performed on the visiting requirements to determine the interesting demand features contained in the visiting requirements. The interesting demand features can be the history of the exhibits, the background of the exhibits, the texture of the exhibits, the texture of the exhibits, etc. The specific interesting demand features are not specifically limited in the embodiments of this application and can be set by relevant technical personnel.

[0069] Step S150: Optimize the initial feedback explanation content based on the interested demand characteristics to obtain the target feedback explanation content, and feed back the target feedback explanation content to the explanation receiving device carried by the visitor.

[0070] Specifically, the initial feedback explanation content includes basic explanation content for each exhibit to be visited, but the basic explanation content may lack descriptions of relevant interest demand characteristics. The initial feedback explanation content is optimized based on the interest demand characteristics to facilitate personalized services for visitors.

[0071] Furthermore, in order to improve the adaptability between the visitor's visiting needs and the received explanation content, the initial feedback explanation content is optimized based on the interested demand characteristics to obtain the target feedback explanation content, which may specifically include:

[0072] Based on the demand feature of interest and the first mapping relationship, determine the linkage type corresponding to the demand feature of interest, the first mapping relationship is the correspondence between the demand feature of interest and the linkage type, and the linkage type includes origin linkage and material linkage; according to the linkage type, identify the linkage content included in the initial feedback explanation content; determine the linkage content as the clue content, optimize the initial feedback explanation content based on the clue content, and obtain the target feedback explanation content.

[0073] Specifically, in addition to directly using the interest demand features to optimize the initial feedback explanation content, the interest demand features can also be analyzed to determine the linkage type corresponding to the interest demand features, and then the initial feedback explanation content can be optimized based on the linkage type. The linkage types corresponding to different interest demand features may be different. For example, when the interest demand features are exhibit history and exhibit background, the linkage type corresponding to the interest demand features can be determined to be origin linkage; when the interest demand features are exhibit texture and exhibit texture, the linkage type corresponding to the interest demand features can be determined to be material linkage. The first mapping relationship contains interest demand features corresponding to different linkage types. The number of interest demand features corresponding to one linkage type can be 1 or more. The specific number is not specifically limited in the embodiment of this application. The linkage type corresponding to the interest demand features in the visitor's visiting needs can be determined through the first mapping relationship. The specific content of the first mapping relationship is not specifically limited in the embodiment of this application and can be set by relevant technical personnel.

[0074] After determining the linkage type corresponding to the feature of interest, in addition to supplementing and enriching the initial feedback explanation content according to the linkage type, it is also necessary to link the basic explanation content of multiple exhibits to be visited contained in the initial feedback explanation content. The specific implementation method can first extract the linkage content containing the linkage type from the initial feedback explanation content according to the linkage type and the trained large language model, and then determine the linkage content as the clue content. The clue content has coherence and key points, and the technical explanation content of multiple exhibits to be visited can be connected in series according to the preset relationship, wherein the preset relationship can be time, person, etc., and the specific relationship can be set by relevant staff. The training process of the large language model can specifically include: using a large amount of text data and label data corresponding to each text data as training data, wherein the text data can be various types of text such as web pages, news articles, novels, etc. The specific training data is not specifically limited in the embodiment of this application. Before training, the text data can be cleaned and preprocessed, such as deleting useless characters contained in the text data. The preprocessed text data is then used to train the large language model to be trained. In each round of training, the large language model to be trained needs to compare the output result with the label data corresponding to the text data, and calculate the loss function between the two to assess whether the large language model to be trained has completed the training, wherein the loss function can be a cross-entropy loss function, and the specific loss function is not specifically limited in the embodiments of the present application.

[0075] In the model training process, an unsupervised learning method can be used, that is, the large language model to be trained can automatically learn the grammar, semantics and context of the language from a large amount of text data, so as to improve the accuracy of the optimization results, thereby improving the quality of the target feedback explanation content. The specific training process is not described in detail in this application embodiment.

[0076] The content of the initial feedback explanation can be optimized based on the clue content.

[0077] By considering and analyzing the linkage types corresponding to the characteristics of interest needs in the process of forming the target feedback explanation content, it is convenient to improve the richness of the target feedback explanation content, thereby improving the visitor's experience. In addition, the linkage types that visitors are interested in are determined through the characteristics of their interest needs, and these linkage contents are used as clues to optimize the initial feedback explanation content. This can not only provide visitors with richer and more interesting information, but also improve the adaptability between visitors' visiting needs and the received explanation content, thereby improving visitor satisfaction.

[0078] After the target feedback explanation content is determined, the target transmission channel can be determined from multiple transmission channels according to the visitor's current location, and the target feedback explanation content is fed back to the explanation receiving device carried by the visitor through the target transmission channel.

[0079] For the embodiments of the present application, by identifying the visitor's visiting route and the exhibit group to be visited, it is possible to provide personalized initial feedback explanation content based on the visitor's specific behavior, and then optimize the initial feedback explanation content based on the received visiting needs, so as to enrich the feedback content while improving the visitor's satisfaction and experience, rather than fixedly feedback the pre-entered explanation content. In addition, after the explanation content is generated in real time according to the visitor's visiting route, the generated content can be fed back to the visitor in real time, without the need for the visitor to enter the fixed explanation area before the corresponding content can be fed back, thereby reducing the probability of visitors receiving incorrect content due to inaccurate positioning, and further improving the accuracy of visitors receiving the explanation content.

[0080] Since the initial feedback explanation content can be optimized according to the visitor's visiting needs, so as to obtain richer target feedback explanation content, when the visitor's visiting needs are not obtained, in order to improve the effectiveness of the target feedback content, the method provided in the embodiment of the present application also includes:

[0081] Determine the visitor's visiting characteristics based on the exhibition images; determine similar persons to the visitor based on the visiting characteristics, and the matching degree between the visiting characteristics of the similar persons and the visiting characteristics of the visitor is higher than the preset matching degree; obtain similar interest demand characteristics corresponding to the similar persons, and optimize the initial feedback explanation content based on the similar interest demand characteristics.

[0082] Specifically, the visitor's visit features include facial features, clothing features, and behavioral features. The visitor's visit features can be used to predict the visitor's areas of interest, and then the visitor's similar personnel can be determined through the visit features. Since similar personnel may be similar to the visitor's visit needs, when the visitor does not send a visit demand, the visitor's initial feedback explanation content is optimized according to the visit needs of similar personnel. Among them, when determining the visitor's visit features based on the exhibition image, computer vision technology can be used to identify the visitor's visit features in the exhibition image. Computer vision technology can include face recognition technology, behavior analysis algorithm, etc. In addition, by combining spatial positioning technology, the visitor's behavior trajectory can be identified from the exhibition image. After identifying the visit features of each visitor, the recognition results will be stored for a limited time, that is, the recognition results will be updated and replaced according to the preset storage frequency to improve the accuracy of determining the visitor's similar personnel. Similar personnel are personnel whose visit features have a matching degree higher than the preset matching degree and have sent a visit demand to the sharing system. The specific preset matching degree is not specifically limited in the embodiment of this application and can be set by relevant technical personnel. According to the visiting demands sent by similar persons to the sharing system, corresponding similar interest demand features are determined. The specific process can refer to the process of identifying the interest demand features included in the visiting demands in the above embodiment, which will not be described in detail here.

[0083] Similar interest demand features are interest demand features corresponding to similar persons, which can characterize the needs of similar persons for participating in the exhibition. When the visitor's visiting needs are not obtained, the initial feedback explanation content is not fed back using any interest demand features, but the visitor's initial feedback explanation content is optimized according to similar interest demand features, so as to improve the adaptability between the optimization results and the actual visiting needs of the visitors while optimizing the initial feedback explanation content, thereby facilitating improving the satisfaction of the visitors. The method of optimizing the initial feedback explanation content according to similar interest demand features can refer to the steps of optimizing the initial feedback explanation content according to the interest demand features in the above embodiment, which will not be repeated here.

[0084] Further, in order to facilitate the convenience of data management or data calling, in the embodiment of the present application, after the target feedback explanation content is fed back to the explanation receiving device carried by the visitor, steps S1 to S3 are also included, such as Figure 2 As shown, where:

[0085] Step S1: identifying the exhibit identifiers contained in the target feedback explanation content, and dividing the target feedback explanation content based on each exhibit identifier to obtain the exhibit feedback explanation content corresponding to each exhibit identifier.

[0086] Specifically, since the target feedback explanation content is formed after linkage and optimization based on the basic explanation content corresponding to each exhibit to be visited, there may be overlaps in the explanation content corresponding to different exhibits to be visited in the target feedback explanation content. For example, the explanation content related to the exhibit a to be visited has been answered at 10:00. Since there is a correlation between the exhibit a to be visited and the exhibit b to be visited, the explanation content answered at 10:05 may also include the explanation content related to the adhesive sheet a to be visited. In order to ensure the integrity of the explanation content corresponding to each exhibit to be visited, when storing the target feedback explanation content, the target feedback explanation content can be divided based on each exhibit identifier to obtain all the explanation content corresponding to each exhibit to be visited. Among them, different exhibits to be visited have different exhibit identifiers. The exhibit identifier can be an exhibit number, an exhibit name, etc. The specific content is not specifically limited in the embodiment of this application, as long as the corresponding exhibit to be visited can be uniquely determined according to the exhibit identifier. The explanation content generally contains the corresponding exhibit identifier, so the target feedback explanation content can be divided by the exhibit identifier.

[0087] Step S2: writing each exhibit feedback explanation content into a temporary circular buffer pool according to the visiting order. The temporary circular buffer pool includes a plurality of data buffers, and each exhibit feedback explanation content corresponds to a data buffer room.

[0088] Specifically, a temporary circular buffer is a data structure that is often used to handle data memory management. The basic idea of ​​a circular buffer is to use a circular array or linked list to store data, where data can be written from one end and read from the other end, forming a "ring" flow. The circular buffer contains multiple data buffers, each of which is an independent space. Therefore, after the exhibit feedback explanation content is written into the temporary circular buffer, there will be no data confusion between different exhibit feedback explanation contents, thereby facilitating the security of the exhibit feedback explanation content during the storage process. When multiple exhibit feedback explanation contents are written into the temporary circular buffer, they can be written according to the visiting order of the exhibits to be visited according to each exhibit feedback explanation content, thereby facilitating the convenience of the writing process. After generating the target feedback explanation content, the target feedback explanation content is stored to facilitate visitors to review the relevant explanation content after the visit.

[0089] In order to reduce the space occupied by data storage, each data buffer has a corresponding clearing time, which is the time corresponding to the preset retention time after the writing time. In the embodiment of the present application, the preset retention time corresponding to different data buffers can be the same, that is, the preset retention time can be uniformly set, and the content written in the corresponding data buffer is cleared after the preset retention time ends. The retention time of each data buffer can also be set according to the visitor's need to review the feedback explanation content of each exhibit. The specific method can be:

[0090] Obtain the visit demand log corresponding to the target feedback explanation content; determine the number of visit demands corresponding to each exhibit identification according to the visit demand log, and determine the retention duration corresponding to each exhibit identification based on the visit demand number and the retention mapping relationship, the retention mapping relationship being the correspondence between the visit demand number and the retention duration; determine the clearing time of each data buffer according to each exhibit identification and the retention duration corresponding to each exhibit identification.

[0091] Specifically, according to the visiting location when the visitor issues a visiting request and the content of the visiting request, the exhibit corresponding to each visiting request can be determined, and the visiting request log records the number of visiting requests corresponding to each exhibit to be visited, that is, the visiting request log records the number of visiting requests corresponding to each exhibit identification. The higher the number of visiting requests, the higher the visitor's interest in the corresponding exhibit, and the greater the probability that the visitor will revisit the exhibit later. Therefore, the retention time corresponding to the exhibit identification is also longer. The retention mapping relationship includes retention times corresponding to different numbers of visiting requests. The specific content is not specifically limited in the embodiment of the present application, and can be determined by relevant staff based on historical experimental data and uploaded to the sharing system.

[0092] After determining the retention time corresponding to each exhibit identifier, the corresponding clearing time is determined based on the writing time of the exhibit feedback content corresponding to each exhibit identifier into the temporary circular buffer pool, and a clearing instruction is generated based on each clearing time, so that the clearing operation between the corresponding data buffers can be completed on time at the clearing time. By analyzing the visit demand log, the calling frequency of different exhibit feedback explanation content can be determined, and finally the corresponding retention time is determined based on the calling frequency of each exhibit feedback explanation content, so as to ensure that the exhibit feedback explanation content can meet the calling demand while reducing the storage space occupancy rate.

[0093] Step S3: When a review instruction is received, the review exhibit identifier contained in the review instruction is identified, and based on the review exhibit identifier, the exhibit feedback explanation content in the corresponding data buffer is fed back to the explanation receiving device carried by the visitor.

[0094] Specifically, the replay instruction can be sent by the visitor through the explanation receiving device carried by the visitor to the sharing system, and the replay exhibit identifier contained in the replay instruction is identified by feature recognition technology, and then the corresponding data buffer is determined from the temporary circular buffer pool based on the identified replay exhibit identifier, and the exhibit identifier corresponding to the data buffer is consistent with the replay exhibit identifier. The generated target feedback explanation content is temporarily stored to meet the visitor's replay needs.

[0095] Furthermore, in order to reduce the probability of visitors missing important information, the method provided in the embodiment of the present application also includes:

[0096] The current visiting position of the visitor is determined based on the exhibition image; the historical visiting demand log generated in the current visiting area within the first preset time period is obtained, and the target interest demand characteristics corresponding to the historical visiting demand log are identified; if the visitor is not detected to have issued a target visiting demand within the second preset time period, prompt information is generated based on the target interest demand characteristics, and the prompt information is fed back to the explanation receiving device carried by the visitor, and the target visiting demand is a visiting demand containing the target interest demand characteristics.

[0097] Specifically, the current visiting position of the visitor can be located by way of landmark recognition. After the current visiting position is determined, the current visiting area of ​​the visitor can be determined according to the pre-entered area division results. Each visiting area corresponds to different exhibits to be visited. Since different visiting areas correspond to different exhibits to be visited, correspondingly, the visiting demands put forward by different visitors in different visiting areas during historical visiting periods are also different. The historical visiting demand log contains the visiting demands of different visitors received from different visiting areas during historical visiting periods. The first preset time period can be 12 hours before the current moment, or 18 hours before the current moment. The specific duration is not specifically limited in the embodiment of the present application and can be set by relevant technical personnel. The target visiting demand is the historical visiting demand with a larger number of providers among all the historical visiting demands generated within the first preset time period. For example, there are 6 historical visiting demands, of which 8 people provide the content for historical visiting demand 1, and it can be determined that the number of providers of historical visiting demand 1 is 8. When determining the target visit demand, multiple historical visit demands can be sorted according to the number of people provided by each historical visit demand, and the historical visit demand before the preset ranking can be determined as the target visit demand, wherein the preset ranking can be the 4th or 5th, and the specific preset ranking is not specifically limited in the embodiment of the present application. After determining the target visit demand, the target interest demand feature corresponding to each target visit demand is identified by feature recognition. The specific identification method can refer to the implementation method of identifying the interest demand feature contained in the visit demand in the above embodiment, which will not be described in detail here.

[0098] The second preset time period can be 30 minutes or 40 minutes after the target visiting demand is determined, and the specific duration is not specifically limited in the embodiment of the present application. The visiting demands issued by visitors within the second preset time period are counted, and it is determined whether there are visiting demands containing the target interest demand characteristics. The specific implementation method can be to identify the interest demand characteristics corresponding to the visiting demands within the second preset time period, and match the identified interest demand characteristics with the target interest demand characteristics to determine whether there are visiting demands containing the target interest demand characteristics in the visiting demands issued within the second preset time period. If not, a prompt message containing the target interest demand characteristics is generated. The prompt message can be played after the target feedback explanation content is played, or it can be played during the process of playing the target feedback explanation content. The specific playback time is not specifically limited in the embodiment of the present application, as long as the prompt message containing the target interest demand characteristics can be fed back to the visitor, so as to reduce the probability of visitors missing important information.

[0099] The present application embodiment provides a sharing system, such as Figure 3 As shown, Figure 3 The sharing system 300 shown includes: a processor 301 and a memory 303. The processor 301 and the memory 303 are connected, such as through a bus 302. Optionally, the sharing system 300 may also include a transceiver 304. It should be noted that in actual applications, the transceiver 304 is not limited to one, and the structure of the sharing system 300 does not constitute a limitation on the embodiments of the present application.

[0100] The processor 301 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. The processor 301 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0101] The bus 302 may include a path to transmit information between the above components. The bus 302 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 302 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 The fact that only one line is used in the diagram does not mean that there is only one bus or only one type of bus.

[0102] The memory 303 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0103] The memory 303 is used to store the application code for executing the solution of the present application, and the execution is controlled by the processor 301. The processor 301 is used to execute the application code stored in the memory 303 to implement the contents shown in the above method embodiment. Figure 3 The sharing system shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0104] An embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer-readable storage medium is run on a computer, the computer can execute the corresponding content in the aforementioned method embodiment.

[0105] An embodiment of the present application provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the method in any of the above embodiments is implemented.

[0106] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0107] The above description is only a partial implementation method of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. An intelligent exhibition hall explanation control method based on a large language model, characterized in that: include: Acquire exhibition images, and determine the visitor's visiting route based on the exhibition images; Identifying a group of exhibits to be visited corresponding to the tour route and a visiting order of each exhibit to be visited in the group of exhibits to be visited, wherein the number of exhibits to be visited in the group of exhibits to be visited is not less than a preset number of exhibits; Determining the initial feedback explanation content of the visitor according to the exhibit groups to be visited and the corresponding visiting order; When the visitor's visiting needs are obtained, identifying the interesting demand features contained in the visiting needs; Optimizing the initial feedback explanation content based on the interested demand characteristics to obtain target feedback explanation content, and feeding back the target feedback explanation content to the explanation receiving device carried by the visitor; The step of optimizing the initial feedback explanation content based on the interested demand feature to obtain the target feedback explanation content includes: Determine, based on the interested demand feature and a first mapping relationship, a linkage type corresponding to the interested demand feature, wherein the first mapping relationship is a correspondence between the interested demand feature and the linkage type, and the linkage type includes an origin linkage and a material linkage; According to the linkage type, identifying linkage content included in the initial feedback explanation content; Determine the linkage content as clue content, optimize the initial feedback explanation content based on the clue content, and obtain the target feedback explanation content; Wherein, when the visitor's visiting demand is not obtained, the method further includes: Determining the visitor's visit characteristics according to the exhibition image; Determining similar persons to the visitor according to the visit characteristics, wherein the matching degree between the visit characteristics of the similar persons and the visit characteristics of the visitor is higher than a preset matching degree; Acquire similar interest demand characteristics corresponding to the similar persons, and optimize the initial feedback explanation content based on the similar interest demand characteristics; Wherein, after feeding back the target feedback explanation content to the explanation receiving device carried by the visitor, it also includes: Identify the exhibit identifiers contained in the target feedback explanation content, and divide the target feedback explanation content based on each exhibit identifier to obtain the exhibit feedback explanation content corresponding to each exhibit identifier; According to the visiting order, each exhibit feedback explanation content is written into a temporary circular buffer pool, wherein the temporary circular buffer pool includes a plurality of data buffers, and each exhibit feedback explanation content corresponds to a data buffer; When a review instruction is received, the review exhibit identifier contained in the review instruction is identified, and based on the review exhibit identifier, the exhibit feedback explanation content in the corresponding data buffer is fed back to the explanation receiving device carried by the visitor.

2. The intelligent exhibition hall explanation control method based on a large language model according to claim 1 is characterized in that: Also includes: Obtaining a visit demand log corresponding to the target feedback explanation content; Determine the number of visit demands corresponding to each exhibit identifier according to the visit demand log, and determine the retention duration corresponding to each exhibit identifier based on the number of visit demands and the retention mapping relationship, wherein the retention mapping relationship is the corresponding relationship between the number of visit demands and the retention duration; According to each exhibit identification and the retention time corresponding to each exhibit identification, the clearing time of each data buffer is determined.

3. The intelligent exhibition hall explanation control method based on a large language model according to claim 1 is characterized in that: Also includes: Determine the current visiting position of the visitor according to the exhibition image; Acquire a historical visit demand log generated in a first preset time period for a current visit area, and identify a target interest demand feature corresponding to the historical visit demand log; If the visitor is not detected to have issued a target visiting demand within the second preset time period, prompt information is generated according to the target interest demand characteristics, and the prompt information is fed back to the explanation receiving device carried by the visitor. The target visiting demand is a visiting demand containing the target interest demand characteristics.

4. A sharing system, characterized in that: The sharing system includes: at least one processor; Memory; At least one application, wherein the at least one application is stored in a memory and configured to be executed by at least one processor, and the at least one application is configured to: execute an exhibition hall intelligent explanation control method based on a large language model as described in any one of claims 1-3.

5. A computer-readable storage medium, characterized in that: include: A computer program is stored which can be loaded by a processor and execute an intelligent exhibition hall explanation control method based on a large language model as described in any one of claims 1-3.

6. A computer program product, characterized in that It includes a computer program, which, when executed by a processor, implements the steps of an exhibition hall intelligent explanation control method based on a large language model as described in any one of claims 1 to 3.

Images