18 results about "Expression pattern" patented technology

The invention relates to an intelligent dialogue system and method based on an AI multi-mode large model. The method comprises the following steps: collecting biological characteristic data of a user in real time; cloning a personalized expression mode of the user by using the generative adversarial network; driving the audio avatar to perform multi-round dialogue interaction with the user, capturing a real-time physiological signal of the user through an expression recognition module, and generating a dynamic response content suggestion in combination with a dialogue context; and analyzing the interaction process based on the reinforcement learning model. The multi-modal deep fusion of the voice rhythm features, the language structure features and the facial dynamic features is realized, so that the limitation of single-modal or simple feature splicing in the prior art is broken through, the emotional state of the user can be accurately and comprehensively captured, the intention is expressed and the fine physiological reaction is expressed, and the user experience is improved. And a solid foundation is laid for constructing high-fidelity user representation.

The invention provides a smart home-oriented user behavior intention prediction method and system, and relates to the field of smart home, and the method comprises the steps: collecting a user voice control signal, and extracting a user voiceprint feature to determine a user identity tag; and synchronously extracting user voice control semantic features, and performing constraint expression based on a user identity tag to determine a current behavior intention. And then, obtaining historical behavior sequence data of the user, and performing deep fusion and prediction in combination with the current behavior intention to obtain a final behavior intention of the user. Therefore, the identity tag is deeply integrated into a semantic understanding process, and the identity tag is adaptive to expression modes of different users, so that the accuracy and robustness of user intention prediction are improved in an auxiliary manner in a complex natural language scene.

The invention relates to the technical field of information processing, and discloses an interactive guiding method and system of an intelligent question answering system and a storage medium. The method comprises the steps that the intelligent question-answering system analyzes and processes expression mode data and rhythm preference data to generate a user reaction sequence and an emotion fluctuation sequence; judging the trend of user intention confidence evaluation, adjusting a dialogue guide strategy, and generating an expression ambiguity mark and a clarification path template; extracting question type classification statistical data of multiple rounds of dialogues, determining an effective guidance opportunity, matching dialogue data of similar scenes, and optimizing a hierarchical progressive strategy and a problem boundary narrowing strategy of a guidance problem; and processing the dialogue data of the current interaction round, judging the narrowing degree of the expression ambiguity mark, generating a multi-round dialogue fluency index, and obtaining a stable state of user intention confidence evaluation. The intention recognition accuracy, ambiguity clarification efficiency and multi-round interaction fluency of the intelligent question answering system are improved.

Presented herein are methods for mapping the effects of gene variants using high throughput sequencing and machine learning to determine the pathogenicity of each variant

The invention provides a space gene expression interpolation method based on a de-noising diffusion probability model, relates to the field of gene data expression, and assumes that scRNA-seq and ST sequencing data in the same tissue area have the same or similar gene expression modes. The method comprises the following steps: firstly, capturing a relationship of gene expression abundance in scRNA-seq data through two Markov chains: in a forward process, introducing noise into the scRNA-seq data through t time steps, and in a reverse process, recovering the data through a denoising network based on attention map convolution and Transform; and then applying the trained model to ST data, and predicting the missing part in the ST data by introducing the ST data expressed by the missing gene in the denoising process so as to complement the transcriptome. According to the method, the ST-deleted gene expression can be more efficiently and accurately deduced from the scRNA-seq reference data.

This computer system is communicably connected to a machine learning model, receives an evaluation instruction including information designating a use case of the machine learning model, an input prompt, and an expected answer, generates a plurality of conversion prompts by changing an expression pattern of the input prompt, inputs the input prompt and each of the plurality of conversion prompts to the machine learning model to acquire a plurality of generated answers, compares the expected answer and the generated answer for each of the plurality of generated answers, calculates a similarity score indicating a similarity between the expected answer and the generated answer, and executes statistical processing using the similarity score of the plurality of generated answers, to thereby calculate an index for evaluating the answer accuracy of the machine learning model in the use case.

The invention discloses a scientific and technical literature hidden problem identification method based on turning semantics and a knowledge graph, and relates to the technical field of natural language processing, the method comprises the following steps: preprocessing a target scientific and technical literature to obtain a structured corpus containing a sentence boundary, a grammatical structure and a term annotation; positioning turning words and identifying semantic questions based on a pre-constructed hierarchical turning word bank to obtain a first candidate question set; processing the negative expression in combination with a negative expression pattern library and a co-occurrence analysis mechanism to obtain a second candidate question set; constructing a science and technology knowledge graph to execute hole detection, and obtaining a third candidate problem set; and after clustering, deduplication and scoring are carried out on the obtained three types of sets, a target scientific and technical literature hidden problem structuring result is output. The technical problem that in the prior art, it is difficult to recognize expression in recessive modes such as turning and negative in the scientific and technical literature is solved, and the technical effects that the recessive technical problems in the scientific and technical literature are precisely recognized, and the comprehensiveness and accuracy of recognition are remarkably improved are achieved.

The application provides a spatial transcriptome cell expression pattern recognition method and system, relates to the field of bioinformatics, and performs pretreatment on acquired spatial transcriptome data to obtain positive sample data composed of image features, gene expression data and an adjacency matrix; the image features and the gene expression data belonging to the same cell have a corresponding relationship; the image features and the gene expression data in the positive sample data are randomly adjusted to obtain negative sample data; a graph autoencoder model is constructed, and the graph autoencoder model is trained by using the positive sample data and the negative sample data; sample data to be recognized is input into the trained graph autoencoder model to obtain a finally recognized cell type; the graph autoencoder is constructed, the multi-modal information in spatial transcriptomics is fully utilized, the cell pattern recognition of spatial transcriptomics is realized, and the expansibility and operability are improved, and the recognition accuracy and reliability of the cell expression pattern in a tissue section are improved.

The invention relates to a dialogue data output method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the steps of obtaining a real dialogue sample set; the real dialogue sample set comprises multiple rounds of dialogue data samples in the real dialogue scene; according to the multi-round dialogue data sample, determining universal man-set information and expression mode information; the universal setup information is used for indicating background setup information of a role played by the dialogue rewriting large model; the expression mode information is used for indicating a language expression style of a role played by the dialogue rewriting large model; according to the general man-set information and the expression mode information, indicating a dialogue rewriting large model to rewrite model generation data in the input multi-round man-machine dialogue data, and outputting the optimized multi-round man-machine dialogue data; the optimized multi-round man-machine dialogue data is used for training and optimizing a dialogue large model. By adopting the method, the quality of the dialogue training sample of the large model can be improved.

The invention provides a maintenance risk entity semi-supervised identification method based on self-adaptive reward guidance. According to the method, an initial model suitable for a maintenance professional term and a risk expression mode is constructed for multi-source maintenance texts such as a maintenance record, a defect report and a work card; and performing multi-dimensional semantic analysis on an unlabeled text based on the model to obtain a tag probability, a scene association degree and a risk coupling degree, and generating a high-confidence pseudo tag under the constraint of lexical-level and sequence-level double thresholds. Performing weighted fusion on the manually labeled data and the pseudo label data according to sample confidence and risk features to form a mixed training set; and finally, introducing a self-adaptive reward guide loss function integrated with a multi-dimensional signal, and carrying out semi-supervised optimization training on the sequence labeling model. According to the method, the key risk entity in the maintenance text can be stably identified under the condition that the marked resources are effective, and technical support is provided for a maintenance unit to realize automatic risk identification, hidden danger classification and safety management.

The application provides a method for characterizing the grain-filling process of rice based on dynamic expression genes of transcriptome and application, the method comprises the following steps: step 1: according to the time node sequence arranged in order, obtaining the transcriptome expression data of rice grains in the grain-filling process; step 2: based on the transcriptome expression data, screening a gene set with specific expression patterns in the grain-filling process; step 3: based on the screened gene set, constructing a grain-filling index; step 4: based on the grain-filling index, realizing the quantitative characterization of the grain-filling process of rice grains. The application screens a gene set with stable dynamic expression characteristics based on transcriptome data, constructs a grain-filling index based on the gene expression amount of the gene set, establishes a functional relationship model between the grain-filling index and the grain-filling time, realizes the continuous quantitative characterization of the grain-filling process, and improves the accuracy and stability of the characterization in the grain-filling stage; and the grain-filling time can be deduced based on the functional relationship model, and the development stage of the sample can be predicted.

The application discloses an interactive method, an electronic device and a computer storage medium. The method comprises the following steps: a first electronic device detects whether it is in a playable voice scene according to one or more of the following: whether a voice message is currently received, or whether specific information is detected through a sensor; when the first electronic device is in the playable voice scene, the first electronic device displays a first expression pattern containing a voice identifier in a first expression interface in response to a first user operation for opening the first expression interface, the first expression pattern being associated with a target voice signal; and the first electronic device sends the first expression pattern and the target voice signal to a second electronic device in response to a second user operation acting on the first expression pattern, so that the convenience can be improved.

An information processing device (2) is provided with: a moving image acquisition unit (231) that acquires a moving image in which a user's face is captured; a detection unit (232) that detects, on the basis of the moving image acquired by the moving image acquisition unit (231), one or more expression patterns of the user and expression intensities indicating the degree of the intensity of the expression in each of the one or more expression patterns; and an output unit (235) that outputs beauty information relating to the beauty of the user's face on the basis of the detection result of the detection unit (232). The information processing device (2) further comprises an estimation unit (234) that estimates the risk of occurrence of wrinkles on the face of the user on the basis of the frequency of occurrence and expression intensity of each of the one or more expression patterns detected by the detection unit (232), and an output unit (235) outputs cosmetic information corresponding to the risk of occurrence estimated by the estimation unit (234).

The loading funnel for the prosthetic heart valve includes a proximal connector configured to releasably secure the funnel to a holding tube for the prosthetic heart valve, a distal end for receiving the prosthetic heart valve in at least a partially expanded state, and a passageway extending between the distal end and the proximal connector. The passageway includes a conical portion that is wider at the distal end and narrower at the proximal end and centered along a cone axis. One or more internal fins each extend from a relatively proximal position within the cone portion to a relatively distal position within the cone portion and project toward the cone axis.

The invention discloses a development dynamic expression mode prediction method, model and system based on gene characteristics and a database structure, and relates to the technical field of biological information and artificial intelligence crossing, and the method comprises the following steps: determining a gene set of a development dynamic expression mode to be detected; acquiring multi-dimensional basis feature information of a to-be-detected gene; and inputting the multi-dimensional basis feature information into a development dynamic expression mode prediction model, and outputting a development dynamic mode corresponding to the target gene. According to the method disclosed by the invention, the range of species related to training data in a gene expression related model is greatly expanded, and the dynamic expression mode of the gene in biological tissue or organ development is quickly and efficiently predicted at low cost by innovatively utilizing the multidimensional gene characteristics.

The present invention relates to an intelligent dialogue system and method based on an AI multimodal large model. The method comprises collecting a user's biometric data in real time; utilizing a generative adversarial network to clone the user's personalized expression pattern; driving an audio avatar to engage in multiple rounds of dialogue interaction with the user, capturing the user's real-time physiological signals through an expression recognition module, and generating dynamic response content suggestions based on the dialogue context; and analyzing the interaction process based on a reinforcement learning model. This system achieves a multimodal deep fusion of speech prosodic features, language structure features, and facial dynamic features, thus breaking through the limitations of single-modality or simple feature splicing in existing technologies. It can accurately and comprehensively capture a user's emotional state, expression intentions, and subtle physiological reactions, laying a solid foundation for building high-fidelity user representations.

The application discloses a gene feature-based developmental dynamic expression pattern prediction method, model, system and database structure, relates to the cross technical field of biological information and artificial intelligence, and comprises the following steps: determining a gene set of a developmental dynamic expression pattern to be detected; acquiring multi-dimensional gene feature information of the genes to be detected; inputting the multi-dimensional gene feature information into a developmental dynamic expression pattern prediction model, and outputting a developmental dynamic pattern corresponding to target genes. The application greatly expands the range of species involved in the training data of the gene expression related model, innovatively uses multi-dimensional gene features, and quickly, efficiently and at low cost predicts the dynamic expression pattern of genes in the development of biological tissues or organs.

An interactive phone holder and intelligent interaction method, the method includes the steps of generating a control signal by a phone and performing a corresponding emotional expression group by the phone and the phone holder which cooperate with each other, wherein the emotional expression group includes: displaying at least one of expression pattern and expression animation by the phone, performing at least one basic action by the phone holder, and adjusting a status of an information indicator light on said phone holder in one or more ways.