Method and system for managing collection online communication appreciation platform
By generating unique identifiers and smart contract records on the collection appreciation platform, combined with Siamese Network and TextRank algorithms, the problem of collection consistency judgment and appreciation report accuracy in the collection appreciation platform is solved, and a more efficient and transparent appreciation process is achieved, improving the improvement of User experience and data security.
Patent Information
- Application Number
- CN202410988434.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-23
- Publication Date
- 2024-11-15
AI Technical Summary
The existing collection appreciation platform is difficult to ensure consistency judgment between the collection and the items sent by users. The lack of a systematic physical appreciation process affects the trust and accuracy of the appreciation report.
By publishing collection appreciation requests on the platform, selecting physical mailing or image appreciation, generating unique identifiers and recording them on the blockchain, performing smart contract recording and secondary appreciation, and using Siamese Network model and TextRank algorithm for appreciation experts to allocate and report Consistency evaluation and generate a final appreciation report.
It improves the accuracy and trust of collection appreciation, ensures transparency and consistency of appreciation reports, reduces subjective factors, and enhances user experience and data security.
Smart Images

Figure CN118968135A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of appreciation platform management, and specifically relates to a management method and system for an online collection communication and appreciation platform. Background Art
[0002] With the rapid development of Internet technology, especially the widespread application of cloud computing, big data, artificial intelligence and other technologies, a solid technical foundation has been provided for the development of the collection online communication and appreciation platform management system. These technologies make it possible to manage the collection digitally and informatized, and also provide users with a more convenient and efficient communication and appreciation experience. With the improvement of people's living standards and cultural literacy, more and more people are beginning to pay attention to and participate in the appreciation of collections. They hope to learn about the detailed information and appreciation data of the collections through online platforms. Therefore, it is particularly important to develop a collection online communication and appreciation platform management system with comprehensive functions and good user experience. The collection appreciation platform uses Internet technology and digital means through a comprehensive software system to centrally manage, display and appreciate various collections.
[0003] Most collection appreciation platforms only have a single appreciation method, lack records of physical unpacking data, and it is difficult to ensure that the platform makes consistency judgments on the collections and items sent by users. At the same time, the lack of a systematic physical appreciation process affects the credibility and accuracy of the appreciation report. Summary of the invention
[0004] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a management method and system for an online communication and appreciation platform for collections, which is used to solve the technical problems that it is difficult to ensure that the platform makes consistency judgments on the collections and the items sent by users, and at the same time, there is a lack of a systematic physical appreciation process, which affects the credibility and accuracy of the appreciation report.
[0005] To solve the above problems, the first aspect of the present invention provides a management method and system for an online collection communication and appreciation platform, comprising the following steps:
[0006] Users post a collection appreciation request on the platform, submit the collection identity data, and choose to have the collection appreciated by mail or by picture;
[0007] When the user chooses to appreciate the picture, the platform directly sends the collection identity data to the appreciation expert's mailbox, and the appreciation expert issues an appreciation report. The platform sends the appreciation report to the user end to complete the collection appreciation request;
[0008] When the user chooses to have the item mailed for appreciation, the platform generates a unique identifier for the item and generates a smart contract on the blockchain to record the item's identity data and request status. After the platform receives the item, it takes another unpacking photo of the item, uploads it to the smart contract, and updates the request status.
[0009] Generate collection classification labels for collections based on collection identity data, and assign appreciation experts to different appreciation groups based on their personal appreciation experience information submitted by them;
[0010] According to the collection classification information, the collections are assigned to different appreciation groups. The appreciation experts will appreciate the collections and generate appreciation reports, evaluate the consistency of the appreciation reports, and conduct a second appreciation of the collections based on the consistency evaluation results of the appreciation reports, and generate a final appreciation report;
[0011] The final appreciation report is uploaded to the blockchain, and a unique identifier and timestamp are generated for the appreciation report. The identifier of the appreciation report is matched with the unique identifier of the collection to complete the collection appreciation request.
[0012] As a further solution of the present invention: the collection identity data includes: the name, material, age, collection method and six views of the collection.
[0013] As a further solution of the present invention: generating collection classification labels for collections according to collection identity data, and assigning appreciation experts to different appreciation groups according to personal appreciation experience information submitted by appreciation experts, including the following steps:
[0014] Based on the collection identity data, the collection is classified according to its texture, age and purpose, and collection classification labels are generated for the collection;
[0015] Based on the personal appreciation experience information provided by the appreciation experts when registering on the platform, the appreciation experts will be assigned to different appreciation groups according to the texture, age and purpose of the collections that they appreciate most.
[0016] As a further solution of the present invention: according to the identity data of the collection, the collection is classified according to the texture, age and purpose of the collection, and a collection classification label is generated for the collection, including the following steps:
[0017] Collect a large number of image pairs of different textures, ages and uses as training sets. The image pairs include: image pairs representing the same collection and image pairs representing different collections. Each image pair is labeled to indicate whether it is from the same collection.
[0018] The image pairs in the training set are input into the two sub-networks of the Siamese Network model respectively, and the Siamese Network model is trained to identify whether the image pairs are images of the same collection;
[0019] Obtain the six-view data in the collection identity data, as well as the collection image information taken again after the platform receives the collection in person;
[0020] Select the front view image data of the six-view data in the collection identity data and the front view image data in the collection image information taken again after the platform receives the collection in person, to form an image pair;
[0021] Input the image pairs of the collection into the trained Siamese Network model to identify whether the image pairs are images of the same collection;
[0022] If the image information of the collection taken again after the recognition platform receives the physical collection is different from the six-view data in the collection identity data, the collection image will be taken again, and the recognition result of the Siamese Network model and the re-taken collection image will be uploaded to the smart contract of the blockchain, and the collection will be returned to the user;
[0023] If the image information of the collection taken again after the recognition platform receives the collection and the six-view data in the collection identity data are the same picture of the collection, the user will continue to publish the collection appreciation request on the platform;
[0024] Get the name, material, age and collection method data of the collection;
[0025] The TextRank algorithm is used to extract keywords from the name, material, and age information of the collection;
[0026] The collection is set according to the keywords in the material information to generate the texture classification label of the collection, the collection is set according to the keywords in the age information to generate the era classification label of the collection, and the collection is set according to the keywords in the name information to generate the use classification label of the collection.
[0027] As a further solution of the present invention: according to the personal appreciation experience information provided by the appreciation experts when registering on the platform, the appreciation experts are assigned to different appreciation groups according to the texture, age and purpose of the collections most appreciated by the appreciation experts, including the following steps:
[0028] According to the personal appreciation experience information provided by the appreciation experts when registering on the platform, the texture, age and use information of the collections appraised by the experts are obtained, and the texture, age and use information of the collections most appreciated by the appreciation experts are counted;
[0029] The TextRank algorithm is used to extract keywords from the texture, age, and usage information of the collections that are most appreciated by the experts;
[0030] According to the extracted keywords of collection texture information, the appreciation experts are grouped into collection texture groups; according to the extracted keywords of collection age information, the appreciation experts are grouped into collection age groups; according to the extracted keywords of collection usage information, the appreciation experts are grouped into collection usage groups.
[0031] As a further solution of the present invention: according to the collection classification information, the collection is assigned to different appreciation groups, and appreciation experts appreciate the collection and generate appreciation reports, evaluate the consistency of the appreciation reports, and according to the consistency evaluation results of the appreciation reports, perform a second appreciation on the collection and generate a final appreciation report, including the following steps:
[0032] According to the collection classification information, the collections are assigned to different appreciation groups, and appreciation experts are randomly selected in each appreciation group to appreciate the collections and generate appreciation reports;
[0033] Generate an appreciation report for each appreciation group, extract the texture, age and use text data of the collection, and calculate the consistency of the appreciation report by comparing the texture, age and use text data of the collection in the appreciation report generated by each appreciation group. If the consistency of the appreciation report is lower than the preset threshold, select appreciation experts from the appreciation group corresponding to the collection again, conduct a second appreciation, and generate an appreciation report; otherwise, no second appraisal will be conducted;
[0034] The appreciation reports with the highest degree of uniformity among all the appreciation reports are screened, and the final appreciation report is generated according to the texture, era and usage text data of the appreciation reports with the highest degree of uniformity.
[0035] As a further solution of the present invention, an appreciation report is generated for each appreciation group, the texture, age and use text data of the collection are extracted, and the consistency of the appreciation report is calculated by comparing the texture, age and use text data of the collection in the appreciation report generated by each appreciation group, including the following steps:
[0036] Obtain the appreciation report generated by each appreciation group, extract the texture, age and use text data of the collection, and extract the keywords of the texture, age and use text data of the collection through the TextRank algorithm;
[0037] Through the Word2Vec language model, the keywords are mapped into a high-dimensional space, and the similarity of the keywords in the texture, era and usage text data in different appreciation reports is calculated;
[0038] By using the TF-IDF algorithm, the TF-IDF values of the extracted keywords in the texture, era and purpose text data are calculated;
[0039] The consistency of the appreciation reports was calculated based on the similarity of the keywords in the texture, era and purpose text data in different appreciation reports, as well as the TF-IDF values of the keywords in the texture, era and purpose text data.
[0040] As a further solution of the present invention: according to the similarity of keywords in the texture, era and purpose text data in different appreciation reports, and the TF-IDF value of the keywords in the texture, era and purpose text data, the consistency of the appreciation reports is calculated, including the following steps:
[0041] According to the similarity of keywords in the texture, era and use text data in different appreciation reports, and the TF-IDF value of keywords in the texture, era and use text data, the consistency of the texture, era and use text data in different appreciation reports is calculated respectively;
[0042] The consistency of the text data on texture, era, and usage in different appreciation reports was calculated using the following formula:
[0043]
[0044] Among them, S is the consistency evaluation value of the texture, era or usage text data in different appreciation reports, R i is the TF-IDF value of the i-th keyword in the texture, era or purpose text data, W i is the similarity of the ith keyword in the texture, era or purpose text data, i∈(1, 2, …, n), n is the total number of keywords in the group;
[0045] According to the consistency of the text data of texture, era and use in different appreciation reports, the consistency of the appreciation reports is calculated by the following formula:
[0046] K=0.4*S1+0.4*S2+0.2*S3
[0047] Among them, K is the consistency evaluation value of the appreciation report, S1 is the consistency evaluation value of the texture data in the appreciation report, S2 is the consistency evaluation value of the age data in the appreciation report, and S3 is the consistency evaluation value of the usage text data in the appreciation report.
[0048] As a further solution of the present invention: screening the appreciation report with the highest degree of uniformity among all appreciation reports, and generating a final appreciation report according to the texture, era and purpose text data of the appreciation report with the highest degree of uniformity, including the following steps:
[0049] All appreciation reports were reviewed twice, and textual data on the texture, age, and use of the collections in each appreciation report were extracted;
[0050] Calculate the consistency evaluation value of each appreciation report with all other appreciation reports in turn, remove the maximum and minimum values of the obtained consistency evaluation value of the appreciation report with all other appreciation reports, calculate the average value of the consistency evaluation value of the remaining appreciation report with all other appreciation reports, and use the obtained value as the uniformity evaluation value of the corresponding appreciation report;
[0051] The appreciation reports with the highest uniformity evaluation values are screened, and the final appreciation report is generated based on the texture, age and usage text data of the collections in the appreciation reports with the highest uniformity evaluation values.
[0052] As another aspect of the present invention: a management system for an online collection communication and appreciation platform, comprising:
[0053] Request release module: users post collection appreciation requests on the platform, submit collection identity data, and choose to have the collection appreciated by mail or by picture;
[0054] Quick Appreciation Module: When a user chooses to appreciate an image, the platform directly sends the collection identity data to the appreciation expert’s mailbox, and the appreciation expert issues an appreciation report. The platform then sends the appreciation report to the user to complete the collection appreciation request;
[0055] Physical appreciation module, including:
[0056] Physical mailing management unit: When a user chooses to have a physical object mailed for appreciation, the platform generates a unique identifier for the collection, and generates a smart contract on the blockchain to record the collection identity data and request status. After the platform receives the physical object, it takes another photo of the collection unpacking, uploads it to the smart contract, and updates the request status;
[0057] Expert allocation unit: Generates collection classification labels for collections based on collection identity data, and allocates appreciation experts to different appreciation groups based on personal appreciation experience information submitted by appreciation experts;
[0058] Appreciation report generation unit: according to the collection classification information, the collection is assigned to different appreciation groups, and the appreciation experts appreciate the collection and generate appreciation reports, evaluate the consistency of the appreciation reports, and perform a second appreciation of the collection based on the consistency evaluation results of the appreciation reports, and generate a final appreciation report;
[0059] Appreciation report synchronization unit: upload the final appreciation report to the blockchain, generate a unique identifier and timestamp for the appreciation report, and match the identifier of the appreciation report with the unique identifier of the collection to complete the collection appreciation request.
[0060] Compared with the prior art, the present invention has the following beneficial effects:
[0061] The present invention allows users to publish collection appreciation requests on the platform, submit collection identity data, and choose to conduct physical mail appreciation or picture appreciation, so that users can choose a suitable appreciation method; picture appreciation is relatively fast, but there is a certain degree of uncertainty, so the platform sends the appreciation report to the user end to complete the collection appreciation request, and there is no need to synchronize the data to the blockchain, which is convenient for quickly sending the appreciation results to the user and also reduces the data that needs to be uploaded to the blockchain; at the same time, when performing physical mail appreciation, it is necessary to ensure that the physical object received is consistent with the one sent by the user. After the platform receives the physical collection, it takes a picture of the collection unpacking again and uploads it to the smart contract, which is convenient for the platform to judge the consistency between the collection and the one sent by the user and leave a record in the smart contract.
[0062] The present invention ensures that each collection receives a professional evaluation and report by assigning the collection to a professional appreciation group and appreciation experts for appreciation, which helps to reduce subjective factors and improve the accuracy and authority of the appreciation results; at the same time, the consistency evaluation of the appreciation report facilitates the discovery of possible evaluation differences or errors, thereby improving the overall appreciation quality and consistency. Through secondary appreciation and comprehensive appreciation reports, it is convenient to examine the collection from the perspectives of multiple experts, integrate the opinions and evaluations of different experts, and draw more comprehensive and objective conclusions. Through this systematic appreciation process, the credibility and transparency of the appreciation report are enhanced. BRIEF DESCRIPTION OF THE DRAWINGS
[0063] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0064] Figure 1 A schematic diagram of the process of the physical object mailing appreciation method of the present invention;
[0065] Figure 2 It is a schematic diagram of the system framework of the present invention. DETAILED DESCRIPTION
[0066] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0067] See also Figure 1-Figure 2 The first embodiment of the present invention provides a management method and system for an online collection communication and appreciation platform, comprising the following steps:
[0068] Users post a collection appreciation request on the platform, submit the collection identity data, and choose to have the collection appreciated by mail or by picture;
[0069] When the user chooses to appreciate the picture, the platform directly sends the collection identity data to the appreciation expert's mailbox, and the appreciation expert issues an appreciation report. The platform sends the appreciation report to the user end to complete the collection appreciation request;
[0070] When the user chooses to have the item mailed for appreciation, the platform generates a unique identifier for the item and generates a smart contract on the blockchain to record the item's identity data and request status. After the platform receives the item, it takes another unpacking photo of the item, uploads it to the smart contract, and updates the request status.
[0071] Generate collection classification labels for collections based on collection identity data, and assign appreciation experts to different appreciation groups based on their personal appreciation experience information submitted by them;
[0072] According to the collection classification information, the collections are assigned to different appreciation groups. The appreciation experts will appreciate the collections and generate appreciation reports, evaluate the consistency of the appreciation reports, and conduct a second appreciation of the collections based on the consistency evaluation results of the appreciation reports, and generate a final appreciation report;
[0073] The final appreciation report is uploaded to the blockchain, and a unique identifier and timestamp are generated for the appreciation report. The identifier of the appreciation report is matched with the unique identifier of the collection to complete the collection appreciation request.
[0074] Specifically, in this embodiment, the user publishes a collection appreciation request on the platform, submits the collection identity data, and chooses to appreciate the collection by mail or by picture, so that the user can choose a suitable appreciation method.
[0075] When the user chooses to appreciate a picture, the platform directly sends the identity data of the collection to the appreciation expert's mailbox, and the appreciation expert issues an appreciation report. The platform sends the appreciation report to the user end to complete the collection appreciation request. Picture appreciation is relatively fast, but there is a certain degree of uncertainty. Therefore, the platform sends the appreciation report to the user end to complete the collection appreciation request. There is no need to synchronize the data to the blockchain, which facilitates the rapid sending of the appreciation results to the user and reduces the data that needs to be uploaded to the blockchain.
[0076] When the user chooses to have the item mailed for appreciation, the platform generates a unique identifier for the collection and creates a smart contract on the blockchain to record the collection's identity data and request status. After the platform receives the item, it takes another photo of the item being unpacked and uploads it to the smart contract, and updates the request status. When appreciating the item by mail, it is necessary to ensure that the item received is consistent with what the user sent. After the platform receives the item, it takes another photo of the item being unpacked and uploads it to the smart contract, so that the platform can judge the consistency between the item and what the user sent, and leave a record in the smart contract.
[0077] Based on the identity data of the collection, a collection classification label is generated for the collection, and based on the personal appreciation experience information submitted by the appreciation experts, the appreciation experts are assigned to different appreciation groups. This facilitates the assignment of the collection to multiple matching appreciation expert groups for appreciation, avoiding negligence in personal appreciation, which may lead to excessive errors in the appreciation results.
[0078] According to the collection classification information, the collection is assigned to different appreciation groups. The appreciation experts will appreciate the collection and generate appreciation reports, evaluate the consistency of the appreciation reports, and perform secondary appreciation on the collections according to the consistency evaluation results of the appreciation reports, and generate the final appreciation reports. By assigning the collections to professional appreciation groups and appreciation experts for appreciation, it is ensured that each collection receives professional level evaluation and reports, which helps to reduce subjective factors and improve the accuracy and authority of the appreciation results. At the same time, the consistency evaluation of the appreciation reports facilitates the discovery of possible evaluation differences or errors, thereby improving the overall appreciation quality and consistency. Through secondary appreciation and comprehensive appreciation reports, it is convenient to examine the collections from the perspectives of multiple experts, integrate the opinions and evaluations of different experts, and draw more comprehensive and objective conclusions. Through this systematic appreciation process, the credibility and transparency of the appreciation reports are enhanced.
[0079] The collection appreciation request is completed by uploading the final appreciation report to the blockchain, generating a unique identifier and timestamp for the appreciation report, and matching the appreciation report identifier with the collection’s unique identifier; blockchain technology ensures that the appreciation report’s information is tamper-proof and transparent. Once uploaded to the blockchain, no one can modify or delete the information, ensuring the integrity and credibility of the appreciation report; matching the appreciation report identifier with the collection’s unique identifier allows the inquirer to view and verify the source and history of the appreciation report, thereby enhancing the credibility and transparency of the appreciation report. Blockchain is a decentralized platform that is not under single control, which means that there is no single bottleneck or risk in the storage and transmission of appreciation reports, improving data security and reliability.
[0080] In one embodiment of the present invention, the collection identity data includes: the name, material, age, collection method and six views of the collection.
[0081] In one embodiment of the present invention, a collection classification label is generated for the collection according to the collection identity data, and the appreciation experts are assigned to different appreciation groups according to the personal appreciation experience information submitted by the appreciation experts, including the following steps:
[0082] Based on the collection identity data, the collection is classified according to its texture, age and purpose, and collection classification labels are generated for the collection;
[0083] Based on the personal appreciation experience information provided by the appreciation experts when registering on the platform, the appreciation experts will be assigned to different appreciation groups according to the texture, age and purpose of the collections that they appreciate most.
[0084] In one embodiment of the present invention, based on the collection identity data, the collection is classified according to the texture, age and purpose of the collection, and the collection classification label is generated for the collection, including the following steps:
[0085] Collect a large number of image pairs of different textures, ages and uses as training sets. The image pairs include: image pairs representing the same collection and image pairs representing different collections. Each image pair is labeled to indicate whether it is from the same collection.
[0086] The image pairs in the training set are input into the two sub-networks of the Siamese Network model respectively, and the Siamese Network model is trained to identify whether the image pairs are images of the same collection;
[0087] Obtain the six-view data in the collection identity data, as well as the collection image information taken again after the platform receives the collection in person;
[0088] Select the front view image data of the six-view data in the collection identity data and the front view image data in the collection image information taken again after the platform receives the collection in person, to form an image pair;
[0089] Input the image pairs of the collection into the trained Siamese Network model to identify whether the image pairs are images of the same collection;
[0090] If the image information of the collection taken again after the recognition platform receives the physical collection is different from the six-view data in the collection identity data, the collection image will be taken again, and the recognition result of the Siamese Network model and the re-taken collection image will be uploaded to the smart contract of the blockchain, and the collection will be returned to the user;
[0091] If the image information of the collection taken again after the recognition platform receives the collection and the six-view data in the collection identity data are the same picture of the collection, the user will continue to publish the collection appreciation request on the platform;
[0092] Get the name, material, age and collection method data of the collection;
[0093] The TextRank algorithm is used to extract keywords from the name, material, and age information of the collection;
[0094] The collection is set according to the keywords in the material information to generate the texture classification label of the collection, the collection is set according to the keywords in the age information to generate the era classification label of the collection, and the collection is set according to the keywords in the name information to generate the use classification label of the collection.
[0095] In one embodiment of the present invention, according to the personal appreciation experience information provided by the appreciation experts when registering on the platform, the appreciation experts are assigned to different appreciation groups according to the texture, age and purpose of the collections most appreciated by the appreciation experts, including the following steps:
[0096] According to the personal appreciation experience information provided by the appreciation experts when registering on the platform, the texture, age and use information of the collections appraised by the experts are obtained, and the texture, age and use information of the collections most appreciated by the appreciation experts are counted;
[0097] The TextRank algorithm is used to extract keywords from the texture, age, and usage information of the collections that are most appreciated by the experts;
[0098] According to the extracted keywords of collection texture information, the appreciation experts are grouped into collection texture groups; according to the extracted keywords of collection age information, the appreciation experts are grouped into collection age groups; according to the extracted keywords of collection usage information, the appreciation experts are grouped into collection usage groups.
[0099] In one embodiment of the present invention, according to the collection classification information, the collection is assigned to different appreciation groups, and appreciation experts appreciate the collection and generate appreciation reports, evaluate the consistency of the appreciation reports, and perform secondary appreciation on the collection according to the consistency evaluation results of the appreciation reports, and generate a final appreciation report, including the following steps:
[0100] According to the collection classification information, the collections are assigned to different appreciation groups, and appreciation experts are randomly selected in each appreciation group to appreciate the collections and generate appreciation reports;
[0101] Generate an appreciation report for each appreciation group, extract the texture, age and use text data of the collection, and calculate the consistency of the appreciation report by comparing the texture, age and use text data of the collection in the appreciation report generated by each appreciation group. If the consistency of the appreciation report is lower than the preset threshold, select appreciation experts from the appreciation group corresponding to the collection again, conduct a second appreciation, and generate an appreciation report; otherwise, no second appraisal will be conducted;
[0102] The appreciation reports with the highest degree of uniformity among all the appreciation reports are screened, and the final appreciation report is generated according to the texture, era and usage text data of the appreciation reports with the highest degree of uniformity.
[0103] Furthermore, in this embodiment, if the consistency of the appreciation report is lower than a preset threshold, appreciation experts are screened again from the appreciation group corresponding to the collection, a secondary appreciation is conducted, and an appreciation report is generated; otherwise, no secondary appraisal is conducted, and finally the appreciation report with the highest degree of uniformity among all the appreciation reports is screened, and the final appreciation report is generated based on the texture, era and use text data of the appreciation report with the highest degree of uniformity. When the collection is difficult to appreciate and the uncertainty is high, the appreciation of the collection is increased and more appreciation reports are generated, which is conducive to selecting a more accurate appreciation report. As for the setting of the consistency threshold of the appreciation report, the consistency evaluation value data of the appreciation report in the historical data can be used to select a value in the top 10% to 20% of the consistency evaluation value data of the appreciation report as the preset threshold of the consistency evaluation value.
[0104] In one embodiment of the present invention, an appreciation report is generated for each appreciation group, and the texture, age and use text data of the collection are extracted. By comparing the texture, age and use text data of the collection in the appreciation report generated by each appreciation group, the consistency of the appreciation report is calculated, including the following steps:
[0105] Obtain the appreciation report generated by each appreciation group, extract the texture, age and use text data of the collection, and extract the keywords of the texture, age and use text data of the collection through the TextRank algorithm;
[0106] Through the Word2Vec language model, the keywords are mapped into a high-dimensional space, and the similarity of the keywords in the texture, era and usage text data in different appreciation reports is calculated;
[0107] By using the TF-IDF algorithm, the TF-IDF values of the extracted keywords in the texture, era and purpose text data are calculated;
[0108] The consistency of the appreciation reports was calculated based on the similarity of the keywords in the texture, era and purpose text data in different appreciation reports, as well as the TF-IDF values of the keywords in the texture, era and purpose text data.
[0109] In one embodiment of the present invention, the consistency of the appreciation reports is calculated based on the similarity of keywords in the texture, era and purpose text data in different appreciation reports, and the TF-IDF value of the keywords in the texture, era and purpose text data, including the following steps:
[0110] According to the similarity of keywords in the texture, era and use text data in different appreciation reports, and the TF-IDF value of keywords in the texture, era and use text data, the consistency of the texture, era and use text data in different appreciation reports is calculated respectively;
[0111] The consistency of the text data on texture, era, and usage in different appreciation reports was calculated using the following formula:
[0112]
[0113] Among them, S is the consistency evaluation value of the texture, era or usage text data in different appreciation reports, R i is the TF-IDF value of the i-th keyword in the texture, era or purpose text data, W i is the similarity of the ith keyword in the texture, era or purpose text data, i∈(1, 2, …, n), n is the total number of keywords in the group;
[0114] According to the consistency of the text data of texture, era and use in different appreciation reports, the consistency of the appreciation reports is calculated by the following formula:
[0115] K=0.4*S1+0.4*S2+0.2*S3
[0116] Among them, K is the consistency evaluation value of the appreciation report, S1 is the consistency evaluation value of the texture data in the appreciation report, S2 is the consistency evaluation value of the age data in the appreciation report, and S3 is the consistency evaluation value of the usage text data in the appreciation report.
[0117] In one embodiment of the present invention, the appreciation report with the highest degree of uniformity is screened out from all appreciation reports, and a final appreciation report is generated according to the texture, age and purpose text data of the appreciation report with the highest degree of uniformity, including the following steps:
[0118] All appreciation reports were reviewed twice, and textual data on the texture, age, and use of the collections in each appreciation report were extracted;
[0119] Calculate the consistency evaluation value of each appreciation report with all other appreciation reports in turn, remove the maximum and minimum values of the obtained consistency evaluation value of the appreciation report with all other appreciation reports, calculate the average value of the consistency evaluation value of the remaining appreciation report with all other appreciation reports, and use the obtained value as the uniformity evaluation value of the corresponding appreciation report;
[0120] The appreciation reports with the highest uniformity evaluation values are screened, and the final appreciation report is generated based on the texture, age and usage text data of the collections in the appreciation reports with the highest uniformity evaluation values.
[0121] In another embodiment of the present invention, a management system for an online collection communication and appreciation platform is provided, including:
[0122] Request release module: users post collection appreciation requests on the platform, submit collection identity data, and choose to have the collection appreciated by mail or by picture;
[0123] Quick Appreciation Module: When a user chooses to appreciate an image, the platform directly sends the collection identity data to the appreciation expert’s mailbox, and the appreciation expert issues an appreciation report. The platform then sends the appreciation report to the user to complete the collection appreciation request;
[0124] Physical appreciation module, including:
[0125] Physical mailing management unit: When a user chooses to have a physical object mailed for appreciation, the platform generates a unique identifier for the collection, and generates a smart contract on the blockchain to record the collection identity data and request status. After the platform receives the physical object, it takes another photo of the collection unpacking, uploads it to the smart contract, and updates the request status;
[0126] Expert allocation unit: Generates collection classification labels for collections based on collection identity data, and allocates appreciation experts to different appreciation groups based on personal appreciation experience information submitted by appreciation experts;
[0127] Appreciation report generation unit: according to the collection classification information, the collection is assigned to different appreciation groups, and the appreciation experts appreciate the collection and generate appreciation reports, evaluate the consistency of the appreciation reports, and perform a second appreciation of the collection based on the consistency evaluation results of the appreciation reports, and generate a final appreciation report;
[0128] Appreciation report synchronization unit: upload the final appreciation report to the blockchain, generate a unique identifier and timestamp for the appreciation report, and match the identifier of the appreciation report with the unique identifier of the collection to complete the collection appreciation request.
[0129] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.
Claims
1. A management method for an online collection communication and appreciation platform, characterized in that: The following steps are involved: Users post a collection appreciation request on the platform, submit the collection identity data, and choose to have the collection appreciated by mail or by picture; When the user chooses to appreciate the picture, the platform directly sends the collection identity data to the appreciation expert's mailbox, and the appreciation expert issues an appreciation report. The platform sends the appreciation report to the user end to complete the collection appreciation request; When the user chooses to have the item mailed for appreciation, the platform generates a unique identifier for the item and generates a smart contract on the blockchain to record the item's identity data and request status. After the platform receives the item, it takes another unpacking photo of the item, uploads it to the smart contract, and updates the request status. Generate collection classification labels for collections based on collection identity data, and assign appreciation experts to different appreciation groups based on their personal appreciation experience information submitted by them; According to the collection classification information, the collections are assigned to different appreciation groups. The appreciation experts appreciate the collections and generate appreciation reports. The consistency of the appreciation reports is evaluated. According to the consistency evaluation results of the appreciation reports, the collections are appreciated again and a final appreciation report is generated. The final appreciation report is uploaded to the blockchain, and a unique identifier and timestamp are generated for the appreciation report. The identifier of the appreciation report is matched with the unique identifier of the collection to complete the collection appreciation request.
2. The management method of the online collection communication and appreciation platform according to claim 1 is characterized in that: The collection identity data includes: the name, material, age, collection method and six views of the collection.
3. The management method of the online collection communication and appreciation platform according to claim 2 is characterized in that: Generate collection classification labels for collections based on collection identity data, and assign appreciation experts to different appreciation groups based on personal appreciation experience information submitted by appreciation experts, including the following steps: Based on the collection identity data, the collection is classified according to its texture, age and purpose, and collection classification labels are generated for the collection; Based on the personal appreciation experience information provided by the appreciation experts when registering on the platform, the appreciation experts will be assigned to different appreciation groups according to the texture, age and purpose of the collections that they appreciate most.
4. The management method of the online collection communication and appreciation platform according to claim 3 is characterized in that: Based on the collection identity data, the collection is classified according to its texture, age and purpose, and collection classification labels are generated for the collection, including the following steps: Collect a large number of image pairs of different textures, ages and uses as training sets. The image pairs include: image pairs representing the same collection and image pairs representing different collections. Each image pair is labeled to indicate whether it is from the same collection. The image pairs in the training set are input into the two sub-networks of the Siamese Network model respectively, and the Siamese Network model is trained to identify whether the image pairs are images of the same collection; Obtain the six-view data in the collection identity data, as well as the collection image information taken again after the platform receives the collection in person; Select the front view image data of the six-view data in the collection identity data and the front view image data in the collection image information taken again after the platform receives the collection in person, to form an image pair; Input the image pairs of the collection into the trained Siamese Network model to identify whether the image pairs are images of the same collection; If the image information of the collection taken again after the recognition platform receives the physical collection is different from the six-view data in the collection identity data, the collection image will be taken again, and the recognition result of the Siamese Network model and the re-taken collection image will be uploaded to the smart contract of the blockchain, and the collection will be returned to the user; If the image information of the collection taken again after the recognition platform receives the collection and the six-view data in the collection identity data are the same picture of the collection, the user will continue to publish the collection appreciation request on the platform; Get the name, material, age and collection method data of the collection; The TextRank algorithm is used to extract keywords from the name, material, and age information of the collection; The collection is set according to the keywords in the material information to generate the texture classification label of the collection, the collection is set according to the keywords in the age information to generate the era classification label of the collection, and the collection is set according to the keywords in the name information to generate the use classification label of the collection.
5. The management method of the online collection communication and appreciation platform according to claim 3 is characterized in that: According to the personal appreciation experience information provided by the appreciation experts when registering on the platform, the appreciation experts are assigned to different appreciation groups according to the texture, age and purpose of the collections that the appreciation experts appreciate the most, including the following steps: According to the personal appreciation experience information provided by the appreciation experts when registering on the platform, the texture, age and use information of the collections appraised by the experts are obtained, and the texture, age and use information of the collections most appreciated by the appreciation experts are counted; The TextRank algorithm is used to extract keywords from the texture, age, and usage information of the collections that are most appreciated by the experts; According to the extracted keywords of collection texture information, the appreciation experts are grouped into collection texture groups; according to the extracted keywords of collection age information, the appreciation experts are grouped into collection age groups; according to the extracted keywords of collection usage information, the appreciation experts are grouped into collection usage groups.
6. The management method of the online collection communication and appreciation platform according to claim 1 is characterized in that: According to the collection classification information, the collection is assigned to different appreciation groups. The appreciation experts appreciate the collection and generate appreciation reports. The consistency of the appreciation reports is evaluated. According to the consistency evaluation results of the appreciation reports, the collection is appreciated again and a final appreciation report is generated, including the following steps: According to the collection classification information, the collections are assigned to different appreciation groups, and appreciation experts are randomly selected in each appreciation group to appreciate the collections and generate appreciation reports; Generate an appreciation report for each appreciation group, extract the texture, age and use text data of the collection, and calculate the consistency of the appreciation report by comparing the texture, age and use text data of the collection in the appreciation report generated by each appreciation group. If the consistency of the appreciation report is lower than the preset threshold, select appreciation experts from the appreciation group corresponding to the collection again, conduct a second appreciation, and generate an appreciation report; otherwise, no second appraisal will be conducted; The appreciation reports with the highest degree of uniformity among all the appreciation reports are screened, and the final appreciation report is generated according to the texture, era and usage text data of the appreciation reports with the highest degree of uniformity.
7. The management method of the online collection communication and appreciation platform according to claim 6 is characterized in that: Generate an appreciation report for each appreciation group, extract the texture, age and use text data of the collection, and calculate the consistency of the appreciation report by comparing the texture, age and use text data of the collection in the appreciation report generated by each appreciation group, including the following steps: Obtain the appreciation report generated by each appreciation group, extract the texture, age and use text data of the collection, and extract the keywords of the texture, age and use text data of the collection through the TextRank algorithm; Through the Word2Vec language model, the keywords are mapped into a high-dimensional space, and the similarity of the keywords in the texture, era and usage text data in different appreciation reports is calculated; By using the TF-IDF algorithm, the TF-IDF values of the extracted keywords in the texture, era and purpose text data are calculated; The consistency of the appreciation reports was calculated based on the similarity of the keywords in the texture, era and purpose text data in different appreciation reports, as well as the TF-IDF values of the keywords in the texture, era and purpose text data.
8. The management method of the online collection communication and appreciation platform according to claim 7 is characterized in that: According to the similarity of keywords in the texture, era and purpose text data in different appreciation reports, and the TF-IDF value of the keywords in the texture, era and purpose text data, the consistency of the appreciation report is calculated, including the following steps: According to the similarity of keywords in the texture, era and use text data in different appreciation reports, and the TF-IDF value of keywords in the texture, era and use text data, the consistency of the texture, era and use text data in different appreciation reports is calculated respectively; The consistency of the text data on texture, era, and usage in different appreciation reports was calculated using the following formula: Among them, S is the consistency evaluation value of the texture, era or usage text data in different appreciation reports, R i is the TF-IDF value of the i-th keyword in the texture, era or purpose text data, W i is the similarity of the ith keyword in the texture, era or purpose text data, i∈(1, 2, …, n), n is the total number of keywords in the group; According to the consistency of the text data of texture, era and use in different appreciation reports, the consistency of the appreciation reports is calculated by the following formula: K=0.4*S1+0.4*S2+0.2*S3 Among them, K is the consistency evaluation value of the appreciation report, S1 is the consistency evaluation value of the texture data in the appreciation report, S2 is the consistency evaluation value of the age data in the appreciation report, and S3 is the consistency evaluation value of the usage text data in the appreciation report.
9. The management method of the online collection communication and appreciation platform according to claim 8 is characterized in that: Screening the most uniform appreciation report among all appreciation reports, and generating a final appreciation report based on the texture, era and usage text data of the most uniform appreciation report, including the following steps: All appreciation reports were reviewed twice, and textual data on the texture, age, and use of the collections in each appreciation report were extracted; Calculate the consistency evaluation value of each appreciation report with all other appreciation reports in turn, remove the maximum and minimum values of the obtained consistency evaluation value of the appreciation report with all other appreciation reports, calculate the average value of the consistency evaluation value of the remaining appreciation report with all other appreciation reports, and use the obtained value as the uniformity evaluation value of the corresponding appreciation report; The appreciation reports with the highest uniformity evaluation values are screened, and the final appreciation report is generated based on the texture, age and usage text data of the collections in the appreciation reports with the highest uniformity evaluation values.
10. A management system for an online collection communication and appreciation platform, characterized in that: include: Request release module: users post collection appreciation requests on the platform, submit collection identity data, and choose to have the collection appreciated by mail or by picture; Quick Appreciation Module: When a user chooses to appreciate an image, the platform directly sends the collection identity data to the appreciation expert’s mailbox, and the appreciation expert issues an appreciation report. The platform then sends the appreciation report to the user to complete the collection appreciation request; Physical appreciation module, including: Physical mailing management unit: When a user chooses to have a physical object mailed for appreciation, the platform generates a unique identifier for the collection, and generates a smart contract on the blockchain to record the collection identity data and request status. After the platform receives the physical object, it takes another photo of the collection unpacking, uploads it to the smart contract, and updates the request status; Expert allocation unit: Generates collection classification labels for collections based on collection identity data, and allocates appreciation experts to different appreciation groups based on personal appreciation experience information submitted by appreciation experts; Appreciation report generation unit: according to the collection classification information, the collection is assigned to different appreciation groups, and the appreciation experts appreciate the collection and generate appreciation reports, evaluate the consistency of the appreciation reports, and perform a second appreciation of the collection based on the consistency evaluation results of the appreciation reports, and generate a final appreciation report; Appreciation report synchronization unit: upload the final appreciation report to the blockchain, generate a unique identifier and timestamp for the appreciation report, and match the identifier of the appreciation report with the unique identifier of the collection to complete the collection appreciation request.
