56results about How to "Match preference" patented technology

The invention discloses an image reconstruction model training method, an image reconstruction method, equipment and a medium. The method comprises the following steps: acquiring an original high-resolution image and an original low-resolution image; inputting the original low-resolution image into a first generation network for image super-resolution reconstruction to obtain a pseudo high-resolution image; inputting the original high-resolution image and the pseudo high-resolution image into a first discrimination network for discrimination, and obtaining a first discrimination result; inputting the original high-resolution image and the pseudo high-resolution image into a perception loss network to obtain a perception loss value; and based on the perception loss value and the first discrimination result, updating model parameters of the first generation network and the first discrimination network, and obtaining a target generation network based on super-resolution reconstruction. The target generation network can reconstruct images with high resolution and containing texture features and structural features of different scales, and has high perception quality.

The embodiments of the invention provide a method and device for controlling vehicle equipment based on passenger personalized information. A vehicle is provided with a vehicle-mounted system, and control parameters of the vehicle equipment are arranged in the vehicle-mounted system. The method for controlling the vehicle equipment based on the passenger personalized information comprises the steps that passenger identity information is recognized, and personalized data of passengers are obtained; and according to the personalized data of the passengers, configuration information of the vehicle equipment is adjusted. According to the control method for controlling the vehicle equipment based on the passenger personalized information, the passenger identity information is recognized, and the personalized data of the passengers are obtained; and according to the personalized data of the passengers, the configuration information of the vehicle equipment is adjusted. The vehicle equipment is controlled based on the passenger personalized information, so that the vehicle conforms to preference of the passengers, and the comfort of the passengers travelling is improved.

The present invention provides a clothes washing or drying machine comprising: means for setting and controlling the operating cycle of the machine, response and user information means, means for processing the commands received from said control and setting means. The control and setting means comprise: selection devices associated to specific control variables that can be selected independently from a plurality of different states, and said response means comprise a plurality of user-information display devices, each one of which is adapted to take a respective multiplicity of states, to each one of which there corresponds a particular combination of the states taken by said control variables. Each combination of said plurality of distinct states of said control variables is associated to a pre-determined set of instructions and programmes residing in said data storage means and adapted to act on respective operating devices of the machine.

The invention discloses a longan fermenting tequila which is characterized in that the longan fermenting tequila is prepared by carrying out the steps of pulping, dissolution assisting and corrosion resisting, pH value controlling, fermenting, distilling and the like on the materials including longan, pectase, a compound fermenting addition agent and active dry yeast. The alcohol content of the tequila is between 15 and 85 degrees; the methanol content (g / L) is equal to or less than 2.0; the copper content (mg / L) is equal to or less than 6.0. The longan fermenting tequila is firstly created in China, is colorless or pale yellow, bright and transparent, has harmony fruit fragrance and full-bodied wine fragrance, is a real invigorant, concentrates the essence of the longan, has a high nutrition value, has the property of wine to meet the favor of Chinese to the wine, does not contain the matters of the wine which assume fragrance and smell, and has a purer taste. Compared with the same kind of products abroad like aquavit, the taste of the longan fermenting tequila is fresher; moreover, the longan fermenting tequila does not contain the extractions of an oaken barrel and meets the taste of Chinese more.

The invention discloses a navigation path framework applicable to reduction of redundant navigation. The navigation path framework comprises a data pre-processing module, a path division module and a path conclusion module, wherein the data pre-processing module is connected with the path division module; the path division module is connected with the path conclusion module; the data pre-processing module is used for extracting user knowledge; the path division module is used for finding an optimal path dividing; the path conclusion module is used for generating individualized path navigation on a familiar path of a user; the data pre-processing module comprises a matching and calibration module, a landmark extracting module and a user knowledge measuring module; the matching and calibration module and the landmark extracting module are connected with the user knowledge measuring module respectively; the user knowledge measuring module is connected with the path division module. Based on the framework disclosed by the invention, a navigation application program is developed and a navigation instruction can be simplified, so that the development cost of the navigation application program is reduced, requirements on resources of navigation application are reduced and navigation information is more easily known by the user.

The invention provides a medical guidance method and system. The method comprises the steps that after a user logs in to a medical management account, basic information of the user is obtained; according to basic information screening, candidate medical institutions are obtained, and the candidate medical institutions are displayed to the user, so that the user can select the target medical institution; according to the target medical institution and the position information of the user, a medical navigation route is planned and generated, and thus the user can go to the target medical institution according to the selected medical navigation route. Accordingly, the manual screening workload is reduced, the time cost is lowered, and meanwhile the user can conduct selection by himself / herself; hospitals conforming to the preference and demands of the user are screened out, and the medical experience is improved.

The embodiment of the invention provides a direct broadcasting room recommendation method and related equipment, which are used for improving the direct broadcasting room recommendation precision. Themethod comprises the steps of determining target score information of a target user for a watching portrait of a target live broadcast room; determining preference information of the target user forthe content label portrait in the target live broadcast room; determining target similarity between the target user and other users in the user set according to the target score information and the watching behavior data; determining the label portrait similarity between the target user and the target live broadcast room according to the content label portrait information of the target live broadcast room and the preference information of the target user for the content label portrait in the target live broadcast room; determining the user portrait similarity between the target user and the target live broadcast room according to the target score information and the target similarity; determining the similarity between the target user and the target live broadcast room according to the label portrait similarity and the user portrait similarity; and recommending the live broadcast rooms with the similarity greater than a preset value to the target user.

The present invention provides a music recommendation method combined with an emotional state of a user for improving music recommendation effect. The method comprises the following steps: extracting and modeling an emotional context based on an emotion dictionary; constructing a music record that comprises the emotional context; and recommending music based on the emotional context of the user. The music recommendation method provided by the present invention primarily uses microblog data issued by the user, common microblogs and music sharing microblogs included, so that music demand and preference of the user in different emotional states are obtained; and the current emotional state of the user and the music preference of the user under the emotional state during recommendation are taken into consideration, thereby making the recommended music more close to the preference of the user under the emotional state.

The invention discloses a navigation route recommending method, system and devices. The navigation route recommending method includes: by a navigation server, determining route nodes where users needto pass from a starting point to a destination point according to information of the starting point and the destination point in a navigation request sent from a navigation client side; to optional two road nodes, matching saved data of users' common routes which are acquired and reported by the navigation client side; according to a matching result and sequence of the road nodes, acquiring commonroutes between the optional two road nodes of the available matched common routes; when there are no available common routes between the optional two road nodes, calculating planned routes between the optional two road nodes; combining the acquired common routes and / or planned routes according to sequence of the nodes to obtain a recommended route from the starting point to the destination point;by the navigation server, pushing the acquired recommended route to the navigation client side. The The navigation route recommending method, system and devices can recommend driving routes which aremore reasonable according to users' preferences well and achieve finer granularity of matching.

The invention discloses a mobile terminal APP usage prediction method combining with time-context based on a sequence mode, comprising steps of using a sequence mode to extract the private app usage relevance of the user from a user history app usage sequence, which is the user application usage mode, combining with the fact whether the user is in the mobile phone usage active period, and comprehensively consider the recent behavior mode of the user during the prediction and the time-context information that the user currently use so as to enable the predicted application usage to accord with the real-time demand and preference of the user.

Embodiments of the invention provide a data object pushing method, device and system. The method comprises the following steps of: receiving a data object access request, associated with a second user, sent by a client corresponding to a first user, wherein the second user has a plurality of associated data objects and a plurality of data object sets determined according to the plurality of data objects; aiming at the access request, determining one or more data objects and determining a target data object set associated with the first user from the plurality of data object sets; and selectingat least one data object from the target data object set and the one or more data objects determined for the access request to send to the client. According to the method, device and system, most associated data objects can be pushed to users by combining practical intentions of the users.

The invention relates to a multimedia resource recommendation method and device, electronic equipment and a storage medium, and relates to the technical field of computers. According to the embodiment of the invention, the interests of the user account are integrated and analyzed through the multimedia resources interacted by the user account, which multimedia resources are the core interests of the user account are analyzed according to the aggregation condition determined by the multimedia resources according to the similarity, and the candidate interests of the user account are analyzed according to the similarity between the multimedia resources and the candidate multimedia resources. According to the method, the multimedia resources which are similar to the candidate multimedia resources can be analyzed, and the multimedia resources which are similar to the candidate multimedia resources and can represent the core interest of the user account can be naturally analyzed from the multimedia resources interacted with the user account by combining the two factors. Therefore, multimedia resource recommendation is performed for the user account, the recommended target multimedia resource can conform to the core interest of the user account and the preference of the user, and the accuracy of a recommendation result is improved.

The application embodiment provides an application program recommend method comprising the following steps: obtaining user behavior information; calculating first similarity between a preset application program and the user behavior information, and using the first similarity to form a first ranking value of the preset application program; calculating second similarity between the preset application program and preset characteristic data, and using the second similarity to form a second ranking value of the preset application program; using the first ranking value, the second ranking value and a preset weight coefficient to form a ranking score of the preset application program; using the ranking score to present the application program. The method fuses the weight coefficient with the user behavior information and preset characteristic data, thus forming the ranking score, better combining the relation between the user behavior information and preset characteristic data, and allowing the recommended application program to better comply with user preference.

The invention discloses a user characteristic prediction method based on a user purchase behavior, and the method comprises the steps: collecting target user characteristic information, historical order information and order commodity information, obtaining triple knowledge related to a commodity from a public knowledge graph through the order commodity information, constructing a knowledge sub-graph, aggregating entity neighbor local features by using a graph convolutional neural network, and fully learning representation vectors of entities. In the user feature prediction model, the similarity between the target user and different commodities and the similarity between the target user and users with similar purchase behaviors are learned according to different commodity features and different features of the users with similar purchase behaviors, feature vectors of the target user and the users with similar purchase behaviors are learned fully according to the similarity between theusers and the similarity between user entities, and personalized requirements of the users are met. According to the method, the accuracy of user feature prediction is improved, so that the user features can be predicted more accurately, and a more complete user portrait is constructed.

The present invention relates to a music recommendation method combined with an emotional state of a user for improving a music recommendation effect. The method primarily uses microblog data issued by a user, common microblogs and music sharing microblogs included, in order to acquire a listening record of the user and corresponding associated data of an emotional context, and uses a graph model to indicate the associated data; According to the music recommendation method disclosed by the present invention, during recommendation, the current emotional state of the user is taken into consideration, and a Personal Rank method is used to traverse the graph model, thereby recommending music that meets the current preference of the user.

The present invention discloses a context-aware music recommendation method based on a graph embedding model. The method comprises: S1. extraction of a music feature based on a graph embedding model; S2. extraction and modeling of a user global music interest and a contextual music interest; and S3. context-aware music recommendation. According to the method disclosed by the present invention, a feature of music is extracted from playback data of a user and metadata of the music by using a graph embedding model; then a global music interest and a contextual music interest of the user are acquired from a full playback record and a recent playback record of the user; and finally, the global interest and the current contextual interest of the user are comprehensively considered during recommendation, so that recommended music can comply with the real-time demand and preference of the user, and therefore, the search cost of the user is reduced and satisfaction of the user is improved.

The invention discloses a keyword Skyline fuzzy query method and system based on a road network. The method comprises the following steps of: constructing a KR-Tree index, converting keywords input by a user into triple forms (L, R and T) which can be identified by a computer, calling the KR-Tree index, quickly retrieving a database according to user query conditions, and finally returning a retrieval result to the user. The system comprises an index construction module for constructing a KR-Tree index, an input module for inputting keywords, a conversion module for converting the keywords into a triple form, a retrieval module for retrieving a database and an output module for displaying retrieval results. According to the method, the preference requirement of a query user is fully met, the error-tolerant rate of query is increased, and the query efficiency is improved by improving the pruning efficiency of irrelevant nodes in the query processThe invention discloses a keyword Skyline fuzzy query method and a keyword Skyline fuzzy query system based on a road network. The keyword Skyline fuzzy query method comprises the following steps: firstly constructing a KR-; tree index is carried out, then the keyword input by the user is converted into a triple form (L, R, T) which can be identified by a computer, and then KR-is called; the Tree index is used for quickly retrieving the database according to the query condition of the user, and finally, the retrieval result is returned to the user; the system comprises the following steps: constructing KR-; the system comprises an input module for inputting keywords, a Tree index construction module, an input module for inputting keywords, a conversion module for converting the keywords into a triple form, a retrieval module for retrieving a database and an output module for displaying retrieval results. The preference requirements of the query user are fully met, the query fault-tolerant rate is increased, and the query efficiency is improved by improving the pruning efficiency of irrelevant nodes in the query process.

The invention discloses a fig distilled formulated wine and a preparation method thereof. The fig distilled formulated wine is prepared from, by mass, 40%-50% of a fig component, 20%-30% of glutinous rice, 70%-90% of water, 6%-10% of distilled liquor, 5%-9% of auxiliary materials, 3%-5% of pectinase, 4%-6% of cellulase, 6%-12% of an additive and acidic solutions, 10% to 20% of active dry yeast, and sweet wine yeast and wine yeast powder. The preparation method of the fig distilled formulated wine includes the steps of sorting, washing, crushing, soaking, pressing and filtering, soaking of the glutinous rice, steam cooking, spraying, tank storing and nesting, fermenting, distillation and pressing, blending, sterilization, and ageing and packaging. The fig distilled formulated wine and the preparation method thereof have the advantages that the raw materials are diverse, the taste is good and is in line with the Chinese people's preference for baijiu, the vitamin C content is high, and the added value of the product is high; besides, the preparation method of the fig distilled formulated wine has a moderate dosage of the fig raw material and takes full advantage of the fig raw material.

The invention discloses a context-sensing music recommendation method based on a neural network model. The method comprises the step of extracting of music characteristics on the basis of the neural network model and modeling of overall situation interest of a user, extracting of context listening interest of the user and context-sensing music recommendation. According to the context-sensing music recommendation method based on the neural network model, the music characteristics and the overall situation interest characteristics of the user are extracted from a music listening sequence of the user by means of the neural network model, the context listening interest of the user is extracted from the complete listening sequence of the user, the overall situation interest of the user and the current context listening interest of the user are comprehensively considered when recommendation is conducted, therefore, the recommended music can meet the real-time requirement and preference of the user, the search cost of the user is reduced, and the satisfaction degree of the user is improved.

The present disclosure is directed towards receiving media asset recommendations for one or more users according to user profile information. A user interface can be manipulated to selected user representations and have such representations placed in different zones (305, 310, 315, 320). Based on the representations selected, corresponding user profile data that corresponds to the selected user representations can be used to acquire media asset recommendations for the one or more users that were selected. The media asset recommendations can be displayed in a recommendation area (330) as pictures, video, and / or poster art, of the recommended media assets.

The invention discloses a backlight adjustment method and device and a storage medium, and is applied to electronic equipment, and the method comprises the steps: determining a region image of a target region of interest of a user for a display screen; performing feature extraction on the regional image to obtain first feature information; obtaining an eye region image of the user, and performing feature extraction on the eye region image to obtain second feature information; determining a target backlight adjustment parameter according to the first feature information and the second feature information; and performing backlight adjustment according to the target backlight adjustment parameter. By adopting the embodiment of the invention, the backlight adjustment intelligence can be improved.

The invention provides an intelligent recommendation method and device for reservation time, a storage medium and a terminal. The method comprises the following steps: acquiring institution management information, employee work information and user reservation project information; based on the institution management information, the employee work information and the user reservation project information, obtaining a user reservable time period; and screening the user reservable time periods based on user historical reservation information and / or an institution reservation strategy to obtain recommended reservation time periods. The method can quickly and intelligently recommend the appointment, considers the information of the user, the institution and the employee, facilitates the improvement of the service satisfaction of the user, improves the profit capability of the institution, and guarantees the service quality of the institution. Moreover, the modes of historical data analysis, reservation related factor weight setting, reservation resource comprehensive scoring and the like are applied, reservation recommendations of two strategies of user priority and institution priority can be given respectively, and the satisfaction degree of the patient and the reservation efficiency of the medical institution are remarkably improved.

The invention provides a picture random distribution device and a picture random distribution method. The picture random distribution device comprises a verification unit and is used for receiving user information from a terminal, a matching unit which is used for selecting pictures matched with the user information from a picture gallery, a synthesizing unit connected to the matching unit and used for synthesizing the pictures and designated cartoon pictures in a cartoon book requested by the terminal to obtain new cartoon pictures, and a replacing unit which uses the new cartoon pictures to replace the designated pictures in the cartoon book and provides the processed cartoon book to the terminal. The picture random distribution device can distribute and push pictures according to the habit and preference of a user, thereby improving cartoon reading experience of the user.

The invention discloses a listening sequence and metadata based context-sensing music recommendation method. The method comprises the steps of S1 extracting music characteristics based on a music listening sequence and music metadata; S2 extracting a global interest and a context listening interest of a user; and S3 performing context-sensing music recommendation. According to the method, the music characteristics are extracted from the music listening sequence and the music metadata of the user by utilizing a neural network model, then the global interest and the context listening interest of the user are extracted from a complete listening sequence and a recent listening sequence of the user, and finally the global interest and the current context listening interest of the user are comprehensively considered, so that recommended music can meet real-time demands and preferences of the user, the search cost of the user can be reduced, and the satisfactory degree of the user can be increased.

The invention discloses a data pushing method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining a data pushing request sent by a client, wherein the data pushing request comprises a user ID; based on the user ID, obtaining a corresponding user role and a historical pushing rule; if the state corresponding to the historical push rule is a default state, creating a default push rule corresponding to the user ID based on the user role, the default push rule comprising at least one editable component corresponding to the user role and at least one default push parameter corresponding to each editable component; based on the default pushing parameter corresponding to each editable component, pushing and displaying the editable component in the current interface; and displaying the editable components in a current interface in a hidden or sequential manner according to each current pushing parameter in the historical pushing rule. According tothe method, the data pushing system better conforms to user preferences, a simpler and more direct current interface is provided for a user, and the information utilization rate is increased.

Embodiments of the present invention provide a live room recommendation method and related equipment, which are used to improve the recommendation accuracy of the live room. The method includes: determining the target score information of the target user for viewing portraits in the target live broadcast room; determining the target user's preference information for the content label portraits in the target live broadcast room; According to the target similarity of other users in the target live room; determine the similarity of the tag portrait between the target user and the target live room according to the content tag portrait information of the target live room and the target user's preference information for the content tag portrait in the target live room; according to the target score information and Target similarity determines the user portrait similarity between the target user and the target live broadcast room; determines the similarity between the target user and the target live broadcast room according to the similarity of label portraits and user portraits; recommends the live broadcast room with a similarity greater than the preset value to the target user.

The invention relates to a method that detects defective printing nozzles in an inkjet printing machine (7) by a computer (6), which comprises the steps of: printing a printing nozzle test chart (12)for detection, the printing nozzle test chart (12) being detected by at least one image sensor, and analyzing the detected printing nozzle test chart (12) by means of the computer to identify the defective printing nozzles, wherein in order to create the printing nozzle test chart (12), every printing nozzle prints a surface (13, 14); the computer (6) determining a centroid of every surface (13, 14) to analyze the detected printing nozzle test chart (12) and, for all surfaces (13, 14), the computer compares an actual position of a determined centroid to a respective target position.

The present invention discloses a context-aware music recommendation method based on a graph embedding model. The method comprises: S1. extraction of a music feature based on a graph embedding model; S2. extraction and modeling of a user global music interest and a contextual music interest; and S3. context-aware music recommendation. According to the method disclosed by the present invention, a feature of music is extracted from playback data of a user and metadata of the music by using a graph embedding model; then a global music interest and a contextual music interest of the user are acquired from a full playback record and a recent playback record of the user; and finally, the global interest and the current contextual interest of the user are comprehensively considered during recommendation, so that recommended music can comply with the real-time demand and preference of the user, and therefore, the search cost of the user is reduced and satisfaction of the user is improved.