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

136results about How to "In line with preferences" patented technology

Method of recommending online music based on playing record of mobile terminal and mobile terminal

The invention discloses a method of recommending online music based on playing records of a mobile terminal. The method of recommending the online music based on the playing records of the mobile terminal comprises the following steps: recording playing times, a singer, and a music style of each song, periods when the song is played every time and geographic positions and moving states of the mobile terminal when the song is played every time to generate a plurality of playing records; extracting the plurality of playing records to form a first set, wherein the number of the playing times of the playing records are larger than a first preset value; extracting a plurality of playing records corresponding to singers or music styles to form a second set, wherein playing times of the songs of the singers or the music styles are maximum; serving the singers or the music styles contained by the playing records in the second set as a personalized label; and recommending the online music according to the personalized label. Compared with the prior art, the method of recommending the online music based on the playing record of the mobile terminal achieves the facts that the online music which is high in matched-degree of a user preference is recommended to the user, the fact that the user carries out cockamamie web search is avoided, time is saved, and user experience is improved. The invention further discloses the mobile terminal of recommending the online music based on playing records.
Owner:GUANGDONG OPPO MOBILE TELECOMM CORP LTD

Target video clip extraction method and device

In order to solve the problem of small relevance between target video clip extraction and user interestingness, the invention provides a target video clip extraction method, which comprises the following steps of: obtaining target scene classification information and target person information; identifying a scene to obtain a scene classification wonderful degree score of each video clip; identifying the character to obtain a target character identification wonderful degree score of each video clip; generating a user option wonderful degree score according to the scene classification wonderfuldegree score and the target character identification wonderful degree score; according to the image difference, obtaining an image wonderful degree score of the video clip; according to the short-timeenergy value of the audio frame, acquiring an audio highlight score of each video clip; obtaining a content highlight score according to the image highlight score and the audio highlight score; obtaining a diversity score between the video clips according to the distance between the video clips; and according to an optimization objective function f(X)=[Sigma]i<XUScore(i)*w1+[Sigma]i<XCScore(i)*w2+[Sigma]i, j<XDScore(i, j)*w3, selecting a preset value Nsel video clips to form a target video clip set X, so that the value of f(X) is maximum.
Owner:WUXI YSTEN TECH

Comprehensive managing and monitoring system and method for advertisement

The invention relates to the technical field of advertisement putting management systems, in particular to a comprehensive managing and monitoring system and method for an advertisement. The comprehensive managing and monitoring system for the advertisement comprises an advertisement putting subsystem, wherein the advertisement putting subsystem comprises an advertising position user acquisition module, a user information acquisition module, a user analysis module and a strategy generation module; the advertising position user acquisition module is used for obtaining the feature data of the mobile equipment of the user of an advertising position surrounding area; the user information acquisition module is used for obtaining user information from a third-party platform according to the feature data; the user analysis module is used for obtaining the passenger flow volume data of the advertising position surrounding area according to the feature data, and the user analysis module is usedfor generating the user representative data of the user of the advertising position surrounding area according to the user information; and the strategy generation module is used for generating the advertisement putting strategy of the advertising position according to the passenger flow volume data of the advertising position surrounding area and the user representative data. By use of the comprehensive managing and monitoring system for the advertisement, the problem in an existing outdoor advertisement putting process that audiences can not be subjected to personalized analysis to cause inaccurate advertisement putting can be solved.
Owner:珠海华发新科技投资控股有限公司

Live broadcast dressing recommending method, device thereof, calculating equipment and storage medium

The invention discloses a live broadcast dressing recommending method, a device thereof, calculating equipment and a computer storage medium. The method comprises the steps of selecting a frame imagewhich comprises a specific object in a video; performing scene segmentation processing on the frame image, and obtaining a to-be-identified area image which aims at the specific object; identifying the to-be-identified area image, and obtaining characteristic information of the specific object; matching the characteristic information and live broadcast dressing by means of a preset matching rule,determining the to-be-recommended dressing; and recommending the to-be-recommended dressing to a user; wherein the characteristic information comprises one kind or multiple kinds of information selected from dressing characteristic information, skin color characteristic information, age characteristic information, gender characteristic information and body form characteristic. According to the method of the invention, high-efficiency and high-accuracy scene segmentation processing is realized by means of a deep learning method. According to the live broadcast dressing recommending method, moreaccurate identification for the characteristic information of a figure in the frame image can be realized based on a scene segmentation result, and live broadcast dressing recommending is performed according to the characteristic information.
Owner:BEIJING QIHOO TECH CO LTD

Frequency modulation broadcast receiving method for vehicle end

A frequency modulation broadcast receiving method for vehicle end comprises determining a geographical position of the vehicle end to determine a current location area of the vehicle end; acquiring, based on the current location area, a local broadcasting radio station list which is corresponding to the current location area and is stored in the vehicle end; if the local broadcasting radio station list is not acquired in the vehicle end, automatically acquiring the local broadcasting radio station list according to radio station radio wave of the current local area and storing the local broadcasting radio station list, and when the vehicle end arrives at the current local area again, retrieving the stored local broadcasting radio station list; and based on the local broadcasting radio station list, determining which radio station is to be selected and playing the selected radio station. When a vehicle end enters a new area, the local broadcasting radio station list that is corresponding to the current location area of the vehicle end and is stored in the vehicle end is retrieved by the vehicle end, and thus a broadcasting radio station searching process is saved, a radio station is selected and determined in a relatively short period, and user experience of the vehicle end is improved.
Owner:SHANGHAI PATEO ELECTRONIC EQUIPMENT MANUFACTURING CO LTD

Video recommendation method and device

The invention discloses a video recommendation method and device. The method comprises the following steps: 1, ; obtaining interaction data sent by the terminal, the interaction data being data generated when the terminal plays an interaction video to a first interaction node for interaction operation, the interaction video comprising the first interaction node and a second interaction node, the first interaction node being in front of the second interaction node, and the interaction video comprising a plurality of branch plot videos at the second interaction node; generating plot decision data of the interactive video according to the interactive data; according to the plot decision data, determining a target branch plot video of the terminal at the second interaction node in the plurality of branch plot videos; and sending the target branch plot information to the terminal, so that the terminal plays the target branch plot video at the second interaction node according to the targetbranch plot information. According to the method, favorite plot branches are provided for users, the user stickiness is improved, and a play space is provided for creators. According to the invention,the favorite branch plot of the user can be recommended through an artificial intelligence method.
Owner:深圳市雅阅科技有限公司

Pushing method and pushing system for management course

The invention provides a pushing method and a pushing system for a management course. The pushing method comprises the following steps that: obtaining time information and user liveness information from a user terminal, sending the time information and the user liveness information to a cloud server, and generating an active time period by the cloud server on the basis of the time information andthe liveness information; sending a standardized questionnaire to the user terminal, and sending an answer, which aims at the standardized questionnaire, of the user to the cloud server; judging the answer by the cloud server, analyzing a judgment result, and obtaining an analysis result, wherein the cloud server is based on the analysis result; when the length of the active time period is smallerthan a first threshold, pushing a first category of courses to the user terminal; when the length of the active time period is greater than the first threshold and small than a second threshold, pushing a second category of courses to the user terminal; and when the length of the active time period is greater than the second threshold and small than a third threshold, pushing a third category ofcourses to the user terminal. By use of the method disclosed by the invention, course pushing is targeted and humanized.
Owner:河南工学院
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