133results about How to "Realize personalized recommendation" patented technology

The invention discloses a personalized network learning resource recommendation method which comprises the following steps: analyzing behavioral data of learners for visiting a network leaning system based on expanded theme pictures, acquiring leaning interest path changing patterns of the learners and groups of the learners on concepts and knowledge elements related to learning contents, and then realizing active personalized recommendation of proper learning resources to the learners according learning interest path changing patterns of the learner individuals and the groups of the learners individuals, sequence of learning objects of the expanded theme pictures and the like. The personalized network learning resource recommendation method can provide personalized learning service for network learners, thereby improving learning efficiency of the network learners.

The present invention discloses a mobile game recommendation system and a recommendation method thereof. The system comprises: a user use behavior acquisition module for acquiring dynamic behavior information and historical use behavior information of a current user based on behavior logs of game use of the current game user; a user behavior analysis module for analyzing user behaviors namely analyzing the use behaviors of the current user and 'friends' associated with the current user to calculate the user similarity and find out similar user groups; a game application list generation module for clustering games into multiple clusters according to labels, keywords and game brief introduction of game application information and description of user evaluations and the like, and generating a similar game application list for each game; and a game recommendation module for generating a recommendation list by using similar game application lists, integrating a recommendation based on a game application and a recommendation result based on a game user in a weighting manner by selecting appropriate weights to form a final recommendation result, setting corresponding recommendation thresholds according to a game application recommendation quantity, and recommending games not installed by the user to the user for installation and use.

The invention discloses a hybrid clustering recommendation method based on Gaussian distribution and distance similarity. The hybrid clustering recommendation method includes the following steps: S1, acquiring a user behavior data set, dividing the behavior data set into labeled data and no-label data, and performing distance metric learning on the labeled data and the no-label data respectively; S2, according to a distance metric weight matrix and a Gaussian hybrid model, performing hybrid calculation to acquire a target function, and performing optimized solving on the target function; S3, after acquiring an optimized solution of the target function, acquiring clustering behavior data through a clustering algorithm, and recommending clustering behavior data to users.

The embodiment of the invention provides a content recommendation method and device. The method comprises the steps that when a video playing request of a player is received, a video, asked to be played, in the video playing request is sent to the player, and the player is triggered to play the video; and meanwhile, after target content in the video is detected and acquired, recommended content matched with the target content is acquired, so that the player is triggered to output prompt information of the recommended content in a corresponding display region of the target content when the target content is played. According to the embodiment, the recommended content has higher pertinence.

The invention discloses an intelligent course selection recommendation method. According to the intelligent course selection recommendation method, a course map tree is established by calculating the similarity with the current student course selection page access, the similarity with the current student course grade, the similarity with the current student course selection record, recommended course selection sets respectively are obtained, then the obtained course selection sets are combined in a weighting mode to obtain the final recommended course selection set. A plurality of dimensionalities are taken into consideration, individual recommendation of courses is achieved, and blindness of student course selection is effectively solved. Existing system data of a school are utilized very well, mature collaborative filtering and the recommendation technology based on the content are utilized, and the method is easy to achieve.

The invention specifically relates to an individualized learning-oriented educational resource recommendation method and system. The method comprises the following steps of: S1, acquiring basic characteristic data of learners and establishing a resource interestingness model and a learning style model of the learners according to the basic characteristic data, wherein the basic characteristic data comprises identity information, learning behavior records and educational resource operation records of the learners; S2, establishing a learner characteristic model according to the resource interestingness model and the learning style model; S3, calculating the similarity between different learners or the similarity between different educational resources according to the learner characteristic model; and S4, generating a target learner-oriented educational resource recommendation result according to the similarity between different learners or the similarity between different educational resources. Through the technical scheme provided by the invention, the target learner-oriented education resource individualized recommendation can be realized.

The invention discloses a personalized pushing method based on user reading preference statistics of reading time. The method includes following steps: step 1, establishing calculating standards of average reading time of electronic reading materials; step 2, judging reading behaviors of a user; step 3, establishing average reading speed of the user; step 4, calculating reading time for a single article of the user; step 5, judging reading completeness according to actual reading time and the average reading time of the user; step 6, establishing preference degree of the user according to the reading completeness of the user; step 7, performing matching similarity operation on user preference vector and article content feature vector to acquire interest evaluation value of the user on an un-read article, and recommending a series of articles with high interest evaluation value to the user. The preference degree of the user is judged mainly according to article reading time and text information of the user, and then articles in which the user has interest are recommended to the user.

The invention belongs to the technical field of website recommendation, and provides a hybrid recommendation system oriented to a local chronicles website. The hybrid recommendation system comprises a behavior recording module, a user classifying module and a user recommendation module; the behavior recording module is used for recording user registration information and behavior information and storing the user registration information and behavior information into a database; the user classifying module is used for dividing users into new users, general users and active users according to the local chronicles access frequency of the users, conducting feature analysis on each user class to select the suitable recommendation method to achieve personalized recommendation; the user recommendation module comprises TopN recommendation, association rule recommendation based on local chronicles labels and improvement and collaborative filtering recommendation based on the users. The system adopts different recommendation algorithms for users of different type according to the classifying result of the user classifying module. Personalized recommendation can be conducted according to different user types.

The invention discloses a recommendation method and system for application extension tools. The method includes: collecting utilization information about installed extension tools applied by a user; analyzing the collected utilization information to determine user's preference degrees on the installed extension tools, and selecting the first extension tool with the preferred degree reaching a first preset condition; from the user's uninstalled extension tools, selecting the second extension tool having the similarity, to the first extension tool reaching the first preset condition, reaching a second preset condition, and recommending the second extension tool to the user. The recommendation method and system has the advantages that personal recommendations on extension tools can be provided for users and effective utilization rate of the extension tools can be increased.

The invention discloses a partial model weight fusion Top-N film recommending method based on user clustering. The method comprises the steps of 1, preprocessing data, wherein inactive users and filmswith very low popularity are subjected to data cleaning, user film label documents are constructed, explicit scoring information is converted into implicit feedback information, and a user-film implicit feedback matrix A is constructed; 2, conducting user clustering, wherein film label information is utilized, user feature vectors are obtained by training an LDA topic model, and user clustering is achieved through a spectral clustering algorithm; 3, determining a local recommending model and training a global recommending model; 4, conducting model weight fusion recommending; 5, proving the effectiveness of the models through a leave-one-out method.

The invention discloses a user dynamic preference oriented commodity sequence personalized recommendation method. The method comprises the steps of extracting a commodity sequence under the same usersimilarity score to construct a commodity score vector; extracting a user sequence of similar scores of the same commodity to obtain a user score vector; combining user personal information, commoditybasic attribute information, user and article comments and commodity pictures; achieving feature extraction of a user and a commodity based on multi-task learning, taking a user feature vector and afeature vector of a historical commodity sequence of the user feature vector as input, achieving generation of the commodity sequence by training a coder-decoder, and accurately learning recommendation of the optimal commodity sequence in combination with a search strategy. According to the invention, based on the multi-modal user-commodity data, the user features and the commodity features are highly extracted and fused, personalized recommendation of the commodity sequence for user preferences is realized, and the user experience is improved.

The invention discloses a cross-border e-commerce platform having high concurrency extensible distributed type architecture. The cross-border e-commerce platform comprises a NODE front end, a NODE background, an interface layer, a service layer, a data access layer, and a data storage layer. A NODE rear end is used to provide a route request, a GET request, or an AJAX request for the NODE front end, and the interface layer is invoked to receive background request parameters, which are converted, and the service layer is invoked, and then the service layer is used to invoke the data access layer, which is used to invoke data from the data storage layer. The service layer comprises a member module, a commodity module, a logistics module, an order module, and a payment module. RPC remote invoking is adopted among the various modules of the service layer.

The invention relates to a parking recommendation method and device. The method comprises the following steps: receiving a parking request, wherein the parking request includes a vehicle identifier; obtaining user image information corresponding to the vehicle identifier; obtaining state information of a target parking lot; determining a parking space matching the vehicle identifier according to the state information and the user image information; and sending the information of the matching parking space to a terminal. By using the method and the device, users can be guided to park vehicles,personalized recommendation of parking spaces is realized, and the parking efficiency of users is improved.

The invention discloses an image recommendation method merging visual features and user ratings. The method includes the first step of crawling a data set and extracting item images from the data setand a user's scoring matrix of the corresponding item images; the second step of utilizing a convolutional neural network (CNN) to extract image visual features of all the collected item images, to obtain a visual feature matrix; the third step of establishing a prediction preference model, and using an element-based alternating least squares method to update the prediction preference model; the fourth step of obtaining the user's preference values of all the item images from the final prediction preference model, sorting the preference values in descending order, selecting item images corresponding to top preference values, and recommending the item images to the user. The image recommendation method merges the visual features and the user ratings so that recommendation accuracy can be improved and personalized recommendations can be achieved.

The invention provides a method and an apparatus for recommending application information based on installed applications. The method comprises the steps of determining the similarity between applications installed by a user and different information in an information flow library; obtaining matching degrees of the user to the different information by utilizing the obtained similarity; and selecting a certain quantity of information as candidate information according to a sequence of the matching degrees from high to low, and recommending the information to the user according to the sequence of the matching degrees from high to low.

The embodiment of the invention provides a method and device for application recommendation and a server. The method comprises the steps of obtaining attribute labels corresponding to target users; according to the attribute labels and a preference value of each application by a user group having the attribute label, obtaining preference values of the target users for each application; and based on the preference values of the target users for each application, recommending applications for the target users. According to the method, under the conditions of no supervision, the preference of the user group under the different attribute labels is automatically identified according to user behavior data, personalized recommendation for the users is performed according to the attribute labels which the users have, the mode greatly reduces human cost, and the recommending accuracy is improved through objective analysis of the user behavior data.

The embodiment of the invention discloses an information recommendation method, device and equipment and a medium, and relates to the technical field of information. The method comprises the followingsteps of determining at least one historical user similar to a target user as a reference user according to the user characteristics; taking the target class article associated with the historical behavior of the reference user as a candidate article; determining the weight of the candidate article according to the historical behavior and historical behavior weight of the reference user on the candidate article and the similarity between the reference user and the target user; and according to the weight of the candidate item, recommending a target item to a target user. According to the information recommendation method, device and equipment and the medium provided by the embodiment of the invention, the personalized recommendation of the credit card information to the user is realized,so that the personalized demand of the user on the credit card is met, and the information recommendation accuracy is further improved.