39results about How to "Improve novelty" patented technology

The invention provides an intelligent dynamic pushing method and device based on marketing canvas construction, equipment and a medium, and the method comprises the following steps: obtaining basic data of a marketing canvas, and constructing a marketing canvas model according to the basic data; adjusting a marketing pushing strategy weight according to the real-time behavior and historical preference of the user; on the basis of the adjusted marketing pushing strategy weight, a pushing path is optimized according to user feedback and context information; based on the generation model, automatically generating personalized contents of the user and evaluating a pushing effect in real time; according to the method, a marketing canvas model based on multiple data is constructed, data such as products, customers, channels and communication strategies are combined, and real-time behaviors of users are dynamically analyzed and predicted by using a deep learning model. In this way, the accuracy of the user portrait can be improved, the pushing strategy can be updated and adjusted in real time, and therefore accurate personalized pushing is achieved.

The invention discloses a personalized recommendation method, an electronic device and a storage medium, the method effectively couples accuracy, diversity and novelty, and calculates a predicted score of a user to an item, and the method comprises the following steps: S1, constructing a user item score matrix; s2, adjusting the score value of the user to the item through a time decay function, and updating the item score of the user; s3, obtaining a user-feature factor matrix and a project-feature factor matrix; s4, calculating a preference value novelty factor of the user to the item novelty; s5, calculating the diversity tendency degree of the user; s6, the SVD + + model is improved, and a DNSVD model is obtained; step S7, training the DNSVD model; and S8, performing prediction on the test set, and calculating a prediction value of the user to the project. According to the method, accuracy, diversity and novelty are considered at the same time, and the diversity and novelty of the recommendation algorithm are optimized on the premise that the accuracy is good.

The invention relates to an object recommendation interaction method and device, electronic equipment and a storage medium. The method comprises the steps that recommended media content corresponding to a target object is displayed on a preset page; in the recommended media content display process, displaying a first part of picture in a first preset recommended picture corresponding to the target object in a preset area in a preset page, and displaying a second part of picture in the first preset recommended picture in the preset page; in response to a preset interaction operation triggered by the target account, moving the second part of the picture to a first preset sub-region in the preset region to enable the preset region to display a complete first preset recommended picture; and skipping to an associated page corresponding to the target object under the condition that the complete first preset recommendation picture is displayed. By utilizing the embodiment of the invention, the presentation style of the object recommendation picture can be enriched, the effectiveness and interestingness of user interaction are improved, the user interaction experience is improved, and the object recommendation effect and the object recommendation efficiency are improved.

The invention provides a game skill generation method and device, a computer storage medium and electronic equipment, and relates to the technical field of games. Comprising the steps that in response to a first game match start instruction, a first game scene corresponding to a first game map is displayed, and the first game scene comprises a controlled virtual object corresponding to a first game account; responding to the controlled virtual object to complete the first game match, and according to a first virtual item obtained from the first game scene, controlling to open the permission of entering a second game map; in response to a start instruction of a second game match, displaying a second game scene corresponding to a second game map, and newly adding a first game skill for the controlled virtual object; and in response to the controlled virtual object, playing the first game skill in the second game scene, and presenting a game effect corresponding to the first game skill. The depth of the game playing method is expanded, and the novelty and the playing durability of the game are improved.

A method and an apparatus for processing molecular scaffold transitions, a computer-readable medium, an electronic device, and a computer program product are provided. The method for processing molecular scaffold transitions includes: generating, according to a connection graph structure corresponding to a reference drug molecule, an atomic latent vector corresponding to the reference drug molecule; performing atom masking processing on the atomic latent vector to obtain a scaffold latent vector and a sidechain latent vector included in the atomic latent vector; generating a target scaffold latent vector with a target transition degree between the scaffold latent vector and the target scaffold latent vector according to a spatial distribution of the scaffold latent vector; and generating a transitioned drug molecule according to the target scaffold latent vector and the sidechain latent vector.

The present invention discloses a human-machine collaborative conceptual design generation method based on FBS theory, comprising the following steps: obtaining a conceptual design case and extracting keywords based on design requirements and the functions, behaviors and structures in the FBS theory to construct a functional space, a behavioral space and a structural space; embedding the functional space, the behavioral space and the structural space into a large language model framework in sequence based on the logical order of the conceptual design process to construct a conceptual design auxiliary model; inputting the design requirements into the conceptual design auxiliary model to obtain a function-structure mind map. The present invention also provides a human-machine collaborative conceptual design generation system. The method provided by the present invention provides guidance for the designer's thinking process in a structured manner, thereby expanding the design ideas and improving the quality of the final generated design solution.

The present invention provides an automobile signal light and a replacement structure for switching to light up different areas using a mirror shutter, comprising a first LED light source, a second LED light source, a first light-emitting system, a second light-emitting system, a mirror shutter, a reflector, an outer lens and a structural plate, wherein the first LED light source is arranged at the focus of the first light-emitting system, the second LED light source is arranged at the focus of the second light-emitting system, the first light-emitting system and the second light-emitting system are arranged on both sides of the mirror shutter, the outer lens is arranged on the light-emitting side of the first light-emitting system, the reflector is arranged on the light-emitting side of the second light-emitting system, and the structural plate is arranged between light-emitting areas one and two above and below the outer lens, and the lighting areas of the light-emitting function one and the light-emitting function two are changed by opening and closing the mirror shutter, so that the lighting areas of the two are switched with each other.

The invention relates to a multimedia resource display method and device, electronic equipment and a storage medium, and the method comprises the steps: displaying a current resource component corresponding to a current hidden display stage in a multimedia resource display process; the current resource component is obtained by hiding an original resource component based on a current hiding degree, and the current hiding degree is determined based on historical resource interaction information before the current hiding display stage; the original resource component and the multimedia resource have display overlap. The display shielding of the resource component on the resource content of the multimedia resource can be reduced, and the novelty and the flexibility of the display form of the resource component in the multimedia resource display process can be improved.

The invention discloses an intelligent grade classification method based on ore image features, and belongs to the technical field of machine vision, and the method comprises the steps: constructing an ore image data set, dividing the ore image data set into a training set, a verification set and a test set, and carrying out the preprocessing operation of the ore image data in each subset; constructing a residual neural network model of a multi-branch structure as an ore grade detection model; training the ore grade detection model based on the ore image data set; and detecting the grade of the to-be-detected ore by using the trained ore grade detection model. According to the method, the ore grade classification model is constructed based on the multi-branch structure and the residual error structure, the intelligent degree of the mine in the mineral separation link can be improved, guidance is provided for production and ore blending, and the method has important practical application value and significance for improving the operation efficiency of the mine.

The invention discloses a drug design optimization and precision medical analysis method and system based on swarm intelligence. According to the method, behaviors of weak meat and strong food groups are simulated, candidate drug molecules or patient pathological models are regarded as agents, and core data of the agents are protected by hiding virtual models; risks such as toxicity and off-target effect are actively identified and avoided through the security defense model; the intelligent agent is guided to escape from the local optimal solution and the invalid research and development path through the escape model; and finally performing fusion and multiple verification on the elite agent through the survival model, and outputting high-quality and low-risk candidate drug molecules or personalized treatment schemes. According to the method, a safe and anti-interference mechanism is embedded in an optimization process, the efficiency and success rate of new drug research and development are remarkably improved, and a powerful analysis tool is provided for precision medical treatment.

The present disclosure relates to a page display method and device, electronic equipment and storage medium. The method comprises: displaying first object material and second object material on a preset display page of preset media content corresponding to a preset business party; the first object material and the second object material are object materials of different objects corresponding to the preset business party; in response to a page jump instruction, the first object material and the second object material are controlled to collide with each other, and a preset animation including a first preset element is displayed; the first preset element is a common element of the two objects corresponding to the first object material and the second object material; in the case where the preset animation display ends, jump to a business page associated with the preset business party. The embodiments of the present disclosure can enrich the recommended objects, meet the diversified recommendation needs of the business party, improve the image shaping effect of the business party, and improve the novelty and interest of the animation in the page jump process, and improve the effect and interactivity of the joint recommendation of the two objects.

The invention belongs to the field of digital libraries, and discloses a digital library book management system, which comprises a behavior recording module for recording user operation behavior data; the interest analysis module is used for carrying out data processing on the user operation behavior data and extracting user interest preference vectors; the book recommendation module is used for constructing a user prediction model according to the user interest preference vector, and retrieving user operation behavior data through a user output by the user prediction model to obtain a recommended book; the recommendation display module is used for displaying recommended books and supporting the user to borrow and check; and the feedback collection module is used for collecting feedback data of the user on the recommended book, optimizing the user prediction model according to the feedback data, and improving the performance of the recommendation system.

The invention provides a long-tail data distribution-oriented recommendation method and system in a federated environment, and the method comprises the steps: firstly, enabling a current popular neural collaborative filtering recommendation model to be federated according to the comprehensive demands of a recommendation system in the federated environment for user privacy security and recommendation accuracy, so as to obtain a federated recommendation model FedNCF; and the privacy and the high efficiency of the federal recommendation result are ensured. Then, in consideration of personalized requirements of the user on recommended items, a long-tail recommendation model of a user side explicitly decouples the attraction of the items into fame attraction and niche attraction by utilizing decoupling representation learning to effectively mine user preferences; according to the model aggregation method of the server side, global attention aggregation is carried out on fame part model parameters through an attention mechanism, self-attention aggregation is carried out on niche part parameters to complete long-tail perception of a global model under federation, and the diversity and novelty of recommendation results are improved.

The present invention relates to the field of beverage production, and in particular to a novel cocktail preparation method and preparation system for improving taste, wherein the preparation system comprises a cocktail preparation module, a cocktail processing module, a quality detection module, a preservation module and a control module, wherein the cocktail preparation module is used to prepare a liquid cocktail, the cocktail processing module is used to process the liquid cocktail prepared by the cocktail preparation module and obtain a finished cocktail, the quality detection module is used to detect the quality of the finished cocktail obtained by the cocktail processing module, and the preservation module is used to preserve the finished cocktail of qualified quality. This solution refers to molecular gastronomy technology, which allows the preservation form of the cocktail to be changed from liquid to solid, provides a new way to taste the cocktail, and greatly improves the plasticity and novelty of the taste of the cocktail.

The invention relates to a live broadcast room interaction method and device, electronic equipment and a storage medium. The method comprises the following steps: displaying a plurality of interaction objects and a target user in a target live broadcast page; in response to a selection operation for a target interaction object, displaying fusion display information of the target interaction object and the target user in the target live broadcast page; the fusion display information represents that interaction exists between the interaction presentation information of the target interaction object and the target user, and the interaction presentation information of the target interaction object is displayed through the target user; the target interaction object is any interaction object in the plurality of interaction objects. The novelty and richness of interactive feedback result information display can be improved.

The invention relates to the technical field of data processing, and particularly provides an interpretable enterprise service matching method, system and device and a storage medium, and the method comprises the steps: obtaining enterprise user demand data, generating a structured enterprise demand vector, and analyzing the potential demand intention of an enterprise; dynamically determining a recall strategy combination according to the intention, and recalling an initial candidate case set from a case library according to the recall strategy combination; performing semantic analysis on the candidate cases to obtain matching features of each case; fusing the enterprise demand vector, the potential demand intention and the matching feature, calculating an interpretable matching score, and generating a sorted recommendation list according to the interpretable matching score; and finally, based on the recall strategy combination, the scoring elements and the case attributes, generating an explanatory recommendation reason for the recommendation case. According to the method, the problems of cold start, information cocoon room and decision black box in traditional enterprise service matching are effectively solved.

The utility model provides a scene toy which comprises a main board, a castle model is installed on the upper surface of the main board, two symmetrically-arranged light columns are installed on the surface of the castle model, a bottom plate is installed on the lower surface of the castle model through a plug connector, and a sound box is installed on the upper surface of the bottom plate. Detachable front plates are installed at the two corners of the front face of the bottom plate correspondingly, a slide is inserted into the main plate in a deviating mode, the lower end of the slide is inserted into one front plate, a plurality of vertical pipes are inserted into the upper surfaces of the front plates, and a pavilion model used for storing toy models is inserted into the upper end of one vertical pipe. A sofa model used for storing toy models is inserted into one side, far away from the castle model, of the upper surface of the main board. The castle toy has the advantages of being high in playability and multiple in creative playing method and linkage playing method.

This utility model discloses a pull-out animated drawer box, comprising an outer box with a window in the middle of its upper surface. A sliding strip is fixedly connected to the upper inner wall of the outer box, and a BOPP film is fixedly connected to the outer surface of the sliding strip. The length of the BOPP film is greater than the length of the sliding strip. This pull-out animated drawer box, through the arrangement of the outer box, window, BOPP film, animated images, film seam, and base box, allows two different animated images to be pasted on the BOPP film. One animated image is partially on the upper surface of the BOPP film on the sliding strip, and the animation is displayed through the window of the outer box. The other animated image is partially on the lower surface of the sliding strip. When the box needs to be opened and the base box is pulled out, the base box pulls out the BOPP film, and the positions of the two animated images are interchanged by the film seam, completing the animation effect and enhancing the fun and novelty of the pull-out box.

The invention provides a decorative doll which comprises a limb part, an ear neck part and a light-transmitting area, the limb part comprises a heating limb cavity and a support, the limb cavity protrudes forwards to form a hanging space, and the rear end of the limb cavity is fixedly connected to the lower section of the support; the ear neck parts are provided with plugging petals which are positioned in the ear holes and can be used for accommodating dry towels; the light-transmitting area comprises a light-transmitting surface, an expression component and a light bar, the expression component is made of a light-proof material and is arranged on the light-transmitting surface, and a preset distance is formed between the light bar and the edge of the light-transmitting surface; wherein the light-transmitting area is located in the ear neck part, the ear neck part is fixedly connected to the upper section of the support, and light emitted by the lamp strip is bright on the light-transmitting surface after passing through a preset distance. The functions of doll decoration, expression brightening, drying and towel storage can be ingeniously combined together, the space decoration brightening atmosphere is improved, dry and wet towel separation is achieved, and novelty, ornamental value and practicability are achieved. The invention further provides a towel rack comprising the decorative doll, so that the towel rack also has the advantages.

The utility model provides a simulated electronic candle. The electronic candle comprises a power management assembly, an airflow sensing assembly, a candlelight simulation assembly and an electronic candle shell. The power management assembly comprises a USB interface, a power management circuit 1, a lithium battery and a manual switch; the airflow sensing assembly comprises an air sensor, a power management circuit 2, an MCU and an electronic switch; the candlelight simulation assembly comprises an LED, an electromagnet drive circuit, an electromagnet and a magnetic flame sheet. The simulated electronic candle provided by the utility model can simulate the luminous effect of the candle, realizes the on-off control by a user in an air blowing manner, and has novelty and certain market application value.

This invention relates to the fields of bioinformatics and tumor immunology, specifically disclosing a small-sample, multi-cell line TE neoantigen screening method based on third-generation sequencing. The method involves preprocessing long-read data from third-generation sequencing to generate a non-redundant transcript set and recording support. Quality control is performed to remove low-quality and artifact transcripts. TE annotations are mapped to identify and classify TE-exon1 events, forming a candidate list. Dual verification of transcription start sites and short-read splicing connections is completed. Multi-dimensional information is summarized to tier the candidates. For high- and medium-confidence candidates, ORF inference, peptide generation, antigen prediction, and risk filtering are performed. Translation and protein layer evidence verification are completed, and a traceable transcript isoform map and neoantigen peptide list are output. This invention is suitable for small-sample, multi-cell line scenarios. Through multi-step quality control, verification, and tiering, it significantly reduces the risk of false positives and off-target effects, improving the confidence level, evidence level, and stability and safety of the screening results.

The invention relates to the technical field of big data and digital advertisements, and particularly discloses an advertisement putting effect optimization method based on big data. According to the method, an advertisement creative gene pool is constructed, copywriting, color matching, layout and background music are coded into gene segments, and iterative evolution and self-optimization of advertisement materials are achieved through creative population initialization, dynamic feedback evaluation, natural selection, gene cross recombination and an environment pressure driven mutation mechanism. The system updates fitness scores in real time based on multi-source user behavior data, dynamically adjusts mutation probability in combination with external environment factors, and introduces a high-quality gene pool to accelerate convergence and ensure compliance. According to the technical scheme, creative fatigue is delayed, the novelty and conversion efficiency of advertisements are improved, and double optimization of the marketing effect and the user experience is achieved.

The present invention belongs to the technical field of small molecule compounds, and discloses a compound with a 7-hydroxychromone structure, a preparation method and an application thereof. The compound has a structure shown in the following formula (I). The preparation method of the compound comprises the following steps: accessing the marine fungus sclerotium Gibberella to a seed culture medium, culturing on a shaking table, and obtaining a seed culture solution; accessing the seed culture solution to a fermentation medium, culturing statically, and obtaining a fermentation product; extracting the fermentation product statically three times with ethyl acetate, extracting the obtained bacteria three times with methanol, concentrating the extract to obtain an extract, and then using column chromatography separation technology to obtain a compound with a structural formula as shown in formula (I). The compound has the activity of inhibiting the production of nitric oxide in LSP-induced macrophages RAW264.7, and its IC50 is 38.26 μM, respectively, and can be used to prepare anti-inflammatory drugs.

The beverage device with the tea bag comprises a bottle body and a bottle cap, a special-shaped sealing gasket is arranged at the inner bottom of the bottle cap, the tea bag is arranged in the upper end of the bottle body, the two ends of the tea bag are fixed to the inner wall of the bottle body through connecting ropes, a cutting mechanism is arranged on the bottle cap, and the tea bag is arranged on the cutting mechanism. By arranging the bottle body, the bottle cap, the special-shaped sealing gasket, the tea bag, the connecting rope and the cutting mechanism, during use, firstly, the bottle cap is in threaded connection with the bottle body, the bottle body can be sealed, then when the bottle cap and the bottle body are tightened, the special-shaped sealing gasket on the bottle cap can be extruded by the bottle body, and the sealing effect on the bottle body is further enhanced; when a user needs to drink tea, the bottle cap can drive the cutting mechanism to rotate by rotating the bottle cap, and then the connecting rope is cut off through the cutting mechanism, so that the tea bag falls into the bottle body.

The present invention provides a process for bonding metal rings and diaphragms used in the production of acoustic devices, and relates to the technical field of acoustic device production. The process for bonding metal rings and diaphragms used in the production of acoustic devices based on the use of adhesive film, the bonding steps are: S1. Continuously unfold and flatten the curled metal ring raw material sheet through an unfolding device, and use a high-pressure airflow through a jet device to remove dust and other impurities on the sheet when the unfolded sheet moves forward; S2. Set an adhesive film roll above the moving path of the sheet, and fix one end of the adhesive film with a clamp. By compounding the adhesive film on the metal plate for processing the metal ring, a large number of metal rings with bonding surfaces can be processed at one time, without the need to use glue to stick the metal rings one by one, and can be quickly assembled, which greatly improves the bonding efficiency between the metal ring and the diaphragm and improves the novelty of the process.

The application discloses a double-target scientific creativity generation method based on hierarchical reinforcement learning, aiming to improve the novelty and feasibility of generated content simultaneously. The method first uses a closed-source large language model to score academic review texts, builds a consensus dataset and trains novelty and feasibility reward models. In the training phase, the open-source model to be optimized is used as a strategy model. Through hierarchical sampling, novelty and feasibility-driven subgroups are constructed. The trained reward models are used to calculate the subgroup rewards and global rewards for the samples. Then, the subgroup advantages and intergroup advantages are estimated, and the final advantage value is obtained by fusion to optimize the GRPO objective function. Through hierarchical sampling and double advantage estimation mechanism, the method explicitly decouples and fuses conflicting goals, enabling the model to directly generate high-quality creativity without complex iterations during reasoning, achieving efficient and balanced scientific creativity generation.

The invention relates to the technical field of recommendation analysis, in particular to a book recommendation system based on artificial intelligence, and the system comprises a user identity recognition module which carries out the user identity verification through a user ID, and extracts the basic information and borrowing historical data of a user after the user logs in; the borrowing historical data extraction module is used for recording the borrowing time, the borrowing frequency and the book type of each book to form a user behavior feature data set; the cognitive deviation recognition module is used for recognizing the selection trend of the user on specific book types and analyzing the deviation degree existing in user preferences; and the diversified recommendation generation module is used for generating a personalized recommendation list based on the potential deviation mode analysis result of the cognitive deviation recognition module. Compared with the prior art, the method and the device have the advantages that the situation that the books with prejudice or limitation are recommended to the user is effectively avoided, the recommendation accuracy is improved, the diversity of recommendation results is ensured, the user can be in contact with wider book types, the reading experience of the user is enriched, and the overall satisfaction degree is enhanced.

The utility model relates to a hexagonal diameter-variable flip cover, including the cross section is regular hexagon and the top opening of box body, the circumscribed circle diameter of box body lower part is greater than the circumscribed circle diameter of box body upper part, the circumscribed circle diameter of box body middle part gradually reduces from below to top, the top side of box body is flexibly connected with a regular hexagon and can be turned over the box cover, and the box cover covers the top opening of box body, beneficial effect is: the cross section of box body is regular hexagon, and the circumscribed circle diameter of box body lower part is greater than the circumscribed circle diameter of box body upper part, and the circumscribed circle diameter of box body middle part gradually reduces from below to top, therefore the box body after forming is the packing box of big below and small above and middle part diameter-variable, improves the novel of box body appearance.

The invention relates to the technical field of intelligent sports equipment management, and discloses a sports fitness equipment management method based on an artificial intelligence algorithm. According to the method, a static description text and a dynamic use log of equipment are fused to generate a mixed information stream; mode slicing is carried out on the equipment, and structural feature slices and functional feature slices of the equipment are separated out; respectively mining a static topological relation network and capturing a dynamic behavior track sequence, and carrying out space-time alignment to form a joint representation vector; a relational graph model with the equipment as the center is constructed, nodes are equipment entities, and edges comprise static and dynamic association; and traversing edges in the model, identifying and aggregating edge clusters with the same association mode, and abstracting each edge cluster into a management dimension, thereby automatically constructing a multi-dimensional management framework. According to the method, the management dimension is automatically found from the multi-source data, and the intelligent level of management is improved.

The invention discloses a monitoring reminding method and system based on a signboard, and relates to the technical field of monitoring reminding, and the method comprises the steps: carrying out the statistics of a first concentration angle value of a crowd sight line concentration region in a target monitoring region, and when it is judged that a monitor exists in the linear angle position direction of the first sight line concentration region, carrying out the monitoring of the first concentration angle value; according to the first execution monitor, carrying out changeable signboard warning, drawing a first to-be-detected data curve graph and a first original data curve graph, carrying out statistics on a first change degree value, setting a first signboard warning scheme, and when it is judged that no monitor exists in the linear angle position direction of the first sight line concentration area, carrying out warning according to the first signboard warning scheme; variable signboard warning is carried out according to the second execution monitor I and the second execution monitor II, a second to-be-detected data curve graph and a second original data curve graph are drawn, a second change degree value is counted, and a second signboard warning scheme is set. According to the monitoring reminding method and system based on the signboard, the monitoring reminding working efficiency of the signboard can be improved.