Block chain digital asset algorithm and opportunity large model digital human interaction incentive algorithm

Abstract

The invention relates to a block chain digital asset algorithm and opportunity large model digital human interaction incentive algorithm, and the digital human interaction incentive algorithm comprises the following steps: a digital human interaction mechanism: carrying out the natural language interaction between a digital human and a user through a large model algorithm, answering the question of the user, and providing related services; evaluation of interaction frequency and quality: the system automatically records the interaction frequency and quality of the user and the digital person, and performs evaluation according to the interaction frequency, the interaction length and the interaction depth; according to the block chain digital asset algorithm and the opportunity large model digital human interaction excitation algorithm, the algorithm can be widely applied to various digital human systems needing user interaction, such as virtual customer service, intelligent assistant, online education and the like. By introducing a digital asset reward mechanism, the user is stimulated to perform more high-quality interaction with the digital human, so that the user experience and the overall efficiency of the system are improved.

Description

Technical Field

[0001] The present invention relates to the field of composite waterproof systems, and specifically to a blockchain digital asset algorithm and a digital human interaction incentive algorithm for an opportunity large model. Background Art

[0002] The digital human interactive incentive algorithm is a digital platform ecosystem built on digital technology in the context of the digital era. It fully releases the incentive contained in the data by collecting the behavior and characteristic data of the motivated object in real time and automatically performs subsequent data processing and analysis, so as to accurately strengthen or weaken the specific behavior of the motivated object. This concept has had a huge impact on traditional incentive mechanisms and methods, and provides an opportunity to solve the dilemma of traditional incentives; the realization of digital incentives depends on several core dimensions, including the background of the times, technical foundation, incentive mechanism, key activities and application scenarios. By understanding the relationship between these core dimensions, we can sort out the concept of digital incentives and analyze their application in different application scenarios. The technical implementation path and organizational implementation path of digital incentives, as well as the human-oriented, benefit-based and system-based laws followed by organizational human resource management, question and reflect on the practice of digital incentives, point out the breakthrough direction of organizational human resource management theory, in order to correctly guide and effectively promote the in-depth development of digital incentives; in addition, the presentation methods of digital humans can now basically be divided into three types: 2D, 2.5D and 3D. The choice of these presentation methods directly affects the interactive experience and visual effects of digital humans. For example, 2D digital humans usually use a voice to drive a photo, while 3D digital humans are realized through UE modeling technology, providing a more realistic and three-dimensional interactive experience; in terms of technical implementation, the dialogue process of digital humans involves ASR speech recognition, NLU natural language understanding, DM dialogue management, NL natural language generation, TTS text-to-speech, STV voice-to-video and other technologies. The integration of these technologies enables digital humans to interact in real time and provide a high-quality dialogue experience; This patent proposes a digital human interaction incentive algorithm that combines the blockchain digital asset algorithm with the opportunity model, aiming to improve the frequency and quality of interaction between users and digital humans through an interactive incentive mechanism. By introducing a digital asset reward mechanism, users can receive digital asset rewards during the interaction with digital humans, thereby encouraging more interactions. The system also uses artificial intelligence algorithms to transform an ordinary image into a three-dimensional 3D digital asset in a few seconds, increasing the user's purchase rate. Summary of the invention

[0003] To achieve the above purpose, the present invention provides the following technical solution: a blockchain digital asset algorithm and a digital human interaction incentive algorithm of an opportunity large model, the digital human interaction incentive algorithm comprising the following steps: Step 1: Digital human interaction mechanism: Digital humans interact with users in natural language through large model algorithms, answering users’ questions and providing related services; Step 2: Interaction frequency and quality assessment: the system automatically records the frequency and quality of interaction between the user and the digital human, and evaluates it based on the number of interactions, interaction length, and interaction depth; Step 3: Digital asset reward mechanism. Combined with the blockchain algorithm, users can obtain digital asset rewards through interaction points, digital asset difficulty adjustment and digital asset reward mechanism during the interaction with digital humans. Step 4: Automated mechanism to ensure efficient operation of the system and user experience. Propose an automated mechanism to automatically evaluate and record. The system automatically evaluates and records the frequency, length, and depth of each interaction, and updates the user's interaction score in real time. Step 5: Dynamic adjustment and notification: The system dynamically adjusts the digital asset algorithm based on the user’s interaction and overall digital asset difficulty, and notifies the user of the latest digital asset progress and rewards in real time through Digital Humans; Step 6: Security and privacy protection. The system uses blockchain technology to ensure the security and privacy of user interaction data to avoid data leakage and abuse.

[0004] Preferably, the interaction mechanism is based on the previously proposed two-stage answer model, including weight vector concatenation and a text-based matching model.

[0005] Preferably, the evaluation in step 2 is performed according to the following criteria: number of interactions, recording the number of conversations between the user and the digital human; length of interaction, counting the number of words and duration of each interaction; depth of interaction, analyzing the complexity and diversity of the conversation content.

[0006] Preferably, the steps of obtaining digital asset rewards through the mechanism include: S1. Interaction points: After each interaction with a digital human, the system calculates interaction points for the user based on the interaction evaluation criteria; S2. Digital asset difficulty adjustment: The system dynamically adjusts the difficulty of digital assets based on the overall user interaction to ensure the fairness and incentive of rewards; S3. Digital asset rewards: The system converts the user's interaction points into corresponding digital asset rewards and records them on the blockchain to ensure transparency and non-tamperability.

[0007] Preferably, the main principle of the digital human interaction mechanism is to first generate a digital human through photos or videos, and use ASR to identify the questions input by the user during the conversation. The questions are then sent to the digital human's brain to obtain answers, and then the answers are converted into audio through TTS. The digital human's lips and face are driven by the audio to form a digital human speaking video, thereby realizing a conversation between a real person and a digital human.

[0008] Preferably, the driving capabilities of the digital human include lips, facial expressions and body movements, and the lip and facial changes of the digital human are driven by audio.

[0009] Preferably, weight vector concatenation may involve the following steps: Step 1: Data preprocessing: First, the raw data needs to be preprocessed, including data normalization and standardization, to ensure that different features or data are compared and calculated on the same scale. Step 2: Feature extraction: Extract relevant features from the preprocessed data. These features are one-dimensional values ​​and multi-dimensional feature vectors. Step 3. Generate weight vectors: Generate corresponding weight vectors according to specific algorithm or model requirements. These weight vectors are usually obtained through training and are used to measure the importance or relevance of different features. Step 4. Concatenation operation: Concatenate the extracted feature vector with the weight vector to form a new vector. This process is a simple vector addition, multiplication or other more complex mathematical operations, depending on how the model is built and what is required.

[0010] Step 5. Model training and optimization: Use the concatenated vectors to train and optimize the model, and adjust the weights through methods such as back propagation so that the model can better fit the data and improve prediction accuracy.

[0011] Preferably, in image processing, the concatenation of weight vectors may involve using a sliding window to split the image into multiple small patches, then performing feature extraction and weight calculation on these patches, and finally concatenating the extracted features with the corresponding weights to form a richer feature representation.

[0012] Preferably, the blockchain system consists of a data layer, a network layer, a consensus layer, an incentive layer, a contract layer, and an application layer.

[0013] Compared with the prior art, the present invention has the following beneficial effects: The blockchain digital asset algorithm of the present invention and the digital human interaction incentive algorithm of the opportunity large model can be widely used in various digital human systems that require user interaction, such as virtual customer service, intelligent assistants, online education, etc. By introducing a digital asset reward mechanism, users are encouraged to have more high-quality interactions with digital humans, thereby improving user experience and the overall efficiency of the system. DETAILED DESCRIPTION

[0014] The following will be combined with 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.

[0015] The present invention provides a technical solution: a blockchain digital asset algorithm and a digital human interaction incentive algorithm of an opportunity large model, the digital human interaction incentive algorithm comprising the following steps: Step 1: Digital human interaction mechanism: Digital humans interact with users in natural language through large model algorithms, answering users’ questions and providing related services; Step 2: Interaction frequency and quality assessment: the system automatically records the frequency and quality of interaction between the user and the digital human, and evaluates it based on the number of interactions, interaction length, and interaction depth; Step 3: Digital asset reward mechanism. Combined with the blockchain algorithm, users can obtain digital asset rewards through interaction points, digital asset difficulty adjustment and digital asset reward mechanism during the interaction with digital humans. Step 4: Automated mechanism to ensure efficient operation of the system and user experience. Propose an automated mechanism to automatically evaluate and record. The system automatically evaluates and records the frequency, length, and depth of each interaction, and updates the user's interaction score in real time. Step 5: Dynamic adjustment and notification: The system dynamically adjusts the digital asset algorithm based on the user’s interaction and overall digital asset difficulty, and notifies the user of the latest digital asset progress and rewards in real time through Digital Humans; Step 6: Security and privacy protection. The system uses blockchain technology to ensure the security and privacy of user interaction data to avoid data leakage and abuse.

[0016] Among them, the interaction mechanism is based on the previously proposed two-stage answer model, including weight vector concatenation and text-based matching model.

[0017] Among them, step 2 is evaluated according to the following criteria: number of interactions, recording the number of conversations between the user and the digital human; interaction length, counting the number of words and duration of each interaction; interaction depth, analyzing the complexity and diversity of the conversation content.

[0018] Among them, the steps of obtaining digital asset rewards through the mechanism include: S1. Interaction points: After each interaction with a digital human, the system calculates interaction points for the user based on the interaction evaluation criteria; S2. Digital asset difficulty adjustment: The system dynamically adjusts the difficulty of digital assets based on the overall user interaction to ensure the fairness and incentive of rewards; S3. Digital asset rewards: The system converts the user's interaction points into corresponding digital asset rewards and records them on the blockchain to ensure transparency and non-tamperability.

[0019] Among them, the main principle of the digital human interaction mechanism is to first generate a digital human through photos or videos, and use ASR to identify the questions input by the user during the conversation. Then the questions are sent to the digital human brain to obtain answers, and then the answers are converted into audio through TTS. The digital human's lips and face are driven by audio to form a digital human speaking video, thereby realizing the conversation between real people and digital humans.

[0020] Among them, the driving capabilities of digital humans include lips, facial expressions and body movements, and the changes in the digital human's lips and face are driven by audio.

[0021] Among them, weight vector concatenation may involve the following steps: Step 1: Data preprocessing: First, the raw data needs to be preprocessed, including data normalization and standardization, to ensure that different features or data are compared and calculated on the same scale. Step 2: Feature extraction: Extract relevant features from the preprocessed data. These features are one-dimensional values ​​and multi-dimensional feature vectors. Step 3. Generate weight vectors: Generate corresponding weight vectors according to specific algorithm or model requirements. These weight vectors are usually obtained through training and are used to measure the importance or relevance of different features. Step 4. Concatenation operation: Concatenate the extracted feature vector with the weight vector to form a new vector. This process is a simple vector addition, multiplication or other more complex mathematical operations, depending on how the model is built and what is required.

[0022] Step 5. Model training and optimization: Use the concatenated vectors to train and optimize the model, and adjust the weights through methods such as back propagation so that the model can better fit the data and improve prediction accuracy.

[0023] In image processing, the concatenation of weight vectors may involve using a sliding window to split the image into multiple small patches, then extracting features and calculating weights on these patches, and finally concatenating the extracted features with the corresponding weights to form a richer feature representation.

[0024] Among them, the blockchain system consists of data layer, network layer, consensus layer, incentive layer, contract layer and application layer.

[0025] This algorithm can be widely used in various digital human systems that require user interaction, such as virtual customer service, intelligent assistants, online education, etc. By introducing a digital asset reward mechanism, users are encouraged to have more high-quality interactions with digital humans, thereby improving user experience and the overall effectiveness of the system.

[0026] Algorithm Advantages Improve user interaction: Through the digital asset reward mechanism, the frequency and quality of interaction between users and digital people can be significantly improved.

[0027] Automated and efficient operation: The system automatically evaluates and records user interactions, dynamically adjusts digital asset difficulty and rewards, and ensures efficient operation of the system.

[0028] Data security and transparency: Blockchain technology is used to ensure the security and transparency of user data and enhance user trust.

[0029] The digital human interactive incentive algorithm proposed in this patent organically combines big models and blockchain technology, providing an innovative user incentive mechanism with significant practical value and commercial potential.

[0030] Embodiment 1 The digital human interaction incentive algorithm of the blockchain digital asset algorithm and the opportunity big model includes the following steps: digital human interaction mechanism, the digital human interacts with the user in natural language through the big model algorithm, answers the user's questions and provides related services; interaction frequency and quality evaluation, the system automatically records the interaction frequency and quality between the user and the digital human, and evaluates according to the number of interactions, interaction length and interaction depth; digital asset reward mechanism, combined with the blockchain algorithm, the user obtains digital asset rewards through interaction points, digital asset difficulty adjustment and digital asset reward mechanism during the interaction with the digital human; automation mechanism, to ensure the efficient operation of the system and user experience, an automation mechanism is proposed, automatic evaluation and recording, the system automatically evaluates and records the frequency, length and depth of each interaction, and updates the user's interaction points in real time; dynamic adjustment and notification, the system dynamically adjusts the digital asset algorithm according to the user's interaction situation and the overall digital asset difficulty, and notifies the user of the latest digital asset progress and reward situation in real time through the digital human; security and privacy protection, the system ensures the security and privacy protection of user interaction data through blockchain technology to avoid data leakage and abuse.

[0031] Embodiment 2 The mechanism for obtaining digital asset rewards includes the following steps: Interaction points: After each interaction with a digital human, the system calculates interaction points for the user based on the interaction evaluation criteria; Digital asset difficulty adjustment: The system dynamically adjusts the digital asset difficulty based on the overall user interaction situation to ensure the fairness and incentive of the reward; Digital asset rewards: The system converts the user's interaction points into corresponding digital asset rewards and records them on the blockchain to ensure transparency and non-tamperability.

[0032] Embodiment 3 Weight vector concatenation may involve the following steps: Data preprocessing: First, the original data needs to be preprocessed, including data normalization and standardization to ensure that different features or data are compared and calculated on the same scale. Feature extraction: Extract relevant features from the preprocessed data. These features are one-dimensional numerical values ​​and multi-dimensional feature vectors. Weight vector generation: Generate corresponding weight vectors according to specific algorithm or model requirements. These weight vectors are usually obtained through training and are used to measure the importance or relevance of different features. Concatenation operation: Concatenate the extracted feature vector with the weight vector to form a new vector. This process is a simple vector addition, multiplication or other more complex mathematical operations, depending on how the model is built and the requirements. Model training and optimization: Use the concatenated vectors to train and optimize the model, and adjust the weights through methods such as back propagation so that the model can better fit the data and improve the prediction accuracy.

[0033] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the attached claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any mark in the claims should not be regarded as limiting the claims involved.

[0034] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. The blockchain digital asset algorithm and the digital human interaction incentive algorithm of the opportunity model are characterized by: The digital human interaction excitation algorithm includes the following steps: Step 1: Digital human interaction mechanism: Digital humans interact with users in natural language through large model algorithms, answering users’ questions and providing related services; Step 2: Interaction frequency and quality assessment: the system automatically records the frequency and quality of interaction between the user and the digital human, and evaluates it based on the number of interactions, interaction length, and interaction depth; Step 3: Digital asset reward mechanism. Combined with the blockchain algorithm, users can obtain digital asset rewards through interaction points, digital asset difficulty adjustment and digital asset reward mechanism during the interaction with digital humans. Step 4: Automated mechanism to ensure efficient operation of the system and user experience. Propose an automated mechanism to automatically evaluate and record. The system automatically evaluates and records the frequency, length, and depth of each interaction, and updates the user's interaction score in real time. Step 5: Dynamic adjustment and notification: The system dynamically adjusts the digital asset algorithm based on the user’s interaction and overall digital asset difficulty, and notifies the user of the latest digital asset progress and rewards in real time through Digital Humans; Step 6: Security and privacy protection. The system uses blockchain technology to ensure the security and privacy of user interaction data to avoid data leakage and abuse.

2. The blockchain digital asset algorithm and the digital human interaction incentive algorithm of the opportunity large model according to claim 1 are characterized by: The interaction mechanism is based on the previously proposed two-stage answer model, which includes weight vector concatenation and text-based matching model.

3. The blockchain digital asset algorithm and the digital human interaction incentive algorithm of the opportunity large model according to claim 1 are characterized by: In step 2, the evaluation was conducted based on the following criteria: number of interactions, which records the number of conversations between the user and the digital human; Interaction length, counts the number of words and duration of each interaction; interaction depth, analyzes the complexity and diversity of the conversation content.

4. The blockchain digital asset algorithm and the digital human interaction incentive algorithm of the opportunity large model according to claim 1 are characterized by: The steps to obtain digital asset rewards through the mechanism include: S1. Interaction points: After each interaction with a digital human, the system calculates interaction points for the user based on the interaction evaluation criteria; S2. Digital asset difficulty adjustment: The system dynamically adjusts the difficulty of digital assets based on the overall user interaction to ensure the fairness and incentive of rewards; S3. Digital asset rewards: The system converts the user's interaction points into corresponding digital asset rewards and records them on the blockchain to ensure transparency and non-tamperability.

5. The blockchain digital asset algorithm and the digital human interaction incentive algorithm of the opportunity large model according to claim 1 are characterized by: The main principle of the digital human interaction mechanism is to first generate a digital human through photos or videos, and use ASR to identify the questions input by the user during the conversation. The questions are then sent to the digital human's brain to obtain answers, and then converted into audio through TTS. The audio drives the digital human's lips and face to form a digital human speaking video, thereby realizing a conversation between a real person and a digital human.

6. The blockchain digital asset algorithm and the digital human interaction incentive algorithm of the opportunity large model according to claim 1 are characterized by: The driving capabilities of digital humans include lips, facial expressions and body movements, and the lip and facial changes of digital humans are driven by audio.

7. The blockchain digital asset algorithm and the digital human interaction incentive algorithm of the opportunity large model according to claim 1 are characterized by: Weight vector concatenation may involve the following steps: Step 1: Data preprocessing: First, the raw data needs to be preprocessed, including data normalization and standardization, to ensure that different features or data are compared and calculated on the same scale. Step 2: Feature extraction: extract relevant features from the preprocessed data. These features are one-dimensional values ​​and multi-dimensional feature vectors. Step 3: Generate weight vectors: Generate corresponding weight vectors according to specific algorithm or model requirements. These weight vectors are usually obtained through training and are used to measure the importance or relevance of different features. Step 4: Concatenation operation: Concatenate the extracted feature vector with the weight vector to form a new vector. This process is a simple vector addition, multiplication or other more complex mathematical operations, depending on how the model is built and what is required. Step 5: Model training and optimization: Use the concatenated vectors to train and optimize the model, and adjust the weights through methods such as back propagation so that the model can better fit the data and improve prediction accuracy.

8. The blockchain digital asset algorithm and the digital human interaction incentive algorithm of the opportunity large model according to claim 1 are characterized by: In image processing, the splicing of weight vectors may involve using a sliding window to split the image into multiple small patches, then performing feature extraction and weight calculation on these patches, and finally splicing the extracted features with the corresponding weights to form a richer feature representation.

9. The blockchain digital asset algorithm and the digital human interaction incentive algorithm of the opportunity large model according to claim 1 are characterized by: The blockchain system consists of data layer, network layer, consensus layer, incentive layer, contract layer and application layer.