9results about How to "Guaranteed classification effect" patented technology

The invention relates to the field of artificial intelligence, in particular to the technical field of natural language processing, provides an information processing method and device, electronic equipment and a storage medium, and aims to improve the accuracy of a large language model question and answer result. The method comprises the steps of receiving a target question input by an object; inputting the target question into a trained classification model, and obtaining a question intention to which the target question belongs; determining a processing flow corresponding to the target problem according to the problem intention; wherein if the problem intention is the first type of intention, the processing flow is that after intermediate processing is carried out on the target problem, an intermediate processing result is input into the finance and taxation large model; if the problem intention is a second type intention, the processing flow is to directly input the target problem into the finance and taxation large model; and taking an output result of the large finance and taxation model as an answer to the target question. The target questions are classified through the classification model, so that question intention recognition can be simply and efficiently realized, and the accuracy of a big language model question and answer result is improved.

The invention discloses an electromechanical equipment cross-domain residual life prediction method and system based on dynamic parameter learning, and the method comprises the steps: obtaining source domain data and target domain data of electromechanical equipment, and carrying out the preprocessing of the source domain data and the target domain data; the method comprises the following steps: independently pre-training on labeled source domain data by using a feature extractor to enable the feature extractor to have feature extraction capability on the source domain data; training a feature extractor by using a domain adaptive method, so that the feature extractor can extract effective high-dimensional data features from the preprocessed source domain data and target domain data; in the reasoning stage, only the feature extractor and the classifier are applied to predict the residual life of the electromechanical equipment. Through real-time residual life prediction, an enterprise can make a preventive maintenance plan in advance, the service life of equipment is prolonged, the failure rate and the maintenance cost are reduced, and the overall operation efficiency and the equipment reliability are improved.

A base classification method comprises: acquiring a fluorescence image to be identified; identifying a position of each DNA nanoball in the fluorescence image; extracting a brightness data feature of the fluorescence image, the brightness data feature comprising brightness data of M dimensions corresponding to the position of each DNA nanoball; and inputting the extracted brightness data feature into a preset base identification model to obtain a base category corresponding to the fluorescence image. The application also provides a gene sequencer and a computer readable storage medium. The application can realize end-to-end classification from brightness data of a fluorescence image to a base category.

This invention discloses an adversarial defense method for a speaker recognition system based on F-ratio adaptive masking. The method first extracts features from the input speech to obtain an amplitude spectrogram; then, it denoises the amplitude spectrogram to obtain a denoised amplitude spectrogram; F-ratio is used to statistically analyze the high-relevance and low-relevance frequency band sets in the denoised amplitude spectrogram to distinguish the speaker; next, masking thresholds for the high-relevance and low-relevance frequency bands are calculated separately; the amplitude spectrogram is then masked to obtain a reconstructed amplitude spectrogram; the reconstructed amplitude spectrogram is transformed using librosa.griffinlim to obtain the corresponding waveform signal as the reconstructed speech; finally, a batch of clean samples is used for speech reconstruction, and the reconstructed speech is used for SRS fine-tuning training to ensure the classification performance of SRS. This invention exhibits significant defensive advantages, and the average defense capability demonstrates the versatility of this scheme against different attacks. Furthermore, because this invention does not involve additional data and training, it possesses low cost.

The invention belongs to but is not limited to the technical field of artificial intelligence, and discloses an Alzheimer's disease diagnosis method based on electroencephalogram signal channel selection and a gating circulation unit, which selects an electroencephalogram channel through a particle swarm algorithm, reduces redundant information in electroencephalogram signal data, and improves the accuracy of electroencephalogram data analysis; feature extraction is carried out on the electroencephalogram signal data through principal component analysis, and the importance and distinction degree of features are improved; and finally, inputting the electroencephalogram signals into a gated circulation unit neural network for classification prediction, thereby reducing the number of electroencephalogram channels and reducing the dimension of data input. And the calculation efficiency is improved while the classification performance is ensured. According to the method, the classification performance of the Alzheimer's disease patients and the healthy subjects is improved, and accurate and efficient classification of the Alzheimer's disease patients and the healthy subjects is achieved.

The low-power pulse wave signal arrhythmia classification method and system of the application comprises the following steps: using a large convolution kernel to extract the bottom local features of the original signal, and reducing the sequence length by half through pooling; adopting a leaky integral-discharge neuron activation to make the data pulsed; using a large convolution kernel and a hollow convolution to expand the receptive field and capture longer-range local timing features; dimensionally reducing the output features, cooperating with batch normalization, leaky integral-discharge neurons and maximum pooling, and aggregating the local features; calculating self-attention in each subsequence, combining a linear mapping layer, focusing the model on the feature correlation in the local subsequence through the aggregation of the sub-attention map, and extracting the waveform dependence in the short time window of the pulse wave signal; calculating QKV interaction on the whole sequence, combining a linear layer and a pulse neuron, capturing long-distance global feature correlation, and mining the waveform dependence in different time periods of the pulse wave signal. The application can achieve a classification performance equivalent to that of a deep artificial neural network.

The invention discloses a breath sound classification detection method based on an improved convolutional neural network, and the method comprises the steps: firstly carrying out the preprocessing and feature extraction of a breath sound signal collected in real time, and constructing a breath sound cepstrum feature matrix; and then the breath sound features are input into an improved convolutional neural network comprising a convolution module, a channel attention mechanism module and a lightweight feature extraction module for training and classification, and accurate recognition of breath sound categories is realized. The method realizes real-time, efficient and high-precision classification detection of breath sound, has the advantages of low calculation complexity, strong real-time performance and high identification accuracy, and is suitable for the field of biomedical signal processing.

The application discloses an encrypted traffic classification model and method based on a dual-mode hierarchical encoder, constructs a dual-domain hierarchical flow representation (DFHR) matrix, extracts and fuses traffic byte and timing depth features through a dual-mode hierarchical encoder (DMHE), completes self-supervised pre-training by using a hierarchical mask multi-task mask autoencoder (HM-MAE), and outputs a classification result through an adaptive gating classification head (AGCH) after supplementing features through a traffic feature enhancement module (TFEM). The application solves the problems of insufficient single-mode feature extraction capability, small-size matrix information loss, large deviation between pre-training and downstream tasks, and poor generalization of the classification head of the prior art, greatly reduces the model training time, maintains high classification accuracy and robustness in small sample and lightweight scenarios, and adapts to the deployment requirements of edge resource-limited devices.

The invention provides a sound event classification model training method, an electronic device, a storage medium and a computer program product, and the method comprises the steps: S1, inputting the audio features of a plurality of pieces of audio data in a specified data set into a first original model and a second original model, obtaining a first prediction category output by the first original model and a second prediction category output by the second original model; s2, determining prediction loss according to the first prediction category, the second prediction category, the category label of the original category data and the category label of the new category data, and adjusting parameters of the second original model based on the prediction loss; and S3, returning to the step S1, and taking the converged second original model as a sound event classification model through multiple iterations. According to the scheme, the problem of knowledge forgetting caused by introduction of new category data is avoided.