2846 results about "Categorization" patented technology

The invention discloses a traffic engineering large model intelligent question answering system and method based on a knowledge graph, and the method comprises the steps: extracting a structured degree feature, a semantic ambiguity feature and a context association feature through receiving and analyzing a natural language query statement inputted by a user, generating a retrieval intention vector, and carrying out the retrieval of the retrieval intention vector; and dynamically selecting a retrieval path according to the intention classification model. And according to the retrieval path, constructing a structured query statement or a semantic vector, and respectively retrieving in the knowledge graph and the vector database to obtain a first retrieval result and a second retrieval result. Further performing bidirectional verification through entity consistency, semantic similarity and relation connectivity indexes, screening a candidate result set, and constructing a reasoning chain; if the inference chain is broken, a large model inference gap complementation mechanism is adopted to generate relay nodes, a complete inference chain is formed, and inference type answer output is generated based on the complete chain. According to the method, the retrieval accuracy and reasoning continuity of the question-answering system are improved.

An AI-based Cybersecurity System and Method enable real-time detection, analysis, and mitigation of cyber threats within computing networks using adaptive artificial intelligence. The system continuously monitors network traffic, extracts behavioral and contextual attributes, and applies deep learning-based inference to identify anomalous activities indicating security breaches. The method integrates several computational units, including a network monitoring unit, feature extraction unit, artificial intelligence processor, contextual reasoning processor, and decision synthesis unit, to compute a composite risk index quantifying threat likelihood and severity. A classification processor categorizes detected threats into types such as ransomware, phishing, or unauthorized access, while a mitigation control processor initiates automated response actions to isolate compromised nodes and restore network integrity. An adaptive learning processor updates AI models using feedback from confirmed incidents. This provides a scalable, self-evolving cybersecurity framework that minimizes human intervention and enhances resilience against dynamic and zero-day threats.

The invention discloses a knowledge base system construction method and device for multi-source heterogeneous data and a medium, and relates to the technical field of artificial intelligence and natural language processing. The method comprises the following steps: integrating a dynamic graph convolutional network and a hierarchical attention mechanism to construct a multi-modal document analysis engine; performing semantic structure analysis on the original heterogeneous document on the basis of a multi-modal document analysis engine to extract document structure features and content semantic features, and constructing an original document relationship model on the basis of the document structure features and the content semantic features; based on the original document relationship model, performing classification fusion on heterogeneous data in the original heterogeneous document to obtain a to-be-stored heterogeneous data corpus, and processing the to-be-stored heterogeneous data corpus by using a graph neural network to establish a cross-modal semantic association index; and based on the cross-modal semantic association index, performing classified storage on the to-be-stored heterogeneous data corpora by utilizing a preset heterogeneous database so as to complete knowledge base system construction of the multi-source heterogeneous data.

The invention discloses a multi-round knowledge-guided question-answering method and system fusing a large language model and a knowledge graph. The method comprises the following steps: S1, constructing the knowledge graph; s2, performing retrieval and clarification; s3, extracting labels and classifying questions; s4, performing scene guidance and dynamic retrieval; and S5, answer generation and formatting output. The method aims at solving the fuzzy problem in complex policy and regulation questions and answers, and the accuracy of user demand analysis and the comprehensiveness of answers are improved. According to the method, a knowledge graph is taken as a core, and multi-path semantic analysis and problem guidance capabilities are provided by constructing a hierarchical structure covering first-level items, second-level items, scene categories and keyword nodes. In combination with strong semantic comprehension and generation capabilities of the pre-trained large language model, the system can dynamically track context information in user input, iteratively access related nodes along different paths of the knowledge graph, and gradually clarify user intentions.

In a described embodiment, a system for information processing is provided including a data acquisition module configured to receive feedback corresponding to one or more outputs generated by a language model. The system further includes a cognitive reasoning module configured to evaluate the reasoning process of the language model, emulate cognitive functions including metacognitive processes, and generate an assessment based on an analysis of the received feedback, wherein the assessment includes classifying the one or more outputs into components, assigning quality scores for each component, and identifying an improvement corresponding to the one or more outputs. Additionally, the system includes a process adjustment module coupled to the cognitive reasoning module for adjusting the reasoning process of the language model based on the assessment is provided. A refinement module coupled to the process adjustment module is provided for iteratively refining the reasoning process based on subsequent updates to the generated assessment until a performance threshold is met.

The invention relates to the technical field of artificial intelligence and intelligent education, and provides a mathematical teaching knowledge graph generation method and system based on artificial intelligence, which are used for optimizing and improving a teaching knowledge graph generation technology to realize more intelligent and accurate teaching knowledge graph generation, and the method comprises the following steps: obtaining a multi-source teaching data set; extracting a mathematical knowledge point entity set from the textbook text data, and generating an association relationship set among mathematical knowledge point entities in the mathematical knowledge point entity set according to the test question structure data; performing hierarchical classification processing on the mathematical knowledge point entity set based on a preset semantic analysis model to obtain knowledge point hierarchical structure data, and calculating weight distribution data of the mathematical knowledge point entities based on the association relationship set; and generating a dynamic knowledge graph topological structure according to the knowledge point hierarchical structure data and the weight distribution data, wherein nodes in the dynamic knowledge graph topological structure comprise semantic vectors and association strength parameters of mathematical knowledge point entities.

The invention discloses an intention classification method and device based on vector retrieval and context awareness and a medium, and relates to the technical field of artificial intelligence. The method comprises the steps of extracting business metadata and associating the business metadata with typical problem examples to generate a standardized service description document; encoding the standardized service description document into a high-dimensional semantic vector through a pre-training language model, and constructing a neighbor search index to store the high-dimensional semantic vector; splicing the user identity information and the current question text into an enhanced query statement, and encoding the enhanced query statement into a context-aware dynamic query vector through a semantic model; performing similarity retrieval based on the dynamic query vector to obtain candidate intelligent services, performing business domain filtering, context weighted sorting and dynamic priority rearrangement, and outputting target recommendation services; by collecting interactive behavior data of a target recommendation service, quality scoring is performed on service descriptions and problem examples based on a preset evaluation rule, and the service descriptions and the problem examples of which the quality scores are lower than a quality threshold value are updated.

The invention provides a knowledge and data fused medical content image-text generation system and method. Relates to the technical field of digital medical treatment and artificial intelligence. The multi-modal feature extraction module is used for extracting multi-modal features from the multi-modal data; the multi-modal feature fusion module is used for carrying out cross-modal dynamic fusion on the multi-modal features to generate cross-modal high-consistency fusion features; the model training optimization module is used for configuring a soft and hard target loss function and a parameter optimizer; the information analysis and reasoning module is used for carrying out semantic matching on the fusion features, the disease classification and the clinical path based on a dynamic medical knowledge network, and generating disease prediction probability distribution and a semantic reasoning path; and the image-text report generation module is used for generating an image-text report containing a diagnosis conclusion and an interpretation basis according to the prediction probability and the semantic reasoning path. According to the invention, intelligent and precise generation of medical contents can be realized.

The invention discloses a multi-modal visual arrangement recommendation method, belongs to the technical field of artificial intelligence and data visualization crossing, and realizes visual arrangement recommendation based on multi-modal input analysis, a dynamic mixed recommendation model and an intelligent optimization algorithm. Comprising the following steps: multi-modal intention analysis: realizing intelligent analysis of multi-modal input through combined use of a base model and a fine tuning model, realizing high-precision intention classification in combination with a pre-training language model and a domain adaptation fine tuning technology, and triggering dynamic prompt word recommendation; performing intelligent layout generation: performing global optimization of component space allocation by adopting a genetic algorithm, performing business rule adaptation by combining a constraint solver, and modeling an interaction relationship between components by utilizing a graph neural network; and dynamic mixed recommendation: constructing a three-level recommendation architecture including collaborative filtering, content matching and reinforcement learning. According to the method, a closed-loop recommendation process of user intention-intelligent recommendation-feedback optimization is realized, and the intelligent level of visual arrangement and the user experience are remarkably improved.

The invention discloses a teaching data management method and system based on artificial intelligence, and the method comprises the steps: obtaining a standardized time series data stream according to a heterogeneous data stream generated by a multi-source teaching platform in real time; based on the standardized time sequence data stream, performing classified encryption on the teaching data through a dynamic hierarchical storage strategy based on attribute-based encryption to obtain a security hierarchical storage topological structure; according to a user query request and a teaching scene label, extracting a target data set from the security hierarchical storage topological structure to obtain an enhanced multi-modal teaching data set; based on the enhanced multi-modal teaching data set, generating an interpretable teaching mode graph through a dynamic sub-graph evolution algorithm; and according to the teaching mode map and the real-time teaching feedback data, generating a personalized teaching recommendation strategy through a course-learner dual-channel adaptive recommendation model. According to the embodiment of the invention, the utilization efficiency of teaching resources can be improved, and personalized and intelligent teaching recommendation and decision can be realized.

A method for structuring the latent space of an autoencoder is provided. The method includes analyzing natural language descriptions related to input data; creating language-guided libraries that categorize and abstract data features based on the analyzed descriptions; mapping input data into the categorized and abstracted features within the latent space of the autoencoder; and training the autoencoder to minimize reconstruction loss while adhering to the structure imposed by the language-guided libraries.

The invention provides a large model cue word design method, system and device in an industrial scene and a medium, and belongs to the technical field of artificial intelligence. The method comprises the following steps: constructing a private cue word template library, and forming a domain special knowledge unit set through multi-source industrial data collection, intelligent semantic analysis and classified storage; customizing a four-layer structured template, sequentially establishing a task overview layer, a step disassembly layer, an instruction refining layer and an output specification layer, and constructing a layered mapping model from task definition to output execution; template library management and iterative optimization are carried out, and dynamic updating and performance improvement of a cue word system are realized through centralized system management, multi-source feedback acquisition and data-driven optimization. According to the method, a cue word design and optimization system based on a hierarchical structure is constructed for complex task requirements in an industrial scene, and the task processing performance of a large model in scenes of industrial production, quality detection, equipment operation and maintenance and the like is improved.

The invention discloses a reverse question guide question answering implementation method and system, and belongs to the technical field of artificial intelligence and natural language processing. Context-aware intention dynamic correction is realized through a three-level intention classification system, and cross-modal knowledge matching is realized by adopting a distributed semantic index technology; based on the reinforcement learning strategy, optimizing a cooperative work mechanism of the dialogue strategy and the knowledge base; comprising the steps of intention recognition: analyzing a session of a user by using an intention recognition model, and constructing a three-level intention classification system based on deep semantic understanding, including main class recognition, fine-grained analysis and context perception; question rewriting: constructing a dynamic rewriting engine to rewrite the user question; recalling and cleaning multi-source item knowledge; generating a reverse question; locking items and acquiring item data; generating questions and answers. According to the method, the robustness, the real-time performance and the scene adaptation capability of a professional question answering system can be improved, and the government affair service question answering accuracy, the intention recognition precision and the cross-region recommendation adoption rate are improved.

The invention provides an APT (Advanced Persistent Threat) attack detection method based on a large language model, which comprises the following steps of: S1, extracting original system call event data from a kernel audit log of an operating system, and preprocessing the data; s2, constructing a multi-model collaborative detection architecture based on a large language model, and realizing fine-grained classification of network entities according to the preprocessed data through prompt construction, model fine tuning and a confidence scoring mechanism; s3, constructing an adaptive graph search algorithm based on multi-modal feature correlation modeling, driving attack path topology reconstruction, and realizing maximum reduction of a malicious sub-graph topology structure; s4, carrying out combination with MITRE ATTamp; the CK tactical knowledge base constructs a cyclic enhancement analysis framework, a cyclic enhancement technology is adopted to drive a large language model to execute hierarchical association reasoning, a mapping relation from malicious subgraphs to attack tactics and tactical chains is derived step by step, and finally an attack report summary and a targeted defense strategy are generated. According to the invention, APT attack detection with high accuracy and high interpretability is realized.

The invention discloses an article recognition system based on computer vision. The article recognition system comprises a multi-modal data acquisition module, a multi-modal data processing module and a computer vision processing module, wherein the multi-modal data acquisition module is used for acquiring multi-modal data through a multi-modal sensor array; the data preprocessing module is used for standardizing a multi-modal data format and generating a time-space aligned multi-modal tensor; the feature extraction module is used for respectively extracting modal specific features from texture, spectrum and geometric dimensions by adopting ResNet50, 3D-CNN and PointNet + +; the multi-modal fusion module is used for constructing cross-modal joint representation; the adaptive sensing module is used for modeling illumination invariance and scene dynamics based on self-supervised comparative learning and a 3D-STMN space-time memory network, predicting a shielded target trajectory by using Kalman filtering in combination with the shielding sensing propagation module, and generating an environment sensing parameter set; and the recognition engine module is used for integrating YOLOv8 detection, Mask R-CNN segmentation and multi-modal decision tree classification, outputting a target bounding box, a category and confidence in combination with the depth data, and generating three-dimensional space coordinates combined with the depth data.

The invention relates to a supporting structure stress state monitoring method based on artificial intelligence, and belongs to the technical field of artificial intelligence and data processing. The method comprises the following steps: acquiring and marking strain data of a supporting structure; after abnormal values are removed, normalizing the multi-sensor data to generate a normalized strain sequence; a state monitoring model is constructed, a deep time sequence neural network architecture is adopted, and the state monitoring model comprises an input layer, a self-adaptive wavelet attention feature mapping layer, a time domain gating convolution module, a global maximum pooling layer, a dynamic feature importance reweighting layer and a full-connection classification layer; inputting a normalized data training model; optimizing a loss function through a quantile interval adaptive learning rate and a momentum updating strategy; after real-time monitoring data is processed, inputting the data into the training model according to time window slices, outputting four types of probabilities, and taking the maximum value as a prediction state; and if a plurality of continuous windows are early-warning and dangerous, triggering the terminal to give an alarm. The accuracy of monitoring the stress state of the supporting structure can be improved.

The invention provides an intelligent customer risk assessment system and method based on a large language model, and relates to the technical field of risk assessment, and the method comprises the steps: obtaining multi-modal data of a customer, carrying out the preprocessing, and extracting structured and unstructured features; constructing a hierarchical risk knowledge system, and realizing adaptive evolution of the knowledge system through a graph neural network and a generative model; constructing an initial negative sample library, and constructing a negative sample database in combination with a non-risk mode labeled by an expert and derivative layer analysis; optimizing the large language model by adopting a strong supervision, weak supervision and reinforcement learning cooperative training mechanism under each classification according to the customer type; mining risk features in a text by using the optimized large language model, processing multi-modal data through a multi-level attention network, and generating a positioning report including contradiction type coding, service influence dimension evaluation and risk level quantification; the accuracy, efficiency and flexibility of customer risk assessment are improved, and the risk management strategy is optimized.

The invention provides a retrieval enhancement generation method and system based on a large language model. The method comprises the steps that input query information of a user is obtained; performing rewriting and key information extraction on the input query information based on a large language model to obtain a first query instruction and a second query instruction; according to the input query information, identifying a user intention based on a large language model and performing intention classification to obtain an intention classification result; according to the first query instruction and the second query instruction, performing multi-path mixed retrieval through a preset multi-source knowledge base to obtain an initial retrieval result; according to the intention classification result, performing result rearrangement and screening on the initial retrieval result to obtain an enhanced cue word; and inputting the input query information and the enhanced cue word into the large language model to obtain a retrieval enhanced generation result. According to the method, the query demand and intention of the user can be accurately recognized, related information is retrieved in an open domain in a high-precision manner, and the instantaneity, the accuracy, the specialty and the reply quality of information retrieval are improved.

The invention discloses a soft contrast learning fault diagnosis method based on a big language model question and answer dialogue, and the method comprises the steps: inputting the structure parameters and working condition information of a target rotating machine part into a thinking chain big language model, and obtaining a strong and weak data enhancement strategy and parameters matched with the part; performing instance-level and time-level hierarchical measurement on the vibration samples without labels by using a hierarchical soft label distribution mechanism, and respectively constructing an instance-level soft label matrix and a time-level soft label matrix; inputting the strongly enhanced samples, the weakly enhanced samples and the soft labels thereof into a soft contrast learning module, and pre-training a feature encoder through hierarchical soft label feature refinement, cross prediction and context consistency tasks; adding a fault diagnosis classification head module to the pre-trained feature encoder, and performing fine tuning by using a small number of vibration samples with labels to obtain a fault diagnosis model; and performing fault diagnosis through the fault diagnosis model. According to the invention, fault diagnosis of automatic data enhancement and hierarchical soft contrast learning can be realized.

The invention discloses an intelligent approval rule modeling method oriented to process automation, and relates to the technical field of business process management, and the method comprises the following steps: S100, in a process of constructing a rule candidate set, extracting scene features, field semantic hierarchy and participation role information of each piece of historical approval data, generating a context semantic tag set, and establishing a rule candidate set; the method is used for subsequent rule difference modeling. According to the method, context semantic tags are introduced to be aligned with ternary features, so that the semantic boundary recognition capability of the rule is enhanced; constructing a rule feature matrix and a differentiation candidate set, and realizing accurate classification and processing of ambiguity rules; in combination with expression sensitivity enhancement and simulation verification, approval offset and risk are identified in advance; finally, the dynamic optimization of the rule model is realized through backtracking correction, the stability and accuracy of the rule model in multiple scenes are improved, and a closed-loop credible intelligent approval rule system is constructed.

The invention discloses a multi-label electrocardiogram classification method based on self-supervised pre-training and multi-modal semantic alignment, which belongs to the technical field of artificial intelligence, and comprises the following steps: realizing self-supervised pre-training of unlabeled data through a single-modal contrast enhancement network, generating global and local contrast views by adopting a multi-scale random cutting strategy, and classifying the global and local contrast views in a multi-scale random cutting mode; in combination with a teacher-student network architecture, the potential invariance features of the ECG signals are learned while negative sample dependence is avoided, the problem of annotation data scarcity is effectively relieved, and the feature robustness is improved. A multi-modal fusion mechanism based on label semantic guidance is provided, a time domain signal and a frequency domain time-frequency graph are mapped to a unified semantic space through fine-grained semantic alignment, local feature enhancement and cross-modal complementary information fusion are realized by using a cross attention mechanism, and the problem of semantic difference caused by modal heterogeneity in a traditional method is overcome. A multi-label comparison loss function based on a disease co-occurrence relation is proposed, a category discrimination boundary is dynamically optimized by modeling a label co-occurrence probability, the feature separability of a tail category is improved while the head category discrimination ability is enhanced, and the problem of sample category imbalance in a multi-label scene is remarkably relieved.

The disclosure relates to systems and methods of identifying concepts in content having natural language text using a Large Language Model (LLM), training neural networks in a discovery phase to classify the identified concepts into categories, sub-categories, or other groupings of concepts, and executing the neural networks in an operational phase to classify identified concepts.

Systems and methods for detecting and mitigating prompt injection attacks on a generative LLM are disclosed. A deployment scenario is considered, in which the generative LLM supports a task automation function. Prompts are received and interpreted by the generative LLM, and outputs from the generative LLM are used to trigger automation actions. The prompts are constructed based on a combination of user input and external data and are, therefore, vulnerable to prompt injection attacks though manipulation of the external data. To mitigate this risk, a separate discriminative classification, decoupled from the generative LLM, engine is configured to identify malicious prompts, and filter out any malicious prompts before they reach the generative LLM.

Aiming at the problems of insufficient language understanding depth, weak personalized dialogue generation ability, lack of continuous learning and long-term user state modeling and the like in the current emotion recognition and psychological intervention technology, the invention provides a user emotion recognition and psychological intervention method combined with a large language model (LLM). According to the method, the potential emotional state is identified by analyzing free text information input by a user by utilizing the powerful capabilities of a large language model in the aspects of natural language understanding, emotional modeling and text generation; constructing a multi-round dialogue context, and reasoning a psychological change trend of the user; in combination with a psychological knowledge base, personalized and mild psychological intervention dialogue content with a dredging effect is generated. The system supports recognition and classification of various emotional states such as depression, anxiety and alonity, is suitable for various interaction scenes (such as APPs, webpages and social robots), and can greatly improve the precision of emotion recognition and the timeliness and effectiveness of psychological intervention. The emotion recognition and psychological intervention method based on the large language model provides solid technical support for constructing an intelligent, continuous and personalized psychological health management system, and has wide application prospects and profound social significance.

The invention relates to the technical field of data classification management, in particular to a knowledge graph-based foreign trade data classification management system, which comprises a term word order modeling module, a graph path generation module, a path cross analysis module, a semantic category judgment module and a classification structure output module. According to the method, morpheme-level word segmentation and word order extraction are performed on foreign trade terms, an original word order template is constructed, the reduction capability of a semantic expression sequence is enhanced, a directed path is established through morpheme-level label sorting, and the level definition of a term structure and the logic relation between nodes are enhanced; a semantic intersection area is extracted through node intersection and end point frequency analysis, the accuracy of term association judgment is improved, semantic affiliation is selected according to end point label frequency, a node classification closed loop is formed by combining labels and path mapping, links of term graph modeling, path construction, intersection analysis and affiliation output are broken through, and the term association judgment accuracy is improved. And the classification management process of foreign trade data is fully improved.

The invention discloses a multi-modal fusion rumor detection method and system based on a dynamic graph convolutional neural network. According to the method, a dynamic feature graph of a language propagation path is constructed, and potential features in the language propagation process are extracted and analyzed by utilizing time sequence changes and key node relations between nodes in a propagation graph. A neural network is adopted to extract and enhance image data, text semantic features are extracted in combination with a text feature modeling network, text feature vectorization expression is achieved based on a BERT model, and rich semantic information is obtained. And a gating mechanism is introduced to dynamically adjust fusion weights of different modal features, and an information fusion strategy is optimized. A collaborative attention mechanism is further adopted for deep fusion, interactive learning of text, image and propagation path features is enhanced, and the relevance of cross-modal and time series data is improved. And finally, inputting the fused feature vectors into a classifier for accurate classification, thereby realizing accurate detection of the social media rumors. According to the method, the multi-modal features are effectively integrated, and the false information identification efficiency is remarkably improved.

The invention discloses a knowledge base question answering method and system based on intention recognition and a medium, and the method comprises the steps: receiving a user question, inputting the user question into a large language model subjected to intention recognition training, and carrying out the preliminary classification; determining a corresponding search strategy according to a classification result obtained by the preliminary classification; if the search strategy is knowledge base search questions and answers, selecting a knowledge base matched with the search strategy for knowledge retrieval, inputting the retrieved reference knowledge and cue word templates into a large language model, and answering the user questions by the large language model; and if the search strategy is large-model direct question answering, selecting a cue word template, inputting the cue word template into the large-language model, and answering the user question by the large-language model. According to the method, user questions can be preliminarily classified through user intention recognition, different search strategies are adopted for different categories, and knowledge base question answering efficiency and accuracy are effectively improved.

The invention discloses an automatic label labeling and classifying method and system oriented to unstructured system documents, and relates to the technical field of artificial intelligence. The method comprises the steps that semantic structure pre-analysis is conducted on an original system text, and a system semantic structure tree is constructed; establishing a system semantic enhancement vector space based on the semantic units and the logic relationship thereof; performing semantic deconstruction on the preset tag and extracting a feature vector; realizing cross-space semantic matching of the document and the tag through a system semantic attention mechanism; a confidence evaluation module is introduced to screen high-confidence labels from the three dimensions of structural integrity, coverage and logic consistency; and outputting a final label and a score through semantic conflict detection and resolution. According to the method, the problems that in the prior art, unstructured system text labeling accuracy is low and large-scale labeling samples are dependent on polysemy ambiguity, high context dependency, complex semantic structure and the like are solved, and labeling accuracy and robustness are remarkably improved.

The invention relates to the technical field of artificial intelligence and data processing, in particular to a roadbed settlement data identification method based on artificial intelligence, and the method specifically comprises the following steps: collecting time sequence monitoring data of a plurality of roadbed settlement sensors, and carrying out the manual marking; carrying out adaptive time domain segmentation normalization on the collected time sequence monitoring data, designing a conditional diffusion process, and generating enhanced data conforming to a soil mass mechanics constraint in a submerged space; constructing a roadbed settlement data classification model based on a one-dimensional convolutional neural network, and inputting the enhanced data into the model for training to obtain a trained roadbed settlement data classification model; and preprocessing new time sequence monitoring data collected by the roadbed settlement sensor, and inputting the preprocessed data into the roadbed settlement data classification model to obtain a settlement category classification result. According to the invention, the roadbed settlement data classification model based on the one-dimensional convolutional neural network is constructed and trained, so that the discrimination capability of the model can be enhanced, and the accuracy and reliability of model identification can be improved.

Systems and methods for a prompt generation and analysis service for generating and identifying a preferred prompt for performing a function of a large language model (LLM) are provided. The prompt generation and analysis service may generate a set of training prompts for performing a function of an LLM. The prompt generation and analysis service may then query the LLM with the generated set of prompts and characterize the output of the LLM for each prompt. Using the characterization of the output and corresponding prompt, the prompt generation and analysis service can train a classifier model to classify the prompts. The prompt generation and analysis service may generate a set of target prompts for performing a function of an LLM, characterize the target prompts using the training classifier model, and identify a preferred prompt for performing the function based on the classifier model's classification.