Patents
Literature
Hiro is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Hiro

48results about How to "Reduce creative costs" patented technology

Multi-label microblog text classification method based on semi-supervised learning

The invention discloses a multi-label microblog text classification method based on semi-supervised learning, and relates to the field of natural language processing. The method comprises the following steps: firstly, preprocessing an original microblog text, and labeling a small amount of texts; generating augmented data of the annotation data set by using reverse translation, generating augmented data of the unannotated data set by using synonym replacement and random noise injection, guessing and generating a pseudo tag of the unannotated data by using a classifier, and forming a new training set together with the augmented annotation data set; converting the multi-label classification task into a plurality of dichotomy tasks, training a semi-supervised microblog text classification model, randomly extracting two samples from a new training set each time during training, generating a new sample in a text hidden space by using a sample mixing technology, calculating a loss value, and updating network parameters; and finally, classifying the microblog texts by comprehensively using a plurality of trained classifiers. The invention has an important application value for fine-grained information extraction of the microblog texts.
Owner:ZHEJIANG UNIV

Music model training method, music creation method, devices, terminal and storage medium

The invention discloses a music model training method. The music model training method comprises the following steps: acquiring a MIDI music data set, wherein the MIDI music data set comprises a plurality of MIDI music scores; extracting the feature vector of each MIDI music score; inputting the feature vectors into a structured support vector machine for training, so that a music model is obtained, the step specifically comprises the following substeps: constructing a discrimination function f(x;w), wherein x is a feature vector, w is a parameter vector, carrying out outputting by adopting the data value (with the calculation formula shown in the description) of the maximal discrimination function f(x;w) as the predicted value, calculating the predicted value and the true value accordingto a preset loss function (shown in the description), wherein P is the probability distribution of data, which is replaced with the empirical risk (shown in the description) obtained through calculation with the trained sample data, solving the unique parameter vector omega by adopting the optimizing formula (as shown in the description) of SVM, so that the empirical risk (shown in the description) obtained through the trained sample data is 0, solving the discrimination function f(x; omega), and finally, outputting the music time sequence. The invention further provides a music creation method, devices, a terminal and a storage medium. In the technical scheme, artificial intelligence is used for music model training for the first time, for the trained music model, the feature extraction capacity of the MIDI music score can be improved.
Owner:PING AN TECH (SHENZHEN) CO LTD

Subject heading classification model creation method and device and storage medium

The invention provides a subject heading classification model creation method. The method comprises the steps that multiple model training documents are obtained, and label terms of the model training documents are extracted; based on a similarity algorithm, core theme phrases corresponding to the label terms are obtained; based on a map content library, a first model training document collection corresponding to the core theme phrases is obtained; based on a machine learning algorithm, the model training documents are subjected to sort operation; based on the map content library, subject type identification of all the model training documents corresponding to the label terms is obtained, and according to the subject type identification corresponding to the label terms, a second model training document collection corresponding to the label terms is determined; repetitive model training documents in the first model training document collection and the second model training document collection corresponding to the label terms are taken as positive samples, other model training documents in the map content library are taken as negative samples, and a subject heading classification model of the label terms is created. The invention further provides a subject heading classification model creation device and a storage medium.
Owner:TENCENT TECH (SHENZHEN) CO LTD

Search-and-rescue positioning system for underground mine disaster

The invention discloses a search-and-rescue positioning system for an underground mine disaster. The search-and-rescue positioning system for the underground mine disaster comprises a wired signal transmission network distributed underground, multiple position finders which are distributed underground and are connected with the wired signal transmission network, a transmitter installed on a movable device, a control center arranged on the ground and a signal receiving terminal used for search and rescue, wherein the control center is connected with the wired signal transmission network and receives positioning signals transmitted by the position finders, the transmitter is worn by an underground operator and moves along with the underground operator, and the signal receiving terminal is carried with a search and rescue person, moves along with the search and rescue person to be close to the underground operator trapped underground and gives a wireless signal response to the transmitter carried with the underground operator. The search-and-rescue positioning system for the underground mine disaster is simple in structure, low in establishment cost and high in positioning speed and positioning precision, and the advancing direction of the underground operator can be confirmed; the position of the trapped person can be found out rapidly when the mine disaster occurs so that rescue can be conducted in time.
Owner:JINLING INST OF TECH

Point cloud processing method, device and equipment, and computer readable storage medium

Embodiments of the invention provide a point cloud processing method, apparatus and device, and a computer readable storage medium. The method comprises steps of constructing a first model and a second model based on obtained point cloud; wherein the first model is arranged in a space formed by the point cloud, and the second model is arranged outside the space formed by the point cloud; expandingand deforming the first model and contracting and deforming the second model; wherein a target distance from the expanded and deformed first model to the point cloud is less than or equal to a firstthreshold, or the target distance from the expanded and deformed first model to the first set is less than or equal to the first threshold; wherein the target distance from the contracted and deformedsecond model to the point cloud is less than or equal to a second threshold, or the target distance from the contracted and deformed second model to the second set is less than or equal to the secondthreshold; wherein the first set comprises point cloud and a second model, and the second set comprises the point cloud and the first model; and fitting and reconstructing a three-dimensional model of the target object based on the point cloud, the deformed first model and the deformed second model.
Owner:HUAWEI TECH CO LTD +1

Pyrotechnic composition for calligraphy and painting creation, and preparation method thereof

The invention relates to a pyrotechnic composition for calligraphy and painting creation, and a preparation method thereof, and belongs to the field of calligraphy and painting creation pyrotechnic compositions. The pyrotechnic composition for calligraphy and painting creation comprises 30-50 parts of ammonium perchlorate, 15-40 parts of metaldehyde, 15-35 parts of copper chloride and 5-10 parts of shellac. The preparation method comprises the following steps: respectively weighing certain amounts of ammonium perchlorate, metaldehyde, copper chloride and shellac according to proportions of the above formula; respectively fully grinding the weighed ammonium perchlorate, metaldehyde and copper chloride in a mortar, respectively sieving by using a 100-200 mesh sieve 3 times, and carrying out full dry mixing on all the obtained sieved reagents until uniformity; and putting weighed shellac in a beaker, adding a proper amount of 95% ethanol to dissolve shellac, pouring the above obtained dry mixed uniform medicine mixture into the beaker after shellac is completely dissolved, fully stirring, and air-drying the obtained medicine in a shady place until no liquid exists. The pyrotechnic composition has the advantages of fast combustion, low combustion temperature, few combustion residues, fire delivery cross the air, short creation time, low creation cost and high safety coefficient.
Owner:SHENYANG LIGONG UNIV

Picture color matching method and device and electronic equipment

Embodiments of the invention provide a picture color matching method and device, and electronic equipment. The method comprises the steps of obtaining a target style data set; for the content elementpart in the target style picture, carrying out probability density estimation based on color features, wherein the color features comprise content element features of the content element part and difference comparison features of the content element part and a product part in a picture to be subjected to color matching; and forming a target scheme according to the probability density estimation result of the target style picture, and re-coloring the content element part in the picture to be subjected to color matching. According to the embodiment of the invention, the color feature-based probability density estimation is performed on the target style picture, and the color matching scheme is automatically formed in combination with the difference between the content element and the productin the picture to be subjected to color matching, so that the creation cost of the commercial picture is reduced, the time spent by a user on content color matching is reduced, and the creation efficiency of the commercial picture is greatly improved.
Owner:ALIBABA GRP HLDG LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products