67 results about "Context knowledge" patented technology

The invention discloses a customer service data quality inspection method and device based on dynamic reasoning, equipment and a medium, and the method comprises the steps: analyzing customer service dialogue data, and generating structured dialogue data containing an intention label and a key problem node; according to an intention label in the structured dialogue data, a framework is generated through retrieval enhancement, and matched domain knowledge is recalled in real time from a standard knowledge base so as to construct a dynamic context knowledge graph; taking the dynamic context knowledge graph as input, utilizing a thinking chain prompt template to guide a large model to carry out step-by-step logical reasoning, and outputting a preliminary quality inspection conclusion chain; performing retrieval verification on assertion nodes in the initial quality inspection conclusion chain to generate a traceable final quality inspection conclusion; and based on the final quality inspection conclusion, calculating a quantitative score of the customer service dialogue data and generating a visual thinking chain report. And through dynamic retrieval of the knowledge base and logical reasoning, the accuracy, interpretability and adaptability of customer service data quality inspection are improved.

The invention provides a risk conduction prediction method and system based on combined deduction of a time sequence diagram and a large model, and relates to the technical field of artificial intelligence and knowledge engineering, and the method comprises the steps: extracting a prospective time sequence fact based on a hierarchical cue word and a structured constraint mechanism; extracting conflicts and generating a sequential relationship; the historical time sequence knowledge graph is updated based on the time sequence relation and a confidence coefficient weighted updating strategy; determining a central entity, and initiating a structured query to the updated historical time sequence knowledge graph by taking the central entity as a starting point to generate a context knowledge sub-graph; converting the context knowledge sub-graph into a natural language description with a logic relationship, injecting a risk hypothesis event, constructing a cue word as a new input of a large model, and outputting to obtain a structured JSON object containing a complete reasoning chain; and converting the structured JSON object based on a risk path extraction and visualization algorithm influencing weight attenuation to obtain risk early warning information and a visual conduction path diagram.

The invention provides a construction method and device of an instruction generation agent based on a knowledge graph, a medium and equipment, and the method comprises the steps: obtaining a text, extracting a triple in the text, connecting the triple data to a graph database, and constructing a text knowledge graph; performing statistical analysis on the text knowledge graph from the aspects of entities, relationships and graph structures based on the text knowledge graph; based on the statistical analysis result, an instruction generation agent is constructed, an agent planning module, an agent tool module, an agent storage module and an agent execution module are constructed, and the agent execution module executes instruction generation operation based on the reasoning planning path and the cue word template of the corresponding category. And evaluating the quality of the instruction pair and executing an instruction rewriting operation on the data which is unqualified in evaluation. According to the method, the triple chain is constructed as the context knowledge point by utilizing the structural relationship of the knowledge graph, so that the illusion generated by the intelligent agent is reduced, and the quality of instruction data is ensured.

The invention discloses a global knowledge extraction method and system based on a large model and an RAG technology, and belongs to the technical field of large model data processing. According to the global knowledge extraction method and system based on the large model and the RAG technology, local sensitive hashing is adopted for duplicate removal, similar texts are combined through dynamic thresholds, and redundant data interference is reduced; a text segmentation strategy based on separator confidence ensures that segmented text blocks maintain semantic integrity and are adaptive to large model input length limitation, similarity retrieval and full-text retrieval are fused, results are reordered in combination with an RRF algorithm, semantic relevance and accuracy are considered, the multi-scene knowledge construction requirement is met, and the multi-scene knowledge construction efficiency is improved. High-correlation entity pairs are screened through mutual information, context knowledge is retrieved from a vector database in combination with an RAG system, semantic reasoning is conducted through a large language model, hidden logic relations among entities are generated, and jumping from data association to knowledge generation is achieved.

The invention discloses a multi-agent-based context knowledge internalization, sharing and multiplexing system and method. The system comprises a multi-source knowledge capture module, a situation information extraction and structuring module, a knowledge internalization and incremental learning module, a situation awareness knowledge reuse and recommendation module and a knowledge quality evaluation and maintenance module, wherein each module is composed of a special intelligent agent or a functional unit and works cooperatively; according to the method, interaction, cooperation and process data and expert input of a task execution full cycle of a multi-agent system are captured through a multi-source module, preprocessing is carried out through a situation extraction module, a situation is associated to generate structured knowledge, and then the structured knowledge is fused into a dynamic knowledge system through an internalization module to achieve incremental learning. In a new task, matching knowledge is retrieved and recommended through the context awareness module, and finally the knowledge quality is periodically maintained by the evaluation module to ensure accuracy and timeliness; according to the method, the autonomous learning ability, the knowledge reuse efficiency and the intelligent level of solving complex problems of the multi-agent system are remarkably improved.

The invention provides a hybrid AI adaptive optimization system based on agent context engineering. The hybrid AI adaptive optimization system comprises a context knowledge base, a generation module, a feedback analysis module and a knowledge integration module. According to the method, the problems that in an existing context optimization technology based on a large language model (LLM), context Bias is simplified, Context Collapse is caused by overall rewriting, and the learning process is high in delay, high in cost, unstable and the like are solved.

The invention relates to the technical field of artificial intelligence and emotion calculation, and discloses an emotion state evaluation method based on multi-modal data, and the method comprises the steps: obtaining and decoupling an implicit signal, an explicit signal and situation data of an evaluated object; and respectively generating implicit features, explicit features and context vectors through double-flow decoupling coding and context knowledge graph coding. And the core engine determines the internal emotional state and the extrinsic emotional representation in parallel, and dynamically regulates and controls a bridging conversion process from the internal state to the predicted extrinsic representation by utilizing the situation vector. Finally, the emotion regulation intensity is calculated by comparing actual and predicted extrinsic representations, and the emotion regulation intensity, the internal state and the extrinsic representations are jointly used as a multi-dimensional evaluation result to be output. According to the method, the dynamic relationship and the adjustment strategy between the internal feeling and the external expression can be disclosed, a deeper and more three-dimensional analysis view angle is provided for the fields of mental health, human-computer interaction and the like, and the method has remarkable application value.

The invention provides an attack link restoration method and system based on a large language model and a traditional AI model, and effectively solves the problem that attack link restoration is incomplete and inaccurate when a current attack link restoration scheme faces complex factors such as network address translation. The method comprises the following steps: acquiring an attack semantic knowledge base containing an attack context knowledge matrix, wherein the knowledge base is generated by processing an attack framework, a security log and a traceability report by a generative large model; obtaining an asset access relation graph generated by analyzing the metadata of the target network infrastructure by the large model and reasoning in combination with an attack semantic knowledge base; regularly acquiring security alarm logs in a first time window and scoring, and acquiring context logs if the security alarm logs exceed a threshold value; based on a time sequence diagram attention network model, determining the confidence degree that edges in an asset access relation graph corresponding to logs containing access pairs in the alarm logs and the context logs belong to attack links; and performing sub-graph extraction in the atlas to obtain candidate sub-graphs, and then determining a target attack link.

The invention discloses a position and knowledge similarity-based software crash error repair scheme recommendation method, which outputs a repair method for error codes according to generated crash stack tracking information and corresponding codes when software crashes, and can help developers to quickly and efficiently solve the problem of software crash. When the method works, firstly, a context knowledge base is constructed, problem posts containing crash stacks on Stack Overflow are preprocessed, a knowledge base and an abnormal dictionary are constructed, and a language model is trained; thirdly, a related question set is preliminarily obtained through stack tracking, the comprehensive similarity of the question and an input stack is calculated through the position related similarity and the additional knowledge similarity, and the most related question and answer set is returned; and finally, performing relevancy calculation on the answer paragraphs based on multiple features, selecting paragraphs related to the crash reason by applying an MMR algorithm, and outputting root cause analysis and repair code snippets related to the crash error.

The invention provides a game real-time intelligent translation method and device based on interactive reinforcement learning. The method comprises the following steps: determining a corresponding screen capture mode according to current game picture information and user operation habit information; dynamically sensing the current game picture according to the screen capture mode; performing text recognition on the image data in the user attention area by using a context-enhanced image recognition model, and obtaining a recognition text according to a context knowledge base exclusive to the game scene and user operation historical information; inputting the recognition text, the corresponding image data and user operation historical information into a multi-modal fusion translation model to generate a target language text; and based on the corresponding relationship between the recognition text and the target language text, rendering the target language text on the original foreign language region in an augmented reality mode. According to the method, real-time, automatic and immersive translation of the game picture content is realized, a large number of continuous texts can be processed, and the translation result better conforms to a game scene.

The invention relates to the field of knowledge enhancement, natural language processing and artificial intelligence, and provides a dialogue emotion recognition model determination method and device and a dialogue emotion recognition method and device. The dialogue emotion recognition model determination method comprises the steps of determining target contextual knowledge corresponding to a dialogue sequence sample based on the dialogue sequence sample and a contextual knowledge prompt template; inputting the dialogue sequence sample and a contextual knowledge prompt template into an initial model to obtain first predicted contextual knowledge corresponding to the dialogue sequence sample; and based on the target context knowledge and the first prediction context knowledge, performing fine tuning on the initial model to obtain a trained dialogue emotion recognition model. The dialogue emotion recognition model determination method provided by the invention can improve the recognition effect of the dialogue emotion recognition model.

The invention belongs to the technical field of natural language processing and power system intelligent information processing. Comprising the following steps: S1, preparing data; s2, performing named entity recognition and relation extraction on the text block by utilizing a pre-trained large language model; s3, constructing a knowledge graph in the urban power distribution field by using the triple summary; s4, generating a semantic abstract with a hierarchical structure for each theme community; s5, querying and retrieving: retrieving related information from the vector index and the knowledge graph in parallel by receiving a natural language query of a user, and fusing results into a context knowledge set; and S6, an answer generation step: inputting the query and the context knowledge set thereof into a large language model, and generating an answer which is coherent in semantics, accurate in facts and covers query requirements through a preset prompt strategy guide model. According to the method, the accuracy and practicability of the question-answering system are remarkably improved, and core technical support is provided for intelligent operation and maintenance of the power distribution network.

The invention discloses an entity attribute prediction method based on dynamic twinborn attention and a KAN network. The method comprises the following steps of entity digital twinborn construction and data mapping; constructing a space-time knowledge graph; carrying out dynamic feature fusion based on DTA-KAN; predicting a future state based on fusion features; according to the method, firstly, a digital twin and a space-time knowledge graph are constructed for a target entity, an attention query signal is dynamically generated by using a real-time state vector of the digital twin, and the signal is used for guiding a graph attention network; according to the method, context knowledge most relevant to the current real-time state is adaptively extracted from a space-time knowledge graph, intelligent prediction of dynamic knowledge extraction and deep fusion under the guidance of the real-time state is realized, and a complete closed loop from original data acquisition to fusion and prediction to comprehensive presentation and intelligent research and judgment is formed; and powerful technical support is provided for realizing prospective and initiative social governance.

The invention discloses a teaching content generation method and system based on large model instruction and context knowledge dual drive. The method comprises the following steps: firstly, analyzing, slicing and vectorizing an original knowledge document in the education field through knowledge preprocessing in the education field; then, defining teaching content to generate a prompt word template, and defining a large model instruction, a variable and a context knowledge content block; secondly, arranging teaching content to generate a workflow, and constructing large model instruction driving and context knowledge driving logic; and finally, implementing dual-drive teaching content generation and performing result post-processing. Based on the method, a system corresponding to the method is constructed, and the system comprises an education field knowledge preprocessing module, a teaching content generation prompt word template definition module, a teaching content generation workflow arrangement module, a teaching content generation implementation and result post-processing module and a user man-machine collaborative interaction operation module. According to the method, the teaching content can be generated according with the education rule, the specific teaching scene requirement is met, and the teaching content generation accuracy and applicability are improved.

The invention discloses a context-aware dialogue system and method based on AI assistant intelligent question answering, and belongs to the technical field of artificial intelligence and man-machine interaction, and the system comprises a user interaction interface module which is used for receiving a query request input by a user and returning a response result to the user; the dialogue context management module is used for capturing, storing and structurally processing historical information in the dialogue session in real time so as to construct and dynamically update a dialogue context knowledge base containing original dialogue records, refined dialogue states, user intentions and entity information; and the context perception understanding module is used for carrying out deep semantic analysis on the current query of the user and carrying out joint understanding in combination with the dialogue context knowledge base so as to analyze anaphora, omission and implicit intentions in the context and generate a complete semantic representation rich in context information. According to the method, the AI assistant can have strong context memorizing, understanding and reasoning capabilities, and human-like and coherent multi-round intelligent dialogues are realized.

The invention discloses a financial field knowledge base construction method, device and equipment. The method comprises the following steps: firstly, identifying a text region, a table region and an image region from a target image; then extracting text data of the text region by using an OCR algorithm, and extracting semantic features and visual features of the image region and a second entity set and an entity type corresponding to the table region by using a multi-modal AI model; carrying out illusion detection by utilizing the similarity of the two entity sets, and when judging that the illusion of the AI model does not trigger a fault-tolerant mechanism by utilizing a detection result, carrying out intelligent alignment processing on the two entity sets, and screening out an optimal entity matching pair; carrying out formatting verification on the entity type; and then context knowledge association processing is carried out on text data corresponding to the text region, semantic features and visual features corresponding to the image region and entity matching pairs passing verification, and a more stable and accurate knowledge base is constructed by utilizing the obtained graph structure document to represent corresponding structured knowledge.

The application discloses a kind of intelligent query control method and electronic equipment based on natural language interaction.The method includes the following steps: building context knowledge base, converting into context vector, and storing in vector database;Convert the input natural language question into query vector;Retrieve the k context vectors most similar to the query vector from the vector database, and determine the corresponding context information;Combine the natural language question and the retrieved context information into a prompt word sequence input into a large language model (LLM) to obtain an SQL query statement, and analyze whether the natural language question contains a control intent;Execute the SQL query statement;If the natural language question contains a control intent, analyze the query result intelligently and perform the corresponding device control operation.The application can analyze the control intent in the natural language question, and after the result is queried, the corresponding device can be controlled according to the control intent to work, improving the work efficiency.

The invention discloses a knowledge retrieval method, device and equipment, and the method comprises the steps: carrying out the complexity evaluation of a query statement inputted by a user, and obtaining a complexity evaluation result; based on the complexity evaluation result, selecting a retrieval strategy corresponding to the complexity evaluation result to perform entity extraction processing on the query statement to obtain target entity information; based on a retrieval strategy, performing selection and combination processing on the target entity information and the query statement to generate a retrieval instruction; based on the retrieval instruction, performing retrieval from a knowledge vector library to obtain a plurality of knowledge retrieval results; and performing score selection processing on the plurality of knowledge retrieval results, and constructing to obtain target context knowledge data.

The invention discloses a knowledge graph-based government affair retrieval enhancement generation framework and method, and relates to the technical field of natural language processing. The method specifically comprises the steps that fine adjustment is conducted on a basic model through government affair corpora, and a government affair field enhanced semantic model is generated; vectorizing a government affair knowledge graph entity by using a semantic model enhanced in the government affair field and storing the vectorized government affair knowledge graph entity into a graph entity vector database; key government affair entities are extracted from user questions, semantic matching is carried out on the extracted government affair entities and graph entities, and accurate context knowledge is retrieved from the knowledge graph through automatically generated graph query statements; and fusing the retrieved context knowledge with the original question to construct an enhanced prompt, and driving a large language model to generate a high-quality natural language government affair answer. According to the method, the accuracy of government affair questions and answers, the understanding depth of complex government affair logic and the retrieval efficiency are remarkably improved by deeply fusing the structural advantages of the knowledge graph and the semantic ability of the large language model.

The invention relates to a database diagnosis method and device based on a large model, electronic equipment and a storage medium. The method comprises the steps that index data, an internal library table and field annotations in a database are obtained, and a knowledge system of the database is set; classifying the index data according to the knowledge system to obtain fault data corresponding to each fault system; performing natural language mapping processing on the fault data to obtain a large-model context knowledge system; training a preset fault large model according to the large model context knowledge system, the internal library table and the field annotation, and then performing fault detection to obtain a diagnosis result; fault data classification is achieved by building a knowledge system, a context knowledge system capable of being recognized by a large model is generated in combination with a natural language processing technology, automatic fault diagnosis is achieved based on the trained fault large model, and the technical problems that a traditional method depends on manual analysis and is low in complex fault processing efficiency are solved. The method has the advantages of improving diagnosis accuracy and reducing operation and maintenance cost.

A question answering method based on a large model is performed by an electronic device. The method includes: determining a first result corresponding to a query statement based on a first large model, in which the first result includes a first response corresponding to the query statement, a first response logic corresponding to the first response, and first context knowledge corresponding to the first response; determining a retrieval result corresponding to the query statement by retrieving in a database based on the query statement; determining a type of a knowledge conflict between the first result and the retrieval result; and determining a target response corresponding to the query statement based on the first result, the retrieval result, and the type of the knowledge conflict.

The invention provides a diversified multi-hop question generation method and system based on knowledge combination sampling, and relates to the technical field of multi-hop questions.The generation method comprises the steps that context knowledge and target answers are obtained, and the target answers are obtained based on a hierarchical neural network architecture; obtaining conditional probability distribution data and an initial knowledge combination according to the context knowledge and the target answer; obtaining a plurality of diversified knowledge combinations according to the initial knowledge combination and the conditional probability distribution data through a preset sampling strategy; and generating a corresponding multi-hop question according to all the diversified knowledge combinations and the target answer through a preset multi-hop question generation model. The whole process does not need complex intermediate links and can be realized by relying on a mature neural network architecture, the manpower and hardware cost of technology landing is reduced, for example, when a multi-hop question bank is constructed in the education field, a teacher only needs to input a textbook text and knowledge points, multi-hop questions of different examination angles can be automatically generated, and the efficiency is improved. And the problem of data sparsity of the multi-hop question-answer data set is effectively relieved.

The application provides a risk transmission prediction method and system based on a time sequence diagram and a large model joint deduction, relates to the technical field of artificial intelligence and knowledge engineering, and comprises the following steps: extracting forward-looking time sequence facts based on hierarchical prompt words and a structured constraint mechanism; extracting conflict events and generating time sequence relationships; updating a historical time sequence knowledge graph based on a time sequence relationship and a confidence weighted update strategy; determining a central entity, initiating a structured query to the updated historical time sequence knowledge graph from the central entity, and generating a context knowledge subgraph; converting the context knowledge subgraph into a natural language description with logical relationships, injecting a risk hypothetical event, constructing a prompt word as a new input of a large model, and outputting a structured JSON object containing a complete reasoning chain; and based on a risk path refining and visualizing algorithm with influence weight attenuation, converting the structured JSON object to obtain risk warning information and a visualized transmission path diagram.

The invention discloses a multi-source heterogeneous data continuous learning evaluation method based on context knowledge enhancement, and relates to a continuous learning technology and a model performance evaluation method in machine learning. The method mainly aims at evaluating the ability of a depth model to continuously learn sequence tasks on multi-source heterogeneous data; the method is an evaluation method which is based on context knowledge enhancement and focuses on key new knowledge learning efficiency and precision. According to the method, a human forgetting curve is simulated, the defect that new knowledge and old knowledge are not distinguished in an existing evaluation mode is broken through, higher calculation weights are given to knowledge needed by the model for processing problems in the current context environment, and the continuous learning ability of the model on multi-source heterogeneous data is evaluated in a mode more fitting the reality.

According to the multi-mode emotion regulation subtitle generation method fused with cognitive intervention, classroom teaching audio is collected through an omnidirectional condenser microphone, multi-dimensional acoustic features are extracted after preprocessing such as mute segment removal, noise suppression and signal normalization, the audio is converted into a text through a DNN-ASR model and a pre-training BERT model, and a semantic embedding vector is generated. An improved CNN-LSTM model is adopted to fuse acoustic and semantic features for emotion recognition, and a primary cognitive tag is generated in combination with a classroom situation knowledge base. Based on an emotion cognition theory, secondary modeling is carried out on emotion signals through an explanatory subtitle generation sub-model and a background annotation adjustment sub-model, an emotion adjustment factor matrix is formed, and weights are dynamically controlled. And finally, according to the adjustment factor, generating a multi-mode subtitle, matching a color code, a dynamic icon and a font style, and presenting the information on a teaching terminal. According to the invention, classroom teaching information processing efficiency can be effectively improved, and deaf students and other groups can be assisted to better understand teaching contents.

The invention relates to the technical field of data processing, and provides a data retrieval method and device, electronic equipment and a readable storage medium. The method comprises the following steps: performing semantic similarity matching processing on query data to obtain context knowledge data corresponding to the query data; performing integrity verification on the context knowledge data to obtain a state identifier, valid data and missing information representation data corresponding to the context knowledge data; under the condition that the state identifier is in an incomplete state, performing closed-loop completion processing on the query data based on the missing information representation data, performing re-query based on the completed query data, and obtaining updated valid data in combination with the valid data; and performing integration generation processing on the updated effective data and query data to obtain a first target retrieval result, thereby improving the information integrity, improving the accuracy of the retrieval result, enhancing the credibility of the retrieval result and improving the robustness.

A machine learning and inference system for generating personalized activity plans includes: a trained world context knowledge base; a trained personal digital memory bank; the target engine is used for establishing a moving target; an activity decomposition engine to decompose the activity into connected logically consistent sub-activities; a weight allocation engine to allocate value weights to the potential sub-activities; a graphics generation engine to generate a multi-layer weighted graph to train a model of personalized results of the sub-activities according to value weights; a scene generation engine to analyze a network of logically consistent potential sub-activities connected by paths to determine a selected scene path to the activity target according to a model that at least personalizes the user results; a feedback engine to apply learning from the scene generation engine to a world context knowledge base and / or a personal digital memory bank; and an output channel for outputting a personalized activity plan comprising recommended actions for implementing the selected scene path.

The present invention provides a device for optimizing emotional responses using a commonsense knowledge graph incorporating emoticons. The device comprises a dictionary graph storage module for storing an emotional lexicon and an emoticon-enhanced commonsense knowledge graph; an emotional context knowledge subgraph generation module for generating an emotional context knowledge subgraph based on the current conversation, the emotional lexicon, and the emoticon-enhanced commonsense knowledge graph; and a response generation module comprising a conditional variational autoencoder for inputting the emotional context knowledge subgraph, the current utterance, and the emotion label into the conditional variational autoencoder to generate an emotional response. In summary, this method can generate emotional responses that are rich in emotional expression and accurate.

The invention provides an intelligent interaction-oriented memory enhancement and reasoning method and system, and belongs to the technical field of intelligent interaction, and the method comprises the steps: firstly capturing an original interaction data stream of a user, segmenting and arranging into an interaction event time sequence, preliminarily determining an interaction intention, and merging into an interaction round unit sequence; extracting a core keyword, performing association expansion to obtain a background support knowledge unit, and generating an interaction situation knowledge set; and fusing the context knowledge coding vector with the interactive content semantics, and retrieving historical memory entries in a long-term interactive memory library. And finally, complementing the current interaction intention according to historical memory, generating response reasoning content and transmitting the response reasoning content to the intelligent interaction terminal for presentation. According to the invention, the accuracy of intelligent interaction and the user experience are improved.

The present invention discloses a contextual knowledge-guided visual relationship detection method and device, comprising: acquiring a scene image, determining the semantic vector and context vector of an object pair in the scene image, extracting a local image containing the object pair from the scene image and determining a position mask of the object pair; utilizing a contextual knowledge generation module to extract the contextual knowledge of the object pair based on the context vector of the object pair; utilizing a visual feature extraction module to extract the visual features of the object pair based on the local image and the position mask of the object pair; utilizing a semantic feature extraction module to extract the semantic features of the object pair based on the semantic vector of the object pair; utilizing a visual relationship detection module to calculate a visual prediction result of the object pair based on the visual features and semantic features of the object pair; utilizing a comprehensive judgment module to comprehensively judge the visual prediction result and the contextual knowledge of the object pair to obtain a contextual knowledge-guided visual relationship of the object pair. The method and device improve the accuracy of visual relationship detection through contextual knowledge guidance.