Platform content identification method and system based on AIGC

Abstract

The invention discloses a platform content identification method and system based on AIGC, and the method comprises the steps: receiving first prompt information through an AIGC management platform, distributing the first prompt information to two AIGC engines for processing, enabling the two engines to generate different feedback information based on the same prompt information, and then enabling the feedback information to be sent to the AIGC management platform. First application scene information corresponding to the first feedback information and second application scene information corresponding to the second feedback information are recognized and distinguished, and when the AIGC management platform determines that the similarity between the two pieces of application scene information exceeds a preset similarity threshold value, the first feedback information and the second feedback information are automatically synthesized; and the first synthetic information with higher comprehensiveness and accuracy is generated, so that the accuracy and reliability of the information are improved, and richer and more diversified information choices are provided for the user.

Description

Technical Field

[0001] The present invention belongs to the field of artificial intelligence technology, and in particular relates to a platform content recognition method and system based on AIGC. Background Art

[0002] With the rapid development of artificial intelligence (AI), content generation and recognition technologies have been widely applied in various fields. In AI-generated content (AIGC) platforms, content recognition technology is a key component in intelligently processing and generating content. However, traditional content recognition methods often rely on a single model or algorithm, which not only limits the accuracy and efficiency of content recognition but also makes it difficult to cope with complex and diverse application scenarios.

[0003] In the current AIGC platform management system, the content recognition process is usually relatively simple, that is, after receiving user input, it is processed through a single recognition model and feedback is generated. This approach has the following shortcomings:

[0004] Limited accuracy: When a single model processes complex or diverse inputs, it may suffer from recognition errors due to insufficient generalization ability.

[0005] Poor adaptability: A single model is difficult to flexibly adjust and optimize to meet the needs of different application scenarios, which limits the scalability and maintainability of the system. Summary of the Invention

[0006] In order to solve the problems existing in the prior art, the present invention proposes a platform content recognition method and system based on AIGC.

[0007] The technical solutions of the present invention are as follows:

[0008] An AIGC-based platform content identification method is applied to an AIGC platform management system including an AIGC engine cluster and an AIGC management platform, wherein each AIGC engine in the AIGC engine cluster is connected to the AIGC management platform. The method includes:

[0009] Obtaining first prompt information through the AIGC management platform, and inputting the first prompt information into a first AIGC engine and a second AIGC engine in the AIGC engine cluster;

[0010] The first AIGC engine generates first feedback information in response to the first prompt information, and submits the first feedback information to the AIGC management platform; the second AIGC engine generates second feedback information in response to the first prompt information, and submits the second feedback information to the AIGC management platform;

[0011] The AIGC management platform determines first application scenario information corresponding to the first feedback information and second application scenario information corresponding to the second feedback information respectively;

[0012] If the AIGC management platform determines that the first scenario similarity between the first application scenario information and the second application scenario information is greater than a preset similarity threshold, the AIGC management platform outputs first synthetic information, where the first synthetic information is synthetic information of the first feedback information and the second feedback information.

[0013] Furthermore, if the AIGC management platform determines that the first scenario similarity between the first application scenario information and the second application scenario information is less than the preset similarity threshold, the AIGC management platform outputs the first feedback information or the second feedback information.

[0014] Furthermore, the AIGC management platform outputs the first feedback information or the second feedback information, including:

[0015] The AIGC management platform inputs the first prompt information to a third AIGC engine in the AIGC engine cluster, so that the AIGC management platform generates third feedback information in response to the first prompt information, and submits the third feedback information to the AIGC management platform;

[0016] The AIGC management platform determines third application scenario information corresponding to the third feedback information, and determines a second scenario similarity between the first application scenario information and the third application scenario information, and a third scenario similarity between the second application scenario information and the third application scenario information;

[0017] If the AIGC management platform determines that the second scenario similarity is greater than the third scenario similarity, the AIGC management platform outputs the first feedback information;

[0018] If the AIGC management platform determines that the second scene similarity is less than the third scene similarity, the AIGC management platform outputs the second feedback information.

[0019] Further, if the AIGC management platform determines that the first scenario similarity between the first application scenario information and the second application scenario information is less than the preset similarity threshold, the AIGC management platform inputs the first prompt information to each AIGC engine in the AIGC engine cluster to generate a feedback information set;

[0020] The AIGC management platform uses a preset classifier to identify application scenario information corresponding to each piece of feedback information in the feedback information set to generate an application scenario information set;

[0021] The AIGC management platform generates a scenario similarity matrix based on the application scenario information set;

[0022] The AIGC management platform determines the classification feature vector corresponding to the feedback information set according to the scene similarity matrix;

[0023] The AIGC management platform determines target feedback information based on the classification feature vector, where the target feedback information is feedback information corresponding to the target classification feature value;

[0024] The AIGC management platform outputs the target feedback information.

[0025] Furthermore, the AIGC management platform determines first application scenario information corresponding to the first feedback information and second application scenario information corresponding to the second feedback information, including:

[0026] The AIGC management platform performs feature extraction on the first feedback information to generate a first feature word sequence; the AIGC management platform performs feature extraction on the second feedback information to generate a second feature word sequence;

[0027] The AIGC management platform inputs the first feature word sequence and the second feature word sequence into a preset application scenario classifier, respectively, to determine the first application scenario information and the second application scenario information, respectively.

[0028] Furthermore, before the AIGC management platform determines the first application scenario information corresponding to the first feedback information and the second application scenario information corresponding to the second feedback information, the method further includes:

[0029] Constructing a training data set, wherein each training sample in the training data set includes question information and annotated application scenario information, wherein the question information includes a question statement about the product;

[0030] According to each training sample in the constructed training data set, a corresponding question feature word sequence is generated including question information, and the question feature word sequence and the corresponding annotated application scenario information are used to train a preset initial classifier to generate the preset application scenario classifier, wherein the preset initial classifier is a classifier model based on a vector machine.

[0031] Furthermore, after the AIGC management platform outputs the target feedback information, the method further includes:

[0032] If the AIGC management platform receives the first feedback instruction and, after re-outputting new target feedback information, receives the second feedback instruction, first feature data is generated, where the first feedback instruction includes an instruction for instructing to regenerate content corresponding to the first prompt information, the second feedback instruction is an instruction for accepting the new target feedback information, and the first feature data includes the first prompt information, the target feedback information, and the new target feedback information.

[0033] The AIGC management platform adds the first feature data to the training data set to iteratively train the preset application scenario classifier.

[0034] An AIGC-based platform content recognition system is applied to an AIGC platform management system including an AIGC engine cluster and an AIGC management platform, wherein each AIGC engine in the AIGC engine cluster is connected to the AIGC management platform. The system includes a first data processing module, a second data processing module, a third data processing module, and a fourth data processing module.

[0035] The first data processing module is configured to obtain first prompt information through the AIGC management platform, and input the first prompt information into the first AIGC engine and the second AIGC engine in the AIGC engine cluster;

[0036] the second data processing module is configured to cause the first AIGC engine to generate first feedback information in response to the first prompt information and submit the first feedback information to the AIGC management platform; and the second AIGC engine to generate second feedback information in response to the first prompt information and submit the second feedback information to the AIGC management platform;

[0037] the third data processing module is configured to enable the AIGC management platform to determine first application scenario information corresponding to the first feedback information and second application scenario information corresponding to the second feedback information;

[0038] The fourth data processing module is configured to output synthesized information based on a first scenario similarity between the first application scenario information and the second application scenario information. If the AIGC management platform determines that the first scenario similarity between the first application scenario information and the second application scenario information is greater than a preset similarity threshold, the AIGC management platform outputs first synthesized information, where the first synthesized information is synthesized information of the first feedback information and the second feedback information.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] This paper proposes a platform content recognition method based on AIGC, which improves recognition accuracy and efficiency. By introducing a diversified AIGC engine, this solution can cope with complex and changing content recognition needs. Different engines may excel at processing different types or fields of information, thus providing more accurate and professional feedback. At the same time, intelligent analysis and merging functions further enhance recognition accuracy. In addition, because this solution uses parallel processing, it can significantly improve content recognition efficiency.

[0041] This approach enhances system flexibility and scalability: Compared to traditional single models or algorithms, the AIGC engine cluster and management platform in this solution are easier to upgrade and optimize. As AI technology continues to advance, new, more advanced engines can be continuously introduced to replace older ones, thereby improving overall system performance. Furthermore, the management platform supports dynamic adjustment and optimization of the recognition process based on evolving user needs.

[0042] The method of the present invention provides a richer and more diverse selection of information: Because it can generate multiple pieces of feedback information and allow users to select the most appropriate answer based on the similarity of application scenarios, it provides users with a richer and more diverse selection of information. This helps users understand the problem more comprehensively and make more informed decisions.

[0043] The method of this invention achieves continuous learning and optimization: This solution also has the ability to continuously learn and optimize. Every time a user provides feedback, the system records this information and uses it to optimize the classifier and improve future recognition accuracy. Thus, over time, the system will better understand the user, providing more attentive and accurate service. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a flowchart of the AIGC-based platform content identification method.

[0045] Figure 2 This is the second flow chart of the platform content identification method based on AIGC. DETAILED DESCRIPTION

[0046] The present invention is further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention. After reading the present invention, various equivalent modifications of the present invention by those skilled in the art fall within the scope defined by the claims attached to this application.

[0047] Example 1:

[0048] The present invention provides an AIGC-based platform content recognition method, which is applied to an AIGC platform management system including an AIGC engine cluster and an AIGC management platform. Each AIGC engine in the AIGC engine cluster is connected to the AIGC management platform. Figure 1 As shown, the method includes:

[0049] S101: Obtain first prompt information through the AIGC management platform, and input the first prompt information to a first AIGC engine and a second AIGC engine in the AIGC engine cluster;

[0050] In this step, the AIGC management platform receives first prompt information input by the user or generated by the system. The prompt information can be a description or keyword about a certain topic, product, question, etc.

[0051] Specifically, users can be allowed to input prompt information about products, services, or questions through text input or voice input. For example, users can enter the product name, description, or related keywords in the search box, or ask questions through the voice assistant. For voice input, voice recognition technology can be used to convert voice signals into text information. This requires the system to have an efficient voice recognition module that can accurately convert the user's voice command into the first prompt information in text form.

[0052] The first prompt information received is pre-processed, including removing irrelevant characters, correcting spelling errors, and performing synonym replacement to improve the accuracy and efficiency of subsequent AIGC engine processing. In particular, for product names, the system may need to identify and process different meanings in different contexts.

[0053] The pre-processed first prompt information is sent to multiple engines in the AIGC engine cluster (such as the first AIGC engine and the second AIGC engine in this example) at the same time. These engines may generate different feedback information based on the same prompt information.

[0054] S102: The first AIGC engine and the second AIGC engine generate first feedback information and second feedback information respectively in response to the first prompt information;

[0055] In this step, the first AIGC engine generates first feedback information in response to the first prompt information and submits the first feedback information to the AIGC management platform; the second AIGC engine generates second feedback information in response to the first prompt information and submits the second feedback information to the AIGC management platform.

[0056] Specifically, the received first prompt information is simultaneously input to the first and second AIGC engines in the AIGC engine cluster. These two engines may generate different feedback information based on the same prompt information. For example, if the first prompt information input by the user is incomplete, the output of different AIGC engines may be inconsistent. The first AIGC engine generates first feedback information in response to the first prompt information and submits it to the AIGC management platform. The second AIGC engine also generates second feedback information in response to the first prompt information and submits it to the AIGC management platform.

[0057] S103. The AIGC management platform determines first application scenario information corresponding to the first feedback information and second application scenario information corresponding to the second feedback information respectively;

[0058] Specifically, the AIGC management platform processes the received first and second feedback information to determine their corresponding first and second application scenario information, respectively. This step can be achieved by extracting features from the feedback information and then inputting them into a preset application scenario classifier. Specifically, the AIGC management platform can extract features from the first feedback information to generate a first feature word sequence, which is then input into the preset application scenario classifier to determine the first application scenario information. Similarly, feature extraction and classification are performed on the second feedback information to determine the second application scenario information.

[0059] S104: The AIGC management platform outputs synthesized information based on the first scenario similarity between the first application scenario information and the second application scenario information. In this step, the AIGC management platform calculates the first scenario similarity between the first application scenario information and the second application scenario information. This step can be achieved by comparing feature words, semantic relationships, etc. in the two application scenario information.

[0060] If the AIGC management platform determines that the first scenario similarity between the first application scenario information and the second application scenario information is greater than a preset similarity threshold, the AIGC management platform outputs first synthetic information, which is a synthetic information of the first feedback information and the second feedback information, and can provide more comprehensive and accurate information. In the case where the similarity of the application scenario information is not high, outputting a single feedback information instead of synthetic information can avoid the trouble caused to users due to redundant or inconsistent information. This processing method improves the system's ability to screen and filter information, making the output information more accurate and targeted. In addition, by choosing to output the first feedback information or the second feedback information, the AIGC management platform can ensure that users receive the information that is most likely to meet their needs.

[0061] Specifically, irrelevant characters, redundant information, and special symbols can be removed from the first and second feedback messages to ensure their purity. Regular expressions, natural language processing, and other technologies can be used to pre-process the first and second feedback messages, such as removing HTML tags and special characters and correcting spelling errors. Key features can then be extracted from the cleaned feedback information. By comparing keywords in the two feedback messages, commonalities and differences can be identified to prepare for integration. The commonalities and differences between the two feedback messages can be integrated to avoid duplication and omissions. A grammar checker can then be used to perform a grammar check on the synthesized information. The synthesized information can then be presented to users via the AIGC management platform.

[0062] In another possible implementation, the first feedback information and the second feedback information may be directly displayed side by side. Furthermore, the first feedback information and the second feedback information may be re-input into the first AIGC engine, the second AIGC engine, or another AIGC engine, and a prompt word may be input to indicate that the first feedback information and the second feedback information should be merged to output the synthesized information.

[0063] In the above solution, the first prompt information is received by the AIGC management platform and distributed to two AIGC engines (the first AIGC engine and the second AIGC engine) for processing. This parallel processing method significantly improves the speed and efficiency of content generation. At the same time, the two engines generate different feedback information based on the same prompt information, providing a content foundation for subsequent application scenario analysis and information synthesis. Then, the first application scenario information corresponding to the first feedback information and the second application scenario information corresponding to the second feedback information are identified and distinguished. When the AIGC management platform determines that the similarity between the two application scenario information exceeds a preset similarity threshold, it automatically synthesizes the first feedback information and the second feedback information to generate a more comprehensive and accurate first synthesized information. This not only improves the accuracy and reliability of the information, but also provides users with a richer and more diverse information selection.

[0064] Furthermore, if the AIGC management platform determines that the first scenario similarity between the first application scenario information and the second application scenario information is less than a preset similarity threshold, the AIGC management platform outputs either the first feedback information or the second feedback information. This selection can be achieved through further analysis and comparison, for example, by introducing a third AIGC engine to generate third feedback information and making a decision based on the similarity between the application scenario information in the third feedback information and the first and second application scenario information.

[0065] In the above solution, when the similarity of application scenario information is not high, outputting a single feedback message instead of synthesized information can avoid user confusion caused by redundant or inconsistent information. This processing method improves the system's ability to screen and filter information, making the output information more accurate and targeted. In addition, by choosing to output the first or second feedback information, the AIGC management platform can ensure that users receive the information most likely to meet their needs.

[0066] Furthermore, the AIGC management platform outputs the first feedback information or the second feedback information, including:

[0067] The AIGC management platform inputs the first prompt information to a third AIGC engine in the AIGC engine cluster, so that the AIGC management platform generates third feedback information in response to the first prompt information, and submits the third feedback information to the AIGC management platform;

[0068] The AIGC management platform determines the third application scenario information corresponding to the third feedback information, and determines the second scenario similarity between the first application scenario information and the third application scenario information, and the third scenario similarity between the second application scenario information and the third application scenario information;

[0069] If the AIGC management platform determines that the second scenario similarity is greater than the third scenario similarity, the AIGC management platform outputs first feedback information;

[0070] If the AIGC management platform determines that the second scene similarity is less than the third scene similarity, the AIGC management platform outputs second feedback information.

[0071] In the above solution, by introducing the third AIGC engine and the third feedback information, the system can obtain more relevant content about the first prompt information, thereby increasing the dimension and accuracy of information recognition. This triple verification method helps to reduce the error rate of information recognition and improve the recognition accuracy of the system. When determining which feedback information to output, the system not only considers the application scenario information of the first and second feedback information, but also introduces the application scenario information of the third feedback information for comparison. This comprehensive consideration of multiple factors makes the system's decision-making more comprehensive and reasonable, and helps to improve the overall performance of the system. By comparing the similarity of the second scenario and the similarity of the third scenario, the system can intelligently choose to output the feedback information that is most relevant to the first prompt information. This personalized output strategy helps to enhance the user experience and satisfaction, and also helps to improve the system's information push efficiency.

[0072] In this embodiment, the first prompt information is received by the AIGC management platform and distributed to two AIGC engines for processing. The two engines generate different feedback information based on the same prompt information. Then, the two engines identify and distinguish the first application scenario information corresponding to the first feedback information and the second application scenario information corresponding to the second feedback information. When the AIGC management platform determines that the similarity between the two application scenario information exceeds a preset similarity threshold, the first feedback information and the second feedback information are automatically synthesized to generate more comprehensive and accurate first synthesized information. This not only improves the accuracy and reliability of the information, but also provides users with richer and more diverse information choices.

[0073] It is worth noting that when the above solution is applied to an e-commerce dialogue management system, it can not only quickly output relevant information about the questions asked by users, but also cross-verify the content generated by multiple different AIGC engines, so that only when it is determined that the information of two application scenarios is highly similar or even the same, the content generated by the AIGC engine is directly output, thereby reducing the situation where the output content does not match the question content input by the user because the product name represents different meanings in different application scenarios.

[0074] Example 2:

[0075] The present invention provides a platform content identification method based on AIGC, comprising:

[0076] S201: Obtain first prompt information through the AIGC management platform, and input the first prompt information to a first AIGC engine and a second AIGC engine in an AIGC engine cluster.

[0077] In this step, the AIGC management platform receives first prompt information, either input by the user or generated by the system. This prompt information can be a description or keywords about a topic, product, question, etc. Optionally, the AIGC platform management system can be used to access the e-commerce dialogue management system. The first prompt information includes product feature words input by the customer. Product feature words are used to represent different target objects in different application scenarios.

[0078] Specifically, users can be allowed to input prompt information about products, services, or questions through text input or voice input. For example, users can enter the product name, description, or related keywords in the search box, or ask questions through the voice assistant. For voice input, voice recognition technology can be used to convert voice signals into text information. This requires the system to have an efficient voice recognition module that can accurately convert the user's voice command into the first prompt information in text form.

[0079] The first prompt information received is pre-processed, including removing irrelevant characters, correcting spelling errors, and performing synonym replacement to improve the accuracy and efficiency of subsequent AIGC engine processing. In particular, for product names, the system may need to identify and process different meanings in different contexts.

[0080] The pre-processed first prompt information is sent to multiple engines in the AIGC engine cluster (such as the first AIGC engine and the second AIGC engine in this example) at the same time. These engines may generate different feedback information based on the same prompt information.

[0081] In addition, in a possible application scenario, the first AIGC engine and the second AIGC engine may both be general AIGC engines, so as to output more accurate information to the user by comparing the application scenario information of the contents of different AIGC engines.

[0082] It is worth noting that in another possible application scenario, the first AIGC engine mentioned above can be an AIGC engine within the e-commerce dialogue management system, and the content it generates is based on a database built by the internal e-commerce company itself. The training data in this database is data that is more suitable for the e-commerce field, while the second AIGC engine can be an external general AIGC engine. Since the database on which the internal AIGC engine is based has high accuracy but weak generalization ability, and the database on which the external AIGC engine is based has a wide range of data but relatively low accuracy, it is possible to compare the application scenario information of the content generated by the external AIGC engine with the content generated by the internal AIGC engine to determine whether the scene directionality of the content generated by the internal AIGC engine is the same as that of the content generated by the external AIGC engine. If the scene directionality is the same, the composite content of the two can be output, thereby ensuring that the generated content has high generalization and good pertinence.

[0083] Furthermore, it's worth noting that the first AIGC engine mentioned above can also be an AIGC engine currently being trained or tested within the e-commerce dialogue management system, while the second AIGC engine can be an external, general-purpose AIGC engine. In this case, the first prompt information mentioned above can be the relevant test prompt statement input by the tester. By comparing the application scenario information of the content generated by the external and internal AIGC engines, it is determined whether the scenario directionality of the content generated by the internal AIGC engine is the same as that of the content generated by the external AIGC engine, thereby iteratively optimizing and training the internal AIGC engine. Here, the generation and iteration of training data in the internal AIGC engine training database can also be combined with tester feedback.

[0084] S202: The first AIGC engine and the second AIGC engine generate first feedback information and second feedback information respectively in response to the first prompt information.

[0085] In this step, the first AIGC engine generates first feedback information in response to the first prompt information and submits the first feedback information to the AIGC management platform; the second AIGC engine generates second feedback information in response to the first prompt information and submits the second feedback information to the AIGC management platform.

[0086] S203. The AIGC management platform determines first application scenario information corresponding to the first feedback information and second application scenario information corresponding to the second feedback information.

[0087] Specifically, the AIGC management platform processes the received first and second feedback information to determine their corresponding first and second application scenario information, respectively. This step can be achieved by extracting features from the feedback information and then inputting them into a preset application scenario classifier. Specifically, the AIGC management platform can extract features from the first feedback information to generate a first feature word sequence, which is then input into the preset application scenario classifier to determine the first application scenario information. Similarly, feature extraction and classification are performed on the second feedback information to determine the second application scenario information.

[0088] In one possible implementation, the AIGC management platform performs feature extraction on the first feedback information to generate a first feature word sequence; and performs feature extraction on the second feedback information to generate a second feature word sequence. The AIGC management platform inputs the first feature word sequence and the second feature word sequence into a preset application scenario classifier to determine the first application scenario information and the second application scenario information, respectively.

[0089] Specifically, first, the AIGC management platform receives first and second feedback information from different sources. This information may contain data in various forms, such as text, images, audio, or video. In order to effectively identify the application scenarios to which this information belongs, the platform first needs to extract features from this feedback information. The platform can use natural language processing technology or corresponding multimedia processing technology to analyze the first feedback information. For text information, steps such as word segmentation, part-of-speech tagging, and named entity recognition are performed to generate a first feature word sequence representing the core content of the information. For non-text information, such as images or videos, computer vision technology may be used to extract key visual features and convert them into a processable feature vector form. Similar to the first feedback information, the platform also performs feature extraction on the second feedback information to generate a second feature word sequence or feature vector. This step ensures that regardless of the form of the information, it can be converted into a format that the platform can understand and compare.

[0090] Next, the AIGC management platform uses a preset application scenario classifier to classify the extracted feature word sequences or feature vectors to determine the application scenario information corresponding to each feedback information. In the preliminary preparation stage, the platform needs to train an efficient application scenario classifier based on a well-labeled data set. This classifier can be built based on a machine learning algorithm (such as a support vector machine, random forest) or a deep learning model (such as a convolutional neural network, a recurrent neural network). The first feature word sequence and the second feature word sequence extracted in the first step are respectively input into the trained application scenario classifier. By analyzing and learning these features, the classifier can identify the application scenario category to which each feedback information is most likely to belong. The classifier outputs two results, corresponding to the application scenario information of the first feedback information and the second feedback information respectively. These application scenarios may include but are not limited to multiple application fields such as food, furniture, electrical appliances, clothing, etc., depending on the design and service scope of the platform.

[0091] Specifically, before the AIGC management platform determines the first application scenario information corresponding to the first feedback information and the second application scenario information corresponding to the second feedback information, it is necessary to construct a training data set. Each training sample in the training data set includes question information and annotated application scenario information. The question information includes a question statement about the product. According to each training sample in the training data set, a corresponding question feature word sequence is generated, and the question feature word sequence and the corresponding annotated application scenario information are used to train a preset initial classifier to generate a preset application scenario classifier, wherein the preset initial classifier is a classifier model based on a vector machine. Among them, a large amount of question information can be collected, and the collected question information and the corresponding annotated application scenario information are organized into a structured data format to form a training data set. Each training sample in the training data set should include question information and annotated application scenario information. The question information in the training data set is preprocessed, including removing stop words, word segmentation, stem extraction, etc., to purify the data and reduce noise. Natural language processing techniques, such as TF-IDF (Term Frequency-Inverse Document Frequency) or word embeddings (such as Word2Vec and BERT), are then used to convert the preprocessed question information into a sequence of feature words or a vector representation. An initial classifier model based on a vector machine (such as a support vector machine (SVM)) is then used as the basis for training. The initial classifier is trained using the question feature word sequence and the corresponding annotated application scenario information in the training dataset. During training, the classifier learns how to map the question feature word sequence to the correct application scenario information. The classifier's performance is then optimized through methods such as cross-validation and hyperparameter adjustment. The validation dataset is used to evaluate the classifier's performance metrics, such as accuracy and recall, to ensure good generalization. After training and optimization, a final pre-set application scenario classifier is generated that accurately classifies question information into the corresponding application scenario. Upon receiving the first and second feedback information, the AIGC management platform first uses the trained application scenario classifier to classify the question information in the feedback information, determining the first application scenario information corresponding to the first feedback information and the second application scenario information corresponding to the second feedback information. It is worth noting that the above-mentioned preset application scenario classifier can be trained based on various general classifier basic models. In this embodiment, there is no specific limitation on its specific form, and the support vector machine based on it is an exemplary implementation method.

[0092] S204: The AIGC management platform inputs the first prompt information to each AIGC engine in the AIGC engine cluster to generate a feedback information set.

[0093] If the AIGC management platform determines that the first scenario similarity between the first application scenario information and the second application scenario information is less than the preset similarity threshold, the AIGC management platform inputs the first prompt information into the AIGC engine cluster E={e1, e2,…, e i ,…,e n} in each AIGC engine to generate a feedback information set M = {m1, m2, ..., m i ,…,m n}, where n is the number of AIGC engines in the AIGC engine cluster E.

[0094] S205. The AIGC management platform uses a preset classifier to identify application scenario information corresponding to each piece of feedback information in the feedback information set to generate an application scenario information set.

[0095] The AIGC management platform uses a preset classifier to identify the application scenario information corresponding to each feedback information in the feedback information set M to generate an application scenario information set D = {d1, d2, ..., d i ,…,d n}.

[0096] S206. The AIGC management platform generates a scenario similarity matrix based on the application scenario information set.

[0097] The AIGC management platform generates a scenario similarity matrix S based on the application scenario information set D:

[0098]

[0099] Among them, s in the scene similarity matrix S ij is the similarity between the i-th application scenario information and the j-th application scenario information in the application scenario information set D.

[0100] S207. The AIGC management platform determines the classification feature vector corresponding to the feedback information set based on the scene similarity matrix.

[0101] The AIGC management platform uses Formula 1 and determines the classification feature vector K = {k1, k2, ..., k i ,…,k n}, where k i is the classification feature value corresponding to the i-th feedback information in the feedback information set M. Formula 1 is:

[0102]

[0103] S208. The AIGC management platform determines target feedback information based on the classification feature vector.

[0104] The AIGC management platform uses formula 2 and determines the target feedback information based on the classification feature vector K. The target feedback information is the target classification feature value The corresponding feedback information, formula 2 is:

[0105]

[0106] S209. The AIGC management platform outputs target feedback information.

[0107] In the above scheme, by comparing the similarity between the first application scenario information and the second application scenario information, and when the similarity is less than a preset threshold, the AIGC engine cluster is started to generate a feedback information set. This step ensures that when the scene information does not match or the difference is large, the possibility of outputting matching content can be increased through AIGC engine cluster recognition.

[0108] The preset classifier is used to identify the application scenario information of each feedback information in the feedback information set, and an application scenario information set is generated, which further refines the granularity of content identification and improves the accuracy of identification. By calculating the similarity between each application scenario information in the application scenario information set, a scenario similarity matrix is constructed. This step can fully reflect the correlation and difference between each application scenario. Then, the classification feature vector is extracted, and the scene similarity matrix is weighted and summed using the above formula 1, which reflects the overall accuracy of each feedback information in the feedback information set generated by the AIGC engine cluster. Then, the target classification feature value is determined using formula 2, and the target feedback information is selected accordingly. This step can automatically screen out the most representative and relevant feedback information.

[0109] Furthermore, after the AIGC management platform outputs target feedback information, if it receives a first feedback instruction and, after re-outputting new target feedback information, receives a second feedback instruction, it generates first feature data. The first feedback instruction includes an instruction for instructing the regeneration of content corresponding to the first prompt information, and the second feedback instruction is an instruction for accepting the new target feedback information. The first feature data includes the first prompt information, the target feedback information, and the new target feedback information. The AIGC management platform adds the first feature data to the training dataset to iteratively train the preset application scenario classifier. In the above scheme, by introducing a user feedback mechanism, when a user requests the regeneration of the initially generated content (i.e., the first prompt information) through the first feedback instruction, the platform generates new target feedback information based on this feedback. If the user accepts this new content (i.e., issues a second feedback instruction), this indicates that the new content better meets user needs. This mechanism ensures that the generated content is closer to user expectations, thereby improving content quality and satisfaction. The platform integrates the user's feedback data (including the first prompt information, the original target feedback information, and the regenerated target feedback information) into the first feature data and adds it to the training data set, so as to continuously enrich and optimize the model's training data and enable the model to learn more about the user's question-based preference information.

[0110] It's also worth noting that after the AIGC management platform outputs the target feedback information, if it receives a second prompt and outputs new target feedback information based on the second prompt, and then receives a third feedback instruction, it generates second feature data. The third feedback instruction is an instruction to accept the new target feedback information. The second feature data includes the first prompt, the second prompt, and the new target feedback information. The AIGC management platform adds the second feature data to the training dataset to iteratively train the preset application scenario classifier. In the above solution, after the AIGC management platform completes content recognition and determines the target feedback information, it outputs this information to the user. This feedback information is generated based on the first prompt information (e.g., question statement) provided by the user and the processing results of the preset application scenario classifier. If the user provides additional information or modifies the previous question after receiving the initial target feedback information, the AIGC management platform will receive this new prompt information, i.e., the second prompt information. The platform inputs the second prompt information into the preset application scenario classifier and, combined with the previous processing results, generates new target feedback information. The new target feedback information is then output to the user to meet the user's adjusted needs based on the second prompt information. If the user accepts or approves the new target feedback information and sends a corresponding feedback instruction through the AIGC management platform, that is, the third feedback instruction. According to the third feedback instruction, the AIGC management platform integrates the first prompt information, the second prompt information, and the finally accepted new target feedback information to generate the second feature data. This data contains the complete process information from the user's initial question to the final acceptance of the feedback information. The generated second feature data is added to the original training data set to enrich the content of the data set so that it contains more data from actual usage scenarios. The preset application scenario classifier is iteratively trained using the updated training data set. This process is designed to enable the classifier to learn more changes and user feedback in actual scenarios, thereby further improving its classification accuracy and adaptability. During the iterative training process, the performance of the classifier is regularly evaluated, and corresponding optimization adjustments are made based on the evaluation results to ensure that the classifier can continue to provide high-quality application scenario classification services.

[0111] Furthermore, after the AIGC management platform outputs the target feedback information, if the AIGC management platform receives a second prompt message and outputs new target feedback information based on the second prompt message, then receives a fourth feedback instruction and determines that the textual similarity between the first prompt message and the second prompt message exceeds a preset similarity threshold, and the second prompt message includes characteristic keywords that do not appear in the first prompt message, then generates third feature data. The third feature data includes the first prompt message, the second prompt message, and the new target feedback information. The fourth feedback instruction is an instruction to accept the new target feedback information. The AIGC management platform adds the third feature data to the training dataset and configures the weight of the characteristic keywords in subsequent training to be greater than the preset weight threshold, thereby iteratively training the preset application scenario classifier. In this solution, when the AIGC management platform receives the user's second prompt message and outputs new target feedback information based on it, if the user accepts this new feedback (i.e., sends the fourth feedback instruction), and the system determines that the first prompt message and the second prompt message have high textual similarity but contain newly added characteristic keywords, the system generates third feature data. This process ensures that the classifier learns that the user's first response to a question doesn't meet their requirements. However, if the user adds qualifiers to the first response and asks a second question, the response based on the second response meets their requirements. After generating the third feature data, the AIGC management platform not only adds it to the training dataset but also specifically sets the weight of the newly added feature keywords in subsequent training to a value higher than the preset weight threshold. This measure emphasizes the importance of these keywords in classifier training, helping the classifier to more accurately identify and process this key information, thereby improving content recognition accuracy. By incorporating third feature data containing new feature keywords and diverse expressions into the training dataset, the AIGC management platform is able to conduct more comprehensive iterative training of the classifier for pre-defined application scenarios. This process not only enriches the diversity of training samples but also enables the classifier to focus more on feature keywords that frequently appear in real applications and have a significant impact on classification results during training. Through this technical process, the AIGC management platform can more accurately understand users' true intentions and provide feedback that better meets their expectations. This helps to improve user satisfaction and loyalty while strengthening the platform's competitiveness and market position. Furthermore, as classifier performance continues to improve, the platform's content recognition and processing capabilities will be significantly enhanced, providing users with a more efficient and accurate service experience. By continuously receiving user feedback, generating feature data, and conducting iterative training, the AIGC management platform achieves self-learning and evolution.

[0112] Example 3:

[0113] The present invention provides an AIGC-based platform content recognition system, which is applied to an AIGC platform management system including an AIGC engine cluster and an AIGC management platform. Each AIGC engine in the AIGC engine cluster is connected to the AIGC management platform. The system includes a first data processing module, a second data processing module, a third data processing module, and a fourth data processing module.

[0114] A first data processing module is configured to obtain first prompt information through the AIGC management platform and input the first prompt information to a first AIGC engine and a second AIGC engine in the AIGC engine cluster;

[0115] a second data processing module configured to cause the first AIGC engine to generate first feedback information in response to the first prompt information and submit the first feedback information to the AIGC management platform; and to cause the second AIGC engine to generate second feedback information in response to the first prompt information and submit the second feedback information to the AIGC management platform;

[0116] a third data processing module, configured for the AIGC management platform to determine first application scenario information corresponding to the first feedback information and second application scenario information corresponding to the second feedback information;

[0117] The fourth data processing module is used to output synthetic information based on the first scenario similarity between the first application scenario information and the second application scenario information. If the AIGC management platform determines that the first scenario similarity between the first application scenario information and the second application scenario information is greater than a preset similarity threshold, the AIGC management platform outputs first synthetic information, where the first synthetic information is synthetic information of the first feedback information and the second feedback information.

[0118] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by a technician familiar with this technical field within the technical scope disclosed by the present invention should be covered by the scope of protection of the present invention.

Claims

1. A platform content recognition method based on AIGC, applied to an AIGC platform management system including an AIGC engine cluster and an AIGC management platform, wherein each AIGC engine in the AIGC engine cluster is connected to the AIGC management platform, characterized in that: The method comprises: Obtaining first prompt information through the AIGC management platform, and inputting the first prompt information into a first AIGC engine and a second AIGC engine in the AIGC engine cluster; The first AIGC engine generates first feedback information in response to the first prompt information, and submits the first feedback information to the AIGC management platform; the second AIGC engine generates second feedback information in response to the first prompt information, and submits the second feedback information to the AIGC management platform; The AIGC management platform determines first application scenario information corresponding to the first feedback information and second application scenario information corresponding to the second feedback information respectively; If the AIGC management platform determines that the first scenario similarity between the first application scenario information and the second application scenario information is greater than a preset similarity threshold, the AIGC management platform outputs first synthetic information, where the first synthetic information is synthetic information of the first feedback information and the second feedback information.

2. The platform content identification method based on AIGC according to claim 1, characterized in that: If the AIGC management platform determines that the first scenario similarity between the first application scenario information and the second application scenario information is less than the preset similarity threshold, the AIGC management platform outputs the first feedback information or the second feedback information.

3. The platform content identification method based on AIGC according to claim 2, characterized in that: The AIGC management platform outputting the first feedback information or the second feedback information includes: The AIGC management platform inputs the first prompt information to a third AIGC engine in the AIGC engine cluster, so that the AIGC management platform generates third feedback information in response to the first prompt information, and submits the third feedback information to the AIGC management platform; The AIGC management platform determines third application scenario information corresponding to the third feedback information, and determines a second scenario similarity between the first application scenario information and the third application scenario information, and a third scenario similarity between the second application scenario information and the third application scenario information; If the AIGC management platform determines that the second scenario similarity is greater than the third scenario similarity, the AIGC management platform outputs the first feedback information; If the AIGC management platform determines that the second scene similarity is less than the third scene similarity, the AIGC management platform outputs the second feedback information.

4. The platform content identification method based on AIGC according to claim 1, characterized in that: If the AIGC management platform determines that the first scenario similarity between the first application scenario information and the second application scenario information is less than the preset similarity threshold, the AIGC management platform inputs the first prompt information to each AIGC engine in the AIGC engine cluster to generate a feedback information set; The AIGC management platform uses a preset classifier to identify application scenario information corresponding to each piece of feedback information in the feedback information set to generate an application scenario information set; The AIGC management platform generates a scenario similarity matrix based on the application scenario information set; The AIGC management platform determines the classification feature vector corresponding to the feedback information set according to the scene similarity matrix; The AIGC management platform determines target feedback information based on the classification feature vector, where the target feedback information is feedback information corresponding to the target classification feature value; The AIGC management platform outputs the target feedback information.

5. The platform content identification method based on AIGC according to claim 1, characterized in that: The AIGC management platform determines, respectively, first application scenario information corresponding to the first feedback information and second application scenario information corresponding to the second feedback information, including: The AIGC management platform performs feature extraction on the first feedback information to generate a first feature word sequence; the AIGC management platform performs feature extraction on the second feedback information to generate a second feature word sequence; The AIGC management platform inputs the first feature word sequence and the second feature word sequence into a preset application scenario classifier, respectively, to determine the first application scenario information and the second application scenario information, respectively.

6. The platform content identification method based on AIGC according to claim 1, characterized in that: Before the AIGC management platform determines the first application scenario information corresponding to the first feedback information and the second application scenario information corresponding to the second feedback information, the method further includes: Constructing a training data set, wherein each training sample in the training data set includes question information and annotated application scenario information, wherein the question information includes a question statement about the product; According to each training sample in the constructed training data set, a corresponding question feature word sequence is generated including question information, and the question feature word sequence and the corresponding annotated application scenario information are used to train a preset initial classifier to generate the preset application scenario classifier, wherein the preset initial classifier is a classifier model based on a vector machine.

7. The platform content identification method based on AIGC according to claim 6, characterized in that: After the AIGC management platform outputs the target feedback information, the method further includes: If the AIGC management platform receives the first feedback instruction and, after re-outputting new target feedback information, receives the second feedback instruction, first feature data is generated, where the first feedback instruction includes an instruction for instructing to regenerate content corresponding to the first prompt information, the second feedback instruction is an instruction for accepting the new target feedback information, and the first feature data includes the first prompt information, the target feedback information, and the new target feedback information. The AIGC management platform adds the first feature data to the training data set to iteratively train the preset application scenario classifier.

8. An AIGC-based platform content recognition system, applied to an AIGC platform management system including an AIGC engine cluster and an AIGC management platform, wherein each AIGC engine in the AIGC engine cluster is connected to the AIGC management platform, characterized in that: The system includes a first data processing module, a second data processing module, a third data processing module and a fourth data processing module; The first data processing module is configured to obtain first prompt information through the AIGC management platform, and input the first prompt information into the first AIGC engine and the second AIGC engine in the AIGC engine cluster; the second data processing module is configured to cause the first AIGC engine to generate first feedback information in response to the first prompt information and submit the first feedback information to the AIGC management platform; and the second AIGC engine to generate second feedback information in response to the first prompt information and submit the second feedback information to the AIGC management platform; the third data processing module is configured to enable the AIGC management platform to determine first application scenario information corresponding to the first feedback information and second application scenario information corresponding to the second feedback information; The fourth data processing module is configured to output synthesized information based on a first scenario similarity between the first application scenario information and the second application scenario information. If the AIGC management platform determines that the first scenario similarity between the first application scenario information and the second application scenario information is greater than a preset similarity threshold, the AIGC management platform outputs first synthesized information, where the first synthesized information is synthesized information of the first feedback information and the second feedback information.

Images