Intelligent copyright automatic authorization transaction platform system

Abstract

The invention belongs to the technical field of digital copyright management, and discloses an intelligent copyright automatic authorization transaction platform system, which comprises a user management unit used for supporting a user to register and log in on an authorization transaction platform and uploading work information data; the digital copyright confirmation unit is used for carrying out original verification and quality evaluation auditing on the work information data to generate a work digital fingerprint; automatically writing the digital fingerprint of the work into a block chain evidence storage network through an intelligent contract, and generating an NFT digital certificate; the intelligent authorization contract generation unit is internally provided with an intelligent contract template and generates personalized terms in a matching manner according to the digital fingerprints of the works and preset authorization requirements; the personalized terms are converted into standardized intelligent contract codes, and corresponding contract texts are generated at the same time; intermediate links and manual operation are reduced, and the transaction efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of digital rights management, and more specifically, to an intelligent copyright automated authorization trading platform system. Background Art

[0002] A patent with the patent publication number CN108769750A discloses a digital content bank system based on blockchain technology, including a content bank node and several audio-visual platform nodes that constitute the blockchain. The content bank node is connected to a content bank server; the audio-visual platform node is connected to an audio-visual platform server; a first smart contract is deployed on the blockchain. The digital content bank system based on blockchain technology provided by the present invention effectively combines a centralized management system with a decentralized recording system, and realizes the complete automation of the operation of copyright strategies and the process of copyright transactions. By recording and deploying smart contracts on the blockchain, automatic monitoring and execution of copyright contracts are carried out, and copyright transaction records, execution records of smart contracts, etc. are all formed into a blockchain, and reliable storage of data is achieved through distributed ledger technology; it provides strong technical support for copyright maintenance, copyright content authorization, and copyright supervision.

[0003] The existing intelligent copyright automated authorization trading platform systems mainly have the following problems:

[0004] The traditional manual review method is easily affected by the subjective judgment of reviewers, resulting in inconsistent review criteria and inaccurate judgments. Since different reviewers may have different aesthetic or judgment criteria, the same work may receive different evaluations in the hands of different reviewers. This subjectivity may make the verification of originality inaccurate and unreliable; the traditional method usually relies on manual comparison and retrieval, and cannot efficiently process massive data and quickly identify similar works. Manually comparing the similarity of works not only takes a long time, but also is difficult to ensure complete objectivity. For the judgment of similarity between works, the traditional method is often less efficient and accurate than algorithm processing; for some complex works, there may be a large number of details and features that need to be judged. Manual review usually can only rely on experience or intuition to make decisions, and it may not be able to accurately capture some subtle similarities or differences in the works, while algorithms can effectively identify and process complex feature data. Without using algorithms for feature vector analysis, it may not be possible to accurately capture the subtle differences between works. Without introducing algorithm automation processing, there are many links that require manual participation in the traditional copyright review process, which is prone to human errors or deviations. Manual operations are not only error-prone, but also time-consuming, resulting in an extended review cycle for works, reducing the work efficiency and user experience of the platform;

[0005] Traditional auction pricing methods are usually fixed prices or one-time bids, without a dynamic pricing mechanism. This means that the price of the work may not adapt to market changes and cannot be adjusted in real time according to market demand and participants' feedback. Such a fixed price may lead to the price deviating from the true market demand, causing some buyers to give up bidding due to the high price, or the seller being unable to maximize their profits due to the low price. The lack of a pricing mechanism with dynamic adjustment will reduce the attractiveness of the auction and fail to give full play to the role of market supply and demand. The traditional auction method fails to respond to market fluctuations and demand changes in real time. Without using a market price dynamic adjustment formula, the auction price may not follow the changes in the market conditions in a timely manner, thus missing the best trading opportunities. Especially in a rapidly changing market environment, the price may be too lagged or distorted to truly reflect the market value of the work. This will lead to insufficient transparency and fairness in the auction, and even damage the market competitiveness of the platform. The changes in the market are not reflected in the price adjustment in a timely manner, and some buyers may feel that the price setting is unreasonable, leading to trust issues. The lack of a real-time adjustment pricing mechanism also easily leads to doubts about the price transparency among participants, thus affecting the fairness and attractiveness of the platform; in traditional auctions, due to the singularity and fixity of pricing, the needs of both buyers and sellers may not be balanced; the traditional auction method cannot adjust price changes, causing buyers to miss bidding opportunities in the high-price area or miss the bidding timing in the low-price area.

[0006] In view of this, the present invention proposes an intelligent copyright automated authorization trading platform system to solve the above problems. Summary of the Invention

[0007] To overcome the above defects of the prior art and to achieve the above object, the present invention provides the following technical solutions: An intelligent copyright automated authorization trading platform system, comprising:

[0008] A user management unit for supporting users to register and log in on the authorization trading platform and upload work information data;

[0009] A digital copyright confirmation unit for conducting originality verification and quality assessment and review on the work information data to generate a work digital fingerprint; automatically writing the work digital fingerprint into the blockchain evidence storage network through a smart contract to generate an NFT digital certificate;

[0010] An intelligent authorization contract generation unit with built-in smart contract templates, which matches and generates personalized terms according to the work digital fingerprint and preset authorization requirements; converts the personalized terms into standardized smart contract codes and generates corresponding contract texts at the same time;

[0011] An online authorization trading unit for obtaining market data by integrating an oracle network, dynamically pricing using an improved Dutch auction algorithm, and obtaining the contract text after the user determines the price to complete the transaction;

[0012] The copyright protection and monitoring unit is used to monitor the entire network for authorized works and provide an infringement warning mechanism. If unauthorized use is detected, the infringement evidence data will be automatically packaged and an electronic evidence package will be generated to notify the user;

[0013] The decision support unit is used to obtain transaction data, build a digital copyright value assessment model, use transaction data and market data as input to the digital copyright value assessment model, predict the digital copyright value index, and evaluate the market value of the authorized work based on the predicted digital copyright value index, formulate marketing strategies for users, and provide decision-making basis for investors.

[0014] Preferably, the work information data includes text work data, image work data, audio work data, video work data and work attribute data.

[0015] Preferably, the method of verifying the originality and evaluating the quality of the work information data to generate the digital fingerprint of the work includes:

[0016] The work information data is grouped by the K-means clustering algorithm. The number of clusters K is selected according to the number of categories of the work information data. The work feature vectors in the work information data corresponding to the number of clusters K are selected as the initial cluster centers. For each work, the distance from the work to each cluster center is calculated and the work is assigned to the nearest cluster group.

[0017] Update the center point of each cluster to the mean of the feature vectors of all works in the cluster; calculate the distance from each work to the center of its cluster, preset a distance threshold, and if the distance from a certain work to the center of its cluster is less than the preset distance threshold, then the work is determined to be an original work; if the distance from a certain work to the center of its cluster is greater than or equal to the preset distance threshold, then the work is determined to be a non-original work;

[0018] The quality score of each work is calculated using the work quality scoring formula, which is: Among them, qs i is the quality score of work i; f(i) is the feature vector of work i; f K (i) is the feature vector of the i-th work in cluster number K; σ K (i) is the standard deviation of the i-th work in cluster number K; |f(i)-f K (i)| is the absolute difference between the feature vector of work i and the feature vector of the i-th work in cluster number K; n is the number of works in cluster number K; i is the index of the work;

[0019] The feature vectors of each work are hashed through a hashing algorithm. Each work's feature vector is mapped to a unique fixed-length hash value through the SHA-256 hashing function. Using the digital fingerprint generation formula through the SHA-256 hashing function, the work digital fingerprint is obtained; the digital fingerprint generation formula is: Fr(i) = Hash(f(i)); where Fr(i) is the digital fingerprint of work i; Hash(·) is the SHA-256 hashing function.

[0020] Preferably, the method for generating the NFT digital certificate includes:

[0021] Write the smart contract logic using the Solidity programming language, define the functions and rules of the smart contract, and deploy the written smart contract to the blockchain deposit network through the development tool Truffle; through the JavaScript library, connect to the blockchain deposit network and interact with the smart contract, and execute the mint function in the smart contract; mint the NFT through the mint function, associate the work digital fingerprint with the NFT, and generate a unique NFT digital certificate.

[0022] Preferably, the method for matching and generating personalized terms according to the work digital fingerprint and the preset authorization requirements includes:

[0023] The preset authorization requirements include the scope of authorization, the set authorization period, the defined geographical scope of authorization, the defined usage method of the work, the set royalty rate to be paid by the licensee, and the compensation terms after default.

[0024] Preset a smart contract template, use natural language processing technology to analyze the work digital fingerprint and the preset authorization requirements, extract relevant clause content from the preset smart contract template, cluster the work digital fingerprint through the density clustering algorithm, extract relevant clauses from the preset smart contract template, and generate personalized terms that match the work digital fingerprint and the preset authorization requirements through the cosine similarity algorithm.

[0025] Preferably, the method for obtaining market data through the integrated oracle network includes:

[0026] Utilize Oracle oracle technology to connect the blockchain deposit network with external data sources, and obtain market data from the financial market. The market data includes the search volume of different works, the view volume of different works, the download volume of different works, the market price, the trading frequency, the authorization fee, the market growth rate, the pricing of similar works, and the royalty rate.

[0027] Preferably, the method for dynamic pricing using the improved Dutch auction algorithm includes:

[0028] An online authorization trading unit is used to obtain market data by integrating an oracle network, adopt an improved Dutch auction algorithm for dynamic pricing, and after the user determines the price to complete the transaction, obtain the contract text;

[0029] It is preset that the market price obtained at time point t is P ma (t), and the market price change amount between time point t and time point t + 1 is ΔP ma It is preset that the initial auction price of the original work on the authorization trading platform is P0, and the price reduction step size for each time interval is ΔP ac , and make real-time adjustment according to the market price P ma (t) through the dynamic price adjustment formula;

[0030] The dynamic price adjustment formula is: Among them, P ac (t is the auction price of the original work at time point t; α is the market price weight factor; Δt is the time interval; is the integer quotient of time point t and time interval Δt; is the floor function symbol;

[0031] Dynamically adjust the market price weight factor through the market price weight adjustment formula. The market price weight adjustment formula is: Among them, α(t) is the dynamically adjusted market price weight factor; e -λ·t represents the time decay term of the market price weight factor; λ represents the time decay factor; N(t represents the number of buyers and sellers participating in the auction at time point t;

[0032] Within each time period, the auction price of the original work gradually decreases over time. When the auction price P of the original work determined by the bidder at time point t1 ac (t1), the transaction is concluded, and the fund escrow process of the fund escrow smart contract is triggered; It is preset that the transaction occurs at time point t1, and the seller and the buyer are A and B respectively;

[0033] The fund escrow process is: F es (t1) = Es(P ac (t1), A, B); Among them, Es(P ac (t1), A, B) represents the fund escrow smart contract; F es (t1) represents the value of the fund escrow smart contract triggered when the transaction is concluded at time point t1; P ac (t1) represents the auction price of the original work determined at time point t1;

[0034] Use cross-chain bridging technology to conduct settlement in different currencies through the cross-chain settlement formula. It is set that the cryptocurrency paid by the buyer is Cbr , the preset exchange rate at settlement is P se ;

[0035] The cross-chain settlement formula is: P se (C br ) = P ac (t1)·P se (C br , ke); where P se (C br , ke) represents the exchange rate between the cryptocurrency C paid by the buyer at settlement and the native currency of the authorized trading platform; ke represents the native currency of the authorized trading platform; br When the cryptocurrency paid by the buyer is converted into the native currency of the authorized trading platform, the smart contract will automatically trigger a fund transfer to send the funds from the escrow account of the authorized trading platform to the seller.

[0036] Preferably, the method for automatically encapsulating infringement evidence data if unauthorized use is detected includes:

[0037] Using the Scrapy web scraping tool to monitor all usage related to the authorized work on the network in real time; the usage includes whether the authorized work is cited, whether the authorized work is downloaded by unauthorized users, whether the authorized work is recreated by unauthorized users, and whether the authorized work is spread by unauthorized users; if unauthorized use is detected, compare the digital fingerprint of the unauthorized work with the digital fingerprint of the authorized work to confirm whether infringement is constituted; if infringement is constituted, automatically encapsulate the infringement evidence data; the infringement evidence data includes the storage location of the infringing work, the web page screenshot of the infringing work, the video of the infringing work, the time when the infringement event occurred, and the ID of the infringing user.

[0038] Preferably, the method for constructing the digital copyright value evaluation model includes:

[0039] Dividing the data set into a training set and a test set to construct a digital copyright value evaluation model; the sample set is a subset of the data set, and each sample set includes historical transaction data, market data, and the corresponding digital copyright value index; the input data of the model is historical transaction data and market data; the output label is the digital copyright value index; the digital copyright value evaluation model is a random forest regressor model;

[0040]

[0041] ​Initialize the digital copyright value evaluation model and set the hyperparameters of the number and depth of the tree; use the training set to train the digital copyright value evaluation model, use the k-fold cross-validation method to adjust the hyperparameters of the model, and optimize the initially set parameters; use the test set data to evaluate the performance of the model, and use the coefficient of determination to evaluate and calculate the difference between the prediction result and the true label.

[0042] According to the model performance feedback, adjust the hyperparameters of the number and depth of the tree, optimize the model, retrain the model using the adjusted hyperparameters, and stop when the training reaches the preset model complexity to obtain the finally trained digital copyright value evaluation model; use the trained digital copyright value evaluation model to predict the current transaction data and market data, and predict the digital copyright value index.

[0043] Preferably, the method for evaluating the market value of the authorized work according to the predicted digital copyright value index includes:

[0044] Preset the first threshold and the second threshold of the digital copyright value index, compare the predicted digital copyright value index with the preset first threshold and the preset second threshold of the digital copyright value index respectively, and evaluate the market value of the authorized work.

[0045] If the predicted digital copyright value index is less than the preset first threshold of the digital copyright value index, it is determined that the market value of the authorized work is at a low level.

[0046] If the predicted digital copyright value index is greater than or equal to the preset first threshold of the digital copyright value index and less than or equal to the preset second threshold of the digital copyright value index, it is determined that the market value of the authorized work is at a medium level.

[0047] If the predicted digital copyright value index is greater than the preset second threshold of the digital copyright value index, it is determined that the market value of the authorized work is at a high level.

[0048] The technical effects and advantages of the intelligent copyright automated authorization trading platform system of the present invention:

[0049] The use of the K-means clustering algorithm in this invention enables efficient originality verification. By clustering works based on feature vectors, works similar or different from known works can be quickly identified, thus automatically judging their originality. The clustering algorithm avoids the subjectivity of manual review and can process a large amount of data, making it suitable for large-scale work verification. By presetting a distance threshold to determine whether a work is original, the error of manual judgment is avoided, making the judgment criteria more unified and improving the accuracy of verification; through automated clustering, updating of the central point, and distance calculation, the process of copyright review is greatly simplified and the efficiency is improved. Without manual intervention, the automatic processing of work features by the algorithm reduces the possibility of misjudgment or human intervention. The introduction of the work quality scoring formula provides a more objective and scientific work evaluation standard by calculating the quality score of each work, avoiding the limitations of relying solely on manual subjective judgment; the design of the work quality score is based on the calculation and comparison of work feature vectors, combined with the standard deviation and feature difference, which can objectively evaluate the quality of works, especially when dealing with a large amount of work information data. The use of the scoring formula can automatically evaluate the similarity between works and identify works with poor quality or those that do not meet the requirements; the digital fingerprint, as the unique identifier of a work, can not only be used for work originality verification, but also for various purposes such as copyright protection and work traceability, enhancing the security and effectiveness of copyright protection; through the K-means clustering algorithm and the quality scoring formula, more refined review of works can be achieved. Especially in the case of a large variety of work categories, by calculating the similarity of works within the cluster, it is possible to more accurately determine whether a certain work meets the requirements of the platform;

[0050] The modified Dutch auction algorithm allows the auction price of original works to gradually decrease over time. This dynamic pricing mechanism can effectively attract buyers with different needs to participate in the bidding. As time goes by, the price of the work gradually decreases, and buyers can choose the best time to buy according to their budget and needs, which improves the flexibility and attractiveness of the auction. This method adopts a market price dynamic adjustment formula to ensure that the auction price is adjusted according to real-time market fluctuations, making the pricing more in line with market demand and actual market conditions, and avoiding price deviations that may be caused by fixed pricing; the dynamic adjustment of the market price weight factor makes the impact of market supply and demand on the auction price more direct and timely, and improves the transparency and fairness of the price mechanism; the auction mechanism adjusts the dynamic impact of prices and participants through time decay factors and market price weight factors, and can adjust the auction strategy according to the popularity of bidding, the number of participants and market changes. This mechanism can not only balance the needs of buyers and sellers, but also complete transactions at the right time, thereby improving the transaction rate and user satisfaction. The strategy of gradually reducing prices within time intervals allows buyers to bid at different prices at different time points, reducing the unfairness of a single price to buyers and promoting more active bidding activities; the use of cross-chain bridging technology enables multi-currency settlement, allowing buyers to use different cryptocurrencies for payment, greatly improving the flexibility and acceptability of the platform. This can not only meet the payment needs of different users, but also expand the platform's international market and attract global users. The cross-chain settlement formula ensures accurate settlement of transactions through exchange rate conversion, avoids exchange problems between different cryptocurrencies, and improves transaction liquidity and the convenience of cross-border transactions. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 A schematic diagram of the structure of an intelligent copyright automated authorization trading platform system of the present invention;

[0052] Figure 2 A schematic diagram of a method flow of an intelligent copyright automated authorization trading platform of the present invention;

[0053] Figure 3 A flow chart of the method for evaluating the market value of authorized works provided by the present invention. DETAILED DESCRIPTION

[0054] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not 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.

[0055] Embodiment 1

[0056] See also Figure 1 As shown, this embodiment 1 further illustrates an intelligent copyright automated authorization transaction platform system proposed by the present invention, including:

[0057] As the digital economy is booming, the scale of the copyright industry continues to expand, and a large number of digital works such as text, images, audio, and video are emerging, and copyright trading activities are becoming more frequent. As a key support for promoting the efficient operation of the copyright industry, the importance of the intelligent copyright automated authorization trading platform is increasing day by day. However, the existing intelligent copyright automated authorization trading platform system has many problems that need to be solved, which seriously restricts the smooth progress of copyright transactions and the healthy development of the industry.

[0058] The existing technology still faces the following problems:

[0059] Inconsistent review standards: Traditional manual review methods are easily affected by subjective factors of reviewers. Different reviewers have significant differences in aesthetic concepts, professional knowledge reserves, and judgment standards, which means that the same work may receive completely different evaluation results from different reviewers. Taking literary works as an example, some reviewers pay more attention to the literary quality of the work, while others pay more attention to the innovation of the story. This subjectivity leads to a significant reduction in the accuracy and reliability of originality verification, and cannot provide solid protection for copyright confirmation.

[0060] Low audit efficiency: Manual comparison and retrieval of similarities between works not only consumes a lot of time and manpower, but also has difficulty coping with massive amounts of work data. When faced with a large number of works, manual audits often seem to be overwhelmed and it is difficult to ensure the comprehensiveness and objectivity of the audit. For some works with complex details, manual audits may not be able to accurately capture the subtle differences between the works, and it is easy to miss infringing works, which gives infringements an opportunity to take advantage and damages the legitimate rights and interests of copyright owners.

[0061] Lack of automated processing: In the traditional copyright review process, there are too many manual steps involved, which not only increases the probability of human error, but also prolongs the review cycle of the work. During the review process, manual operations may result in data entry errors, omissions in the review process, and other problems, affecting the platform's work efficiency and user experience. After users upload their works, they may have to wait for a long time to get the review results, which is undoubtedly a huge time cost for users who are in urgent need of copyright authorization.

[0062] Lack of dynamic price adjustment: Traditional auction pricing methods usually adopt fixed prices or one-time bidding models, and are unable to flexibly adjust the work prices according to real-time market changes. In a rapidly changing market environment, the market demand and value of works may change at any time. Fixed prices may cause the work prices to deviate from the real market demand. When the market demand for a certain type of work suddenly increases, the fixed price may prevent the seller from obtaining due returns; when the market demand decreases, the excessively high fixed price will also deter buyers and reduce the success rate of transactions.

[0063] Unable to balance supply and demand: Due to the singularity and fixity of pricing, traditional auction methods are difficult to balance the needs of both buyers and sellers. In the high-price area, buyers may give up bidding due to the high price; in the low-price area, sellers may be reluctant to sell due to low returns. This supply-demand imbalance not only reduces the attractiveness of auctions but also may lead to misallocation of market resources and affect the normal order of the copyright trading market.

[0064] Poor market adaptability: Traditional auction methods fail to respond promptly to market fluctuations and demand changes, making auction prices unable to truly reflect the market value of works. When the market conditions change rapidly, auction prices may be too lagged or distorted to provide accurate price references for both buyers and sellers. This not only affects the fairness and transparency of transactions but also may lead to a decline in the trust of participants in the platform, weakening the platform's market competitiveness.

[0065] Non-standard contract terms: In the process of copyright trading, the standardization and integrity of contract terms are crucial. However, traditional copyright trading contracts often rely on manual drafting, which is prone to problems such as vague terms and loopholes. The expressions of key terms such as the scope of authorization and the term of authorization may not be clear enough, which lays a hidden danger for subsequent copyright disputes. Once a dispute occurs, both parties may have disputes due to inconsistent understandings of the contract terms, increasing the difficulty and cost of safeguarding rights.

[0066] Lack of an effective fund escrow mechanism: Traditional copyright trading platforms have deficiencies in fund escrow and cannot fully guarantee the fund safety of both parties to the transaction. During the transaction process, there may be risks in the transfer of funds, such as misappropriation of funds and delayed receipt of funds. For the seller, the inability to receive payments in a timely manner will affect their creative enthusiasm; for the buyer, the lack of guarantee of fund safety may cause concerns about the transaction and hinder the smooth progress of the transaction.

[0067] Insufficient copyright protection and monitoring: After the copyright transaction is completed, the copyright protection and monitoring of the authorized works are insufficient. Traditional methods are difficult to achieve real-time monitoring of the entire network and cannot promptly detect unauthorized uses. Even if an infringement is discovered, it may be difficult to safeguard rights due to untimely and incomplete evidence collection. This fails to effectively protect the rights and interests of copyright owners and affects the healthy development of the copyright trading market.

[0068] Inaccurate value assessment: Existing copyright value assessment methods are often not scientific and accurate enough, and it is difficult to comprehensively consider various factors affecting copyright value. During the assessment process, it may only rely on the surface characteristics of the work or simple market data, and cannot deeply analyze the potential value and market prospects of the work. This results in a deviation between the assessment result and the actual value of the work, and cannot provide reliable decision-making basis for users and investors.

[0069] Difficulties in marketing and investment decisions: Due to the lack of accurate copyright value assessment and market analysis, users often lack direction when formulating marketing strategies and are difficult to develop effective promotion plans according to the characteristics of the work and market demand. Investors also face greater risks when making copyright investments and cannot accurately judge the return rate and risk level of the investment. This not only affects the copyright operation efficiency of users, but also limits the investment vitality and development potential of the copyright industry.

[0070] To effectively solve the above problems, the present invention proposes an intelligent copyright automated authorization trading platform system, including:

[0071] A user management unit, which is used to support users to register and log in on the authorization trading platform and upload work information data;

[0072] A digital copyright confirmation unit, which is used to verify the originality and evaluate the quality of the work information data, generate a digital fingerprint of the work; automatically write the digital fingerprint of the work into the blockchain evidence storage network through a smart contract to generate an NFT digital certificate;

[0073] An intelligent authorization contract generation unit, which has a built-in smart contract template, matches and generates personalized terms according to the digital fingerprint of the work and the preset authorization requirements; converts the personalized terms into standardized smart contract code and generates the corresponding contract text at the same time;

[0074] An online authorization trading unit, which is used to obtain market data by integrating an oracle network, adopt an improved Dutch auction algorithm for dynamic pricing, and obtain the contract text after the user determines the price to complete the transaction;

[0075] A copyright protection and monitoring unit, which is used to monitor the entire network for the authorized works, provide an infringement warning mechanism, and if unauthorized use is detected, automatically package the infringement evidence data and generate an electronic evidence package to notify the user;

[0076] A decision support unit, which is used to obtain transaction data, construct a digital copyright value evaluation model, take the transaction data and market data as the input of the digital copyright value evaluation model, predict the digital copyright value index, evaluate the market value of the authorized works according to the predicted digital copyright value index, formulate marketing strategies for users, and provide decision-making basis for investors.

[0077] The work information data includes text work data, image work data, audio work data, video work data and work attribute data; the text work data includes text files (such as TXT, DOC, PDF, etc.) and text types (such as novels, papers, scripts, etc.); the image work data includes image files (such as JPG, PNG, GIF, etc.), image resolution, color mode and image size; the audio work data includes audio files (such as MP3, WAV, AAC, etc.), audio duration, sampling rate and bit rate; the video work data includes video files (such as MP4, AVI, MOV, etc.), video resolution, frame rate and video duration; the work attribute data includes work title, work tags, creation time and creation location.

[0078] The method for generating the work digital fingerprint by performing originality verification and quality assessment review on the work information data includes:

[0079] Group the work information data through the K-means clustering algorithm, select the number of clusters K according to the number of categories of the work information data, select the work feature vectors in the work information data corresponding to the number of clusters K as the initial cluster centers, calculate the distance of each work to each cluster center, and assign it to the nearest cluster group;

[0080] It should be noted that for various types of data in the work information data, the methods for obtaining work feature vectors are different. For text works, the bag-of-words model can be used to count the occurrence times of words to form vectors, or TF-IDF can be used to consider word frequency and inverse document frequency to generate vectors, and pre-trained word vector models such as Word2Vec and BERT can also be used to convert words into vectors and combine them; for image works, vectors can be constructed by calculating color histograms and extracting texture features using gray-level co-occurrence matrices; for audio works, features such as short-time energy and zero-crossing rate are extracted from the time domain, and features such as Mel frequency cepstral coefficients are calculated from the frequency domain to form vectors; for video works, each frame of image can be processed according to the image feature extraction method and then averaged or spliced frame feature vectors, or 3D-CNN can be used to process the video sequence to obtain feature vectors; for the title and tags in the work attribute data, text feature extraction methods can be used, the creation time can be quantified and encoded, and the creation location can be geocoded and converted into a numerical vector.

[0081] Update the center point of each cluster to the mean of all the work feature vectors in that cluster; calculate the distance from each work to the center of its belonging cluster, preset a distance threshold, if the distance from a certain work to the center of its belonging cluster is less than the preset distance threshold, then determine that work as an original work; if the distance from a certain work to the center of its belonging cluster is greater than or equal to the preset distance threshold, then determine that work as a non - original work;

[0082] Calculate the quality score of each work through the work quality scoring formula, and the work quality scoring formula is: where qs i is the quality score of work i; f(i) is the feature vector of work i; f K (i) is the feature vector of the i - th work within the cluster number K; σ K (i) is the standard deviation of the i - th work within the cluster number K; |f(i) - f K (i)| is the absolute difference between the feature vector of work i and the feature vector of the i - th work within the cluster number K; n is the number of works within the cluster number K; i is the index of the work;

[0083] The quality scoring formula is evaluated based on the deviation degree of the work from other works within its category. Its design purpose is to generate a comprehensive score reflecting the work quality by calculating the absolute difference (deviation degree) between the feature vector of the work and the features of other works within the category, as well as the relationship between the standard deviation of the features of works within the category. In any creative field, the quality of a work is relative and often needs to be compared with works of the same kind. To ensure the accuracy of the quality score, the quality scoring formula adopts the following two main factors:

[0084] Deviation degree of work features: By calculating the difference between the work feature vector and the feature vectors of other works within the category, it measures the uniqueness and difference of the work within the category. A higher deviation degree usually means stronger uniqueness of the work, but it may also mean poorer work quality (if the deviation from high - quality works is large);

[0085] Standard deviation within the category: The standard deviation of the features of other works within the category reflects the quality fluctuation of works within the category. The smaller the standard deviation, the more consistent the quality of the works; the larger the standard deviation, the greater the quality difference of the works. Therefore, the ratio of the work to the standard deviation can help identify whether it is a high - quality work.

[0086] Through the absolute difference |f(i) - f K(i) It can measure the difference between the feature vector of a work and the feature vectors of other works within the category. This method can quantify the "deviation degree" of a work, that is, the difference between its "position" in the category and other works. By normalizing the difference with the standard deviation, it ensures that the scoring result not only considers the degree of deviation of the work, but also takes into account the quality distribution of works within the category. For categories with a smaller standard deviation, the influence of the deviation degree on the score is greater, while for categories with a larger standard deviation, the influence of the deviation degree is relatively smaller, making the scoring more balanced. In the formula, the deviation degrees of all works are accumulated and averaged. This method takes into account the performance of all works within the category, avoids the extreme influence of a single work, and ensures the fairness and stability of the scoring system.

[0087] The advantages of the quality scoring formula compared with the prior art are as follows: Compared with existing quality assessment methods (such as simple similarity scoring or directly scoring works), this quality scoring formula can more accurately evaluate the relative quality of works. Existing methods often only focus on the similarity or score of a work with other works, while ignoring the "deviation degree" of the work and the influence of the quality fluctuation of works within the category. By introducing the standard deviation as a normalization factor, this quality scoring formula makes the scoring more refined and scientific, and can effectively handle the situation where the quality of works within the category fluctuates greatly. In addition, through quantification and standardization processing, this formula shows better adaptability and flexibility when dealing with the quality differences of works within the category. Its design principle is more in line with the requirements in practical applications, and can dynamically adjust the scoring weights according to the specific situation within the category, avoiding the limitations of traditional linear or fixed-standard scoring models. This not only improves the accuracy of scoring, but also enhances the adaptability of the model to works of different categories, making it more in line with the actual creation and evaluation needs.

[0088] The feature vector of each work is hashed through a hashing algorithm. The feature vector of each work is mapped to a unique fixed-length hash value through the SHA-256 hashing function. The digital fingerprint of the work is obtained by using the digital fingerprint generation formula through the SHA-256 hashing function. The digital fingerprint generation formula is: Fr(i) = Hash(f(i)); where Fr(i) is the digital fingerprint (hash value) of work i; Hash(·) is the SHA-256 hashing function.

[0089] The method for generating NFT digital vouchers includes:

[0090] Write the smart contract logic using the Solidity programming language, define the functions and rules of the smart contract, and deploy the written smart contract to the blockchain evidence storage network through the development tool Truffle; connect to the blockchain evidence storage network and interact with the smart contract through JavaScript libraries (such as Web3.js, Ethers.js) to execute the mint function in the smart contract; mint NFTs through the mint function, associate the digital fingerprint of the work with the NFT, and generate a unique NFT digital certificate; mint is a contract function used to create a new NFT on the blockchain. When this contract function is called, the smart contract will: generate a new NFT and give it a unique identifier (usually an ID); associate NFT data (usually including the digital fingerprint of the work), assign the ownership of the NFT to the user of the specified work information data, the NFT digital certificate contains the digital fingerprint information of the work, and assign the ownership of the NFT digital certificate to the user who uploads the work information data.

[0091] The method for matching and generating personalized terms based on the digital fingerprint of the work and the preset authorization requirements includes:

[0092] The preset authorization requirements include the scope of authorization, the set period of authorization, the limited geographical scope of authorization, the limited usage method of the work, the set royalty rate to be paid by the licensee, and the compensation terms in case of default.

[0093] Preset a smart contract template, use natural language processing technology to analyze the digital fingerprint of the work and the preset authorization requirements, extract relevant clause content from the preset smart contract template, cluster the digital fingerprint of the work through the density clustering algorithm, extract relevant clauses from the preset smart contract template, and generate personalized terms that match the digital fingerprint of the work and the preset authorization requirements through the cosine similarity algorithm according to the clustered digital fingerprint of the work.

[0094] The method for obtaining market data by integrating an oracle network includes:

[0095] Utilize Oracle oracle technology to connect the blockchain evidence storage network with external data sources and obtain market data from the financial market. The market data includes the search volume of different works, the view volume of different works, the download volume of different works, market prices, trading frequencies, authorization fees, market growth rates, pricing of similar works, and royalty rates.

[0096] The method for dynamic pricing using the improved Dutch auction algorithm includes:

[0097] An online authorization trading unit is used to obtain market data by integrating an oracle network, perform dynamic pricing using the improved Dutch auction algorithm, and after the user determines the price and completes the transaction, obtain the contract text.

[0098] The market price preset to be obtained at time point t is P ma (t), and the market price change amount between time point t and time point t + 1 is ΔP ma (t). The initial auction price of the original work on the authorized trading platform is preset as P0, and the price reduction step size for each time interval is ΔP ac , and it is adjusted in real time according to the market price P ma (t) through the dynamic price adjustment formula;

[0099] The dynamic price adjustment formula is: where P ac (t) is the auction price of the original work at time point t; α is the market price weight factor, which controls the influence degree of the market price on the initial auction price of the original work; Δt is the time interval; is the integer quotient of time point t and time interval Δt; is the floor function symbol;

[0100] The dynamic price adjustment formula incorporates the market price change amount, reflecting the changes in the external market environment. The market price fluctuations directly affect the value and supply - demand relationship of the works. Incorporating it can closely combine the auction price with the market dynamics. For example, when the market price of popular works rises, the auction price can be adjusted accordingly.

[0101] In order to break the limitations of traditional fixed - price or one - time bidding pricing, enable the auction price of the works to be adjusted in real time according to market changes, more accurately reflect the market value of the works, and give full play to the role of the market supply - demand relationship. Through the gradual reduction of price over time and the dynamic adjustment mechanism, the buyer can choose the best purchase timing according to their budget and needs, improving the flexibility and attractiveness of the auction; the seller can complete the transaction at the right time, maximizing the revenue and balancing the interests of both parties. Make the auction price more in line with the actual market conditions, avoid price deviation, improve the transparency and fairness of the auction, enhance the market competitiveness of the platform, and promote the conclusion of transactions.

[0102] The dynamic price adjustment formula comprehensively considers various key factors such as market price, time, and initial price, comprehensively reflecting the main aspects affecting the auction price of the works, making the pricing more in line with the actual market situation. When the market price rises, through the adjustment of the market price weight factor, the auction price can be increased accordingly; during the passage of time, the price will decline according to the set rhythm, comprehensively reflecting the dynamic changes in the market.

[0103] The dynamic adjustment of the market price weight factor enables it to flexibly adjust the price according to real-time market changes. The time decay term ensures that the price adjustment will not be overly dependent on the initial state, and the influence of the number of participants enables the price to adapt to changes in the degree of market competition, enhancing the rationality and adaptability of price adjustment. The strategy of gradually reducing the price over time conforms to the general rules of the auction market, can stimulate the buyer's desire to buy, promote active bidding activities, and increase the transaction rate, which is in line with the logic of market transactions.

[0104] Compared with the existing technology, the technical effect is:

[0105] The existing pricing methods are single and difficult to adapt to market changes. However, this formula can more accurately reflect the market value of the work by dynamically adjusting multiple factors, avoiding prices from deviating from the real market demand and improving the accuracy of pricing.

[0106] Traditional pricing lacks timely response to market changes. This formula can dynamically adjust prices according to market price fluctuations, time changes and the number of participants, so that auction prices can quickly adapt to market changes and better meet market demand.

[0107] It provides buyers with more options and reduces the unfairness of a single price to buyers; sellers can also maximize profits through reasonable pricing, which improves the satisfaction of both buyers and sellers and promotes active transactions.

[0108] Through a more reasonable pricing mechanism, the transparency and fairness of the auction are enhanced, the platform's market competitiveness is improved, more users are attracted to participate, and the sustainable development of the platform is promoted.

[0109] The market price weight factor is dynamically adjusted through the market price weight adjustment formula. The market price weight adjustment formula is: Among them, α(t) is the market price weight factor after dynamic adjustment; e -λ·t represents the time decay term of the market price weight factor, and its value will gradually decrease as time t increases; λ represents the time decay factor, which controls the degree to which the market price influence decreases over time; N(t) represents the number of buyers and sellers participating in the auction at time point t;

[0110] The market price weight adjustment formula is a weight factor formula that is dynamically adjusted based on multiple factors such as market price changes, time decay, and the number of participants. It is designed to ensure that the pricing mechanism in the auction system can flexibly respond to market changes in practical applications and adapt to the influence of time and participant behavior.

[0111] α, as the basic market price weight factor, reflects the initial market conditions or the weights set by the designer, ensuring that the system can maintain the basic pricing rules without external changes; as the initial reference value, it acts together with the market price and the number of participants in the dynamic pricing process. As market conditions change, the system dynamically adjusts the weight factor; the change amount of the market price directly affects the price adjustment in the Dutch auction. When the market price fluctuates violently, the auction price should respond. This part enables the auction price to better integrate with the external market dynamics by introducing the real-time price changes in the market; the market price changes reflect the changes in the external market environment and directly affect the flexibility of the auction price; over time, the influence of the market on the initial weight will gradually weaken. The time decay factor simulates the influence of time in the form of exponential decay, so that the system does not overly rely on the initial weight and can better adapt to the market changes in long-term auctions; by introducing time decay, the system can gradually abandon the initially set weight, increase the dependence on real-time market data, and avoid the pricing being stagnant at a fixed initial value for a long time; the change in the number of participants has an important impact on the competitiveness of the auction and the liquidity of the market. As the number of participants increases, the price fluctuations in the market may intensify, resulting in more frequent adjustments to the auction price. The introduction of the number of participants enables the system to dynamically adjust the influence of the market in a highly competitive situation; the weight is adjusted according to the number of participants to ensure that the system can handle market participation of different scales, increasing the influence of the market on the auction price when there are more participants and reducing the impact of market fluctuations on price adjustment when there are fewer participants;

[0112] The advantages of the market price weight adjustment formula compared with the prior art are as follows:

[0113] Dynamic adjustment ability: Different from the traditional fixed weight factor, this formula can react according to the actual market conditions and the passage of time by dynamically adjusting the weight factor. This enables the system to more flexibly respond to market changes and avoids the situation of pricing rigidity caused by the fixed weight factor.

[0114] Taking into account the comprehensive influence of multiple factors: Traditional auction systems usually adjust prices only based on simple market data or time, while this formula takes into account the influence of market price changes, time decay, and the number of participants, enabling the system to operate in a more complex market environment and more realistically simulate the dynamic process of actual transactions.

[0115] Time Decay and Participant Influence: Existing technologies often overlook the impact of time factors or simply use linear models to adjust prices. However, a design with an exponential decay factor can more naturally represent the gradual reduction of auction prices over time and their decreasing dependence on the initial weight. Additionally, the introduction of the number of participants increases the adaptability of the system, enabling auction prices to more closely match the actual market situation whether there are many or few participants.

[0116] For example, the market price weight factor α is 0.5; the time decay factor λ is 0.1, indicating that the number of buyers and sellers participating in the auction at time point 5, N(5), is 10; the market price change ΔP ma (t) is 10%, then the dynamically adjusted market price weight factor

[0117] Within each time period, the auction price of the original work gradually decreases over time. When the auction price of the original work P ac (t1) determined by the bidder at time point t1 is reached, the transaction is concluded, and the escrow process of the escrow smart contract is triggered; it is assumed that the transaction occurs at time point t1, and the seller and buyer are A and B respectively;

[0118] The escrow process is: F es (t1) = Es(P ac (t1), A, B); where Es(P ac (t1), A, B) represents the escrow smart contract; F es (t1) represents the value of the escrow smart contract triggered when the transaction is concluded at time point t1, indicating the escrow status between the seller and the buyer; P ac (t1) represents the auction price of the original work determined at time point t1;

[0119] Using cross-chain bridging technology for settlement in different currencies through a cross-chain settlement formula, the buyer can pay in different cryptocurrencies. It is set that the cryptocurrency paid by the buyer is C br , and the preset exchange rate at settlement is P se , the ratio of converting one cryptocurrency to another;

[0120] The cross-chain settlement formula is: P se (C br ) = P ac (t1) · P se (C br , ke); where P se (C br , ke) represents the cryptocurrency C paid by the buyer at settlement brThe exchange rate with the native currency of the authorized trading platform; ke represents the native currency of the authorized trading platform;

[0121] When the cryptocurrency paid by the buyer is converted into the native currency of the authorized trading platform, the smart contract will automatically trigger a fund transfer, sending the funds from the escrow account of the authorized trading platform to the seller.

[0122] The method for automatically encapsulating infringement evidence data in case of unauthorized use is as follows:

[0123] Use the Scrapy web scraping tool to monitor all uses related to the authorized works on the network in real time; the uses include whether the authorized works are cited, whether the authorized works are downloaded by unauthorized users, whether the authorized works are re-created by unauthorized users, and whether the authorized works are disseminated by unauthorized users; if unauthorized use is detected, compare the digital fingerprint of the unauthorized work with the digital fingerprint of the authorized work to confirm whether infringement has occurred; if infringement has occurred, automatically encapsulate the infringement evidence data; the infringement evidence data includes the storage location of the infringing work, the web page screenshot of the infringing work, the video of the infringing work, the time when the infringement event occurred, and the ID of the infringing user.

[0124] The method for constructing a digital copyright value evaluation model includes:

[0125] Divide the dataset into a training set and a test set to construct a digital copyright value evaluation model; the sample set is a subset of the dataset, and each sample set includes historical transaction data, market data, and the corresponding digital copyright value index; the input data of the model is historical transaction data and market data; the output label is the digital copyright value index; the digital copyright value evaluation model is a random forest regressor model;

[0126] Initialize the digital copyright value evaluation model and set the hyperparameters of the number and depth of the trees; use the training set to train the digital copyright value evaluation model, use the k-fold cross-validation method to adjust the hyperparameters of the model, and optimize the initially set parameters; use the test set data to evaluate the performance of the model, and use the coefficient of determination to evaluate and calculate the difference between the prediction result and the true label;

[0127] According to the model performance feedback, adjust the hyperparameters of the number and depth of the trees, optimize the model, retrain the model with the adjusted hyperparameters, and stop when the training reaches the preset model complexity to obtain the finally trained digital copyright value evaluation model; use the trained digital copyright value evaluation model to predict the current transaction data and market data, and predict the digital copyright value index.

[0128] The method for evaluating the market value of an authorized work based on the predicted digital copyright value index includes:

[0129] Preset a first threshold value of the digital copyright value index and a second threshold value of the digital copyright value index, compare the predicted digital copyright value index with the preset first threshold value and the preset second threshold value of the digital copyright value index respectively, and evaluate the market value of the authorized work;

[0130] If the predicted digital copyright value index is less than the preset first threshold value of the digital copyright value index, it is determined that the market value of the authorized work is at a low level;

[0131] If the predicted digital copyright value index is greater than or equal to the preset first threshold value of the digital copyright value index and less than or equal to the preset second threshold value of the digital copyright value index, it is determined that the market value of the authorized work is at a medium level;

[0132] If the predicted digital copyright value index is greater than the preset second threshold value of the digital copyright value index, it is determined that the market value of the authorized work is at a high level.

[0133] The preset first threshold value of the digital copyright value index is set by the staff. Different digital copyright value indexes are collected through the terminal of the authorization trading platform, and the average value of multiple digital copyright value indexes is taken as the preset first threshold value of the digital copyright value index; similarly, the preset second threshold value of the digital copyright value index and the preset distance threshold value are set.

[0134] In this embodiment, the use of the K-means clustering algorithm enables efficient originality verification. By clustering works based on feature vectors, works similar to or different from known works can be quickly identified, thereby automatically judging their originality. The clustering algorithm avoids the subjectivity of manual review and can process a large amount of data, making it suitable for large-scale work verification. By presetting a distance threshold to determine whether a work is original, the error of manual judgment is avoided, the judgment criteria are made more unified, and the accuracy of verification is improved; through automated clustering, updating of the center point, and distance calculation, the process of copyright review is greatly simplified and the efficiency is improved. Without manual intervention, the automatic processing of work features by the algorithm reduces the possibility of misjudgment or human intervention. The introduction of the work quality scoring formula provides a more objective and scientific work evaluation standard by calculating the quality score of each work, avoiding the limitations of relying solely on manual subjective judgment; the design of the work quality score is based on the calculation and comparison of work feature vectors, combined with the standard deviation and feature difference, which can objectively evaluate the quality of works, especially when dealing with a large amount of work information data. The use of the scoring formula can automatically evaluate the similarity between works and identify works with poor quality or those that do not meet the requirements; the digital fingerprint, as the unique identifier of a work, can be used not only for work originality verification but also for various purposes such as copyright protection and work traceability, enhancing the security and effectiveness of copyright protection; through the K-means clustering algorithm and the quality scoring formula, more refined review of works can be achieved. Especially in the case of a large variety of work categories, by calculating the similarity of works within a cluster, it is possible to more accurately determine whether a certain work meets the requirements of the platform;

[0135] The modified Dutch auction algorithm allows the auction price of original works to gradually decrease over time. This dynamic pricing mechanism can effectively attract buyers with different needs to participate in the bidding. As time goes by, the price of the work gradually decreases, and buyers can choose the best time to buy according to their budget and needs, which improves the flexibility and attractiveness of the auction. This method adopts a market price dynamic adjustment formula to ensure that the auction price is adjusted according to real-time market fluctuations, making the pricing more in line with market demand and actual market conditions, and avoiding price deviations that may be caused by fixed pricing; the dynamic adjustment of the market price weight factor makes the impact of market supply and demand on the auction price more direct and timely, and improves the transparency and fairness of the price mechanism; the auction mechanism adjusts the dynamic impact of prices and participants through time decay factors and market price weight factors, and can adjust the auction strategy according to the popularity of bidding, the number of participants and market changes. This mechanism can not only balance the needs of buyers and sellers, but also complete transactions at the right time, thereby improving the transaction rate and user satisfaction. The strategy of gradually reducing prices within time intervals allows buyers to bid at different prices at different time points, reducing the unfairness of a single price to buyers and promoting more active bidding activities; the use of cross-chain bridging technology enables multi-currency settlement, allowing buyers to use different cryptocurrencies for payment, greatly improving the flexibility and acceptability of the platform. This can not only meet the payment needs of different users, but also expand the platform's international market and attract global users. The cross-chain settlement formula ensures accurate settlement of transactions through exchange rate conversion, avoids exchange problems between different cryptocurrencies, and improves transaction liquidity and the convenience of cross-border transactions.

[0136] Embodiment 2

[0137] See also Figure 2 As shown, the part not described in detail in this embodiment is described in Example 1, which provides an intelligent copyright automated authorization transaction platform method, including:

[0138] S1. Support users to register and log in to the authorized trading platform and upload work information data;

[0139] S2. Verify the originality and quality of the work information data to generate a digital fingerprint of the work; automatically write the digital fingerprint of the work into the blockchain evidence storage network through smart contracts to generate NFT digital certificates;

[0140] S3, built-in smart contract template, matching and generating personalized terms according to the digital fingerprint of the work and the preset authorization requirements; converting personalized terms into standardized smart contract code and generating the corresponding contract text at the same time;

[0141] S4. Obtain market data through an integrated oracle network, adopt an improved Dutch auction algorithm for dynamic pricing. After the user determines the price to complete the transaction, obtain the contract text;

[0142] S5. Conduct a full-network monitoring of the authorized works, provide an infringement warning mechanism. If unauthorized use is detected, automatically encapsulate the infringement evidence data and generate an electronic evidence package to notify the user;

[0143] S6. Obtain transaction data, construct a digital copyright value evaluation model. Use the transaction data and market data as the input of the digital copyright value evaluation model to predict the digital copyright value index. According to the predicted digital copyright value index, evaluate the market value of the authorized works, formulate a marketing strategy for the user, and provide a decision-making basis for investors.

[0144] Since the electronic device introduced in this embodiment is the electronic device adopted in the intelligent copyright automated authorization trading platform system according to the embodiments of the present application, based on the intelligent copyright automated authorization trading platform system introduced in the embodiments of the present application, those skilled in the art can understand the specific implementation manners and various variations of the electronic device in this embodiment. Therefore, the specific implementation of how this electronic device realizes the method in the embodiments of the present application will not be described in detail here. As long as those skilled in the art implement the electronic device adopted in the intelligent copyright automated authorization trading platform system according to the embodiments of the present application, it falls within the protection scope of the present application.

[0145] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula that is closest to the real situation. The preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.

[0146] The above is only the preferred implementation manner of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the idea of the present invention belong to the protection scope of the present invention. It should be noted that for ordinary technical users in the technical field, several improvements and refinements made without departing from the principle of the present invention should also be regarded as within the protection scope of the present invention.

Claims

1. An intelligent copyright automated authorization trading platform system, characterized in that, include: User management unit, used to support users to register and log in to the authorization trading platform and upload work information data; Digital copyright confirmation unit, used to verify the originality and quality of work information data and generate digital fingerprints of works; The digital fingerprint of the work is automatically written into the blockchain evidence storage network through smart contracts to generate NFT digital certificates; Smart authorization contract generation unit, with built-in smart contract templates, generates personalized terms based on the digital fingerprint of the work and the preset authorization requirements; converts personalized terms into standardized smart contract codes and generates corresponding contract texts at the same time; The online authorized transaction unit is used to obtain market data through the integrated oracle network and adopt the modified Dutch auction algorithm for dynamic pricing. After the user determines the price and completes the transaction, he will obtain the contract text; The copyright protection and monitoring unit is used to monitor the entire network for authorized works and provide an infringement warning mechanism. If unauthorized use is detected, the infringement evidence data will be automatically packaged and an electronic evidence package will be generated to notify the user; The decision support unit is used to obtain transaction data, build a digital copyright value assessment model, use transaction data and market data as input to the digital copyright value assessment model, predict the digital copyright value index, and evaluate the market value of the authorized work based on the predicted digital copyright value index, formulate marketing strategies for users, and provide decision-making basis for investors.

2. The intelligent copyright automated authorization trading platform system according to claim 1, wherein The work information data includes text work data, image work data, audio work data, video work data and work attribute data.

3. The intelligent copyright automated authorization trading platform system according to claim 2, characterized in that, The method of verifying the originality and evaluating the quality of the work information data and generating the digital fingerprint of the work includes: The work information data is grouped by the K-means clustering algorithm. The number of clusters K is selected according to the number of categories of the work information data. The work feature vectors in the work information data corresponding to the number of clusters K are selected as the initial cluster centers. For each work, the distance from the work to each cluster center is calculated and the work is assigned to the nearest cluster group. Update the center point of each cluster to the mean of the feature vectors of all works in the cluster; calculate the distance from each work to the center of its cluster, preset a distance threshold, and if the distance from a certain work to the center of its cluster is less than the preset distance threshold, then the work is determined to be an original work; if the distance from a certain work to the center of its cluster is greater than or equal to the preset distance threshold, then the work is determined to be a non-original work; Calculate the quality score of each work through the work quality scoring formula, and the work quality scoring formula is: where qs i is the quality score of work i; f(i) is the feature vector of work i; f K (i) is the feature vector of the i-th work within the cluster number K; σ K (i) is the standard deviation of the i-th work within the cluster number K; |f(i) - f K (i)| is the absolute difference between the feature vector of work i and the feature vector of the i-th work within the cluster number K; n is the number of works within the cluster number K; i is the index of the work; The feature vector of each work is hashed by a hash algorithm, and the feature vector of each work is mapped to a unique hash value of fixed length by a SHA-256 hash function. The digital fingerprint of the work is obtained by using a digital fingerprint generation formula using the SHA-256 hash function; the digital fingerprint generation formula is: Fr(i)=Hash(f(i)); wherein Fr(i) is the digital fingerprint of work i; and Hash(·) is the SHA-256 hash function.

4. The intelligent copyright automated authorization trading platform system according to claim 3, characterized in that, The method for generating an NFT digital certificate includes: Write the smart contract logic using the Solidity programming language, define the functions and rules of the smart contract, and deploy the written smart contract to the blockchain evidence storage network through the development tool Truffle; Connect to the blockchain evidence storage network through a JavaScript library and interact with the smart contract to execute the mint function in the smart contract; Mint an NFT through the mint function, associate the digital fingerprint of the work with the NFT, and generate a unique NFT digital certificate.

5. The intelligent copyright automated authorization trading platform system according to claim 4, wherein The method for matching and generating personalized terms according to the digital fingerprint of the work and the preset authorization requirements includes: The preset authorization requirements include the scope of authorization, the set term of authorization, the limited geographical scope of authorization, the limited usage method of the work, the royalty rate that the licensee needs to pay, and the compensation clause after default; Preset a smart contract template, use natural language processing technology to analyze the digital fingerprint of the work and the preset authorization requirements, extract relevant clause content from the preset smart contract template, cluster the digital fingerprint of the work through the density clustering algorithm, extract relevant clauses from the preset smart contract template, and generate personalized terms that match the digital fingerprint of the work and the preset authorization requirements through the cosine similarity algorithm.

6. The intelligent copyright automated authorization trading platform system according to claim 5, wherein The method for obtaining market data by integrating the oracle network includes: Utilize Oracle oracle technology to connect the blockchain evidence storage network with external data sources, and obtain market data from the financial market. The market data includes the search volume of different works, the view volume of different works, the download volume of different works, the market price, the trading frequency, the authorization fee, the market growth rate, the pricing of similar works, and the royalty rate.

7. The intelligent copyright automated authorization trading platform system according to claim 6, wherein The method for dynamic pricing using the improved Dutch auction algorithm includes: An online authorization trading unit is used to obtain market data by integrating the oracle network, dynamically price using the improved Dutch auction algorithm, and obtain a contract text after the user determines the price and completes the transaction; The market price preset to be obtained at time point t is P ma (t), and the market price change amount between time point t and time point t + 1 is ΔP ma The initial auction price of the original work on the authorized trading platform is P0, and the price reduction step size for each time interval is ΔP ac , and perform real-time adjustment according to the market price P ma (t) through the dynamic price adjustment formula; The dynamic price adjustment formula is as follows: where P ac (t is the auction price of the original work at time point t; α is the market price weight factor; Δt is the time interval; is the integer quotient of time point t and time interval Δt; is the floor function symbol; Dynamically adjust the market price weight factor through the market price weight adjustment formula, and the market price weight adjustment formula is: Among them, α(t) is the market price weight factor after dynamic adjustment; e -λ·t represents the time decay term of the market price weight factor; λ represents the time decay factor; N(t) represents the number of buyers and sellers participating in the auction at time point t; During each time period, the auction price of the original work gradually decreases over time. At the auction price P of the original work determined by the bidder at time point t1 ac (t1), the transaction is concluded, and the fund escrow process of the fund escrow smart contract is triggered; it is preset that the transaction occurs at time point t1, and the seller and the buyer are A and B respectively; The fund custody process is as follows: F es Es(P ac (t1), A, B); where Es(P ac (t1), A, B) represents the fund custody smart contract; F es (t1) represents the value of the fund custody smart contract triggered when the transaction is completed at time point t1; P ac (t1) represents the auction price of the original work determined at time point t1; Use cross-chain bridging technology to conduct settlement in different currencies through a cross-chain settlement formula. Set the cryptocurrency paid by the buyer as C br , and the preset exchange rate at the time of settlement is P se ; The cross-chain settlement formula is: P se (C br ) = P ac (t1) · P se (C br , ke); where P se (C br , ke) represents the exchange rate between the cryptocurrency C br paid by the buyer at the time of settlement and the native currency of the authorized trading platform; ke represents the native currency of the authorized trading platform; When the cryptocurrency paid by the buyer is converted into the native currency of the authorization trading platform, the smart contract will automatically trigger a fund transfer and send the funds from the escrow account of the authorization trading platform to the seller.

8. The intelligent copyright automated authorization trading platform system according to claim 7, characterized in that The method for automatically encapsulating infringement evidence data if unauthorized use is detected includes: Use the Scrapy web scraping tool to monitor all usage related to the authorized work on the network in real time; The usage includes whether the authorized work is cited, whether the authorized work is downloaded by unauthorized users, whether the authorized work is recreated by unauthorized users, and whether the authorized work is spread by unauthorized users; If unauthorized use is detected, compare the digital fingerprint of the unauthorized work with the digital fingerprint of the authorized work to confirm whether infringement is constituted; If infringement is constituted, automatically encapsulate the infringement evidence data; The infringement evidence data includes the storage location of the infringing work, the web page screenshot of the infringing work, the video of the infringing work, the time when the infringement event occurred, and the ID of the infringing user.

9. The intelligent copyright automated authorization trading platform system according to claim 8, characterized in that, The method for constructing the digital copyright value evaluation model includes: Divide the dataset into a training set and a test set, and construct a digital copyright value evaluation model; the sample set is a subset of the dataset, and each sample set includes historical transaction data, market data, and the corresponding digital copyright value index; the input data of the model is historical transaction data and market data; the output label is the digital copyright value index; the digital copyright value evaluation model is a random forest regressor model; Initialize the digital copyright value evaluation model and set the hyperparameters of the number and depth of the trees; use the training set to train the digital copyright value evaluation model, use the k-fold cross-validation method to adjust the hyperparameters of the model, and optimize the initially set parameters; use the test set data to evaluate the performance of the model, and use the coefficient of determination to evaluate and calculate the difference between the prediction result and the true label; According to the model performance feedback, adjust the hyperparameters of the number and depth of the trees, optimize the model, retrain the model with the adjusted hyperparameters, and stop when the training reaches the preset model complexity to obtain the finally trained digital copyright value evaluation model; use the trained digital copyright value evaluation model to predict the current transaction data and market data, and predict the digital copyright value index.

10. The intelligent copyright automated authorization trading platform system according to claim 9, characterized in that The method for evaluating the market value of the licensed work according to the predicted digital copyright value index includes: Preset the first threshold of the digital copyright value index and the second threshold of the digital copyright value index, compare the predicted digital copyright value index with the preset first threshold of the digital copyright value index and the preset second threshold of the digital copyright value index respectively, and evaluate the market value of the licensed work; If the predicted digital copyright value index is less than the preset first threshold of the digital copyright value index, it is determined that the market value of the licensed work is at a low level; If the predicted digital copyright value index is greater than or equal to the preset first threshold of the digital copyright value index and less than or equal to the preset second threshold of the digital copyright value index, it is determined that the market value of the licensed work is at a medium level; If the predicted digital copyright value index is greater than the preset second threshold of the digital copyright value index, it is determined that the market value of the licensed work is at a high level.

Images