1458 results about "Recommendation model" patented technology

The invention provides a personalized film recommendation system and method based on attribute description, comprising a candidate film data base, a viewing record data base, a user interest preference model calculation module and a film recommendation model. The personalized film recommendation system uses the attribute description to represent the film feature and is hierarchized according to the types and analyses the user interest hot spot of concentration ratios of different hierarchy and tracks the user interest change using the time forget factor and film model linear superposition method, responds the feedback of the user on the film at real time, adjusts the user interest preference model and increases the accuracy and adaptability of recommended film.

The invention discloses a personalized recommendation method based on deep learning. The method comprises the steps of according to the viewing time sequence behavior sequence of the user, predictingthe next movie that the user will watch, including three stages of preprocessing the historical behavior characteristic data of the user watching the movie, modeling a personalized recommendation model, and performing model training and testing by using the user time sequence behavior characteristic sequence; at the historical behavior characteristic data preprocessing stage when the user watchesthe movie, using the implicit feedback of interaction between the user and the movie to sort the interaction data of each user and the movie according to the timestamp, and obtaining a corresponding movie watching time sequence; and then encoding and representing the movie data,wherein the personalized recommendation model modeling comprises the embedded layer design, the one-dimensional convolutional network layer design, a self-attention mechanism, a classification output layer and the loss function design. According to the method, the one-dimensional convolutional neural network technologyand the self-attention mechanism are combined, so that the training efficiency is higher, and the number of parameters is relatively small.

The present invention is directed towards systems and methods for generating recommendations in response to one or more users based on user search queries. The method of the present invention comprises generating a recommendation model based on aggregate activity generated though use of a network resource. A user profile is generated based on an individual user's interaction with said network resource. A user query is received and the previously generated recommendation model in combination with the previously generated user profile are utilized to provide a recommendation relevant to the user search query and global statistics.

The invention relates to an individuation catering recommendation method and system based on multiple targets. The individuation catering recommendation method mainly includes the steps that essential information and behavior information (including diet records and browsing behaviors on webpages) of a user are collected and recorded into a database; a user model is built according to user information in the database, and individuation nutrition recipes are recommended; nutrition balanced diet recipes are generated through multi-target optimization catering models, the similarity between the recipes based on nutrient elements is calculated through a collaborative filtering recommendation model and an equivalence interchange model, various interchange recipe lists are generated from recipe bases, and the dietary structure is enriched; the nutrition balance of the recipes is detected, improvement measures are provided by comparing practical contents and recommended nutrient intakes of various nutrients in the generated recipes, and the designed recipes trend to be reasonable.

The invention discloses a sequence recommendation method based on a self-attention mechanism, and the method comprises the following steps: obtaining a historical behavior sequence of a target user, and dividing the historical behavior sequence into a long-term behavior sequence and a short-term behavior sequence; inputting the long-term behavior sequence and the short-term behavior sequence intoa recommendation model for recommendation learning to obtain a target recommendation object, wherein the recommendation model is a multi-layer self-attention network sequence recommendation model integrating long-term and short-term preferences of the user; and pushing the target recommendation object to the target user. According to the method, the long-term preference and the short-term demand of the target user can be learned, so that the target recommendation object better conforms to the time-based change preference of the target user, the recommendation is more accurate, and the user experience can be improved. The invention further discloses a sequence recommendation device and equipment based on the self-attention mechanism and a readable storage medium which have corresponding technical effects.

The invention discloses a network resource recommendation method and apparatus, an electronic device, a server and a storage medium, and belongs to the technical field of networks. By utilizing a resource commendation model obtained by training according to network resource information, function usage behavior information of a user and historical usage information of the user, the user to be subjected to recommendation is subjected to preference prediction and recommendation; and the sources of sample data collected during model training are more diversified, so that the resource recommendation model can describe a relationship between the user and preferred network resources from different perspectives, the preference degree of the user to the network resources is accurately represented,the accuracy of recommending the preferred network resources to the user is relatively high, and a powerful basis is provided for commercial operation and the like.

The invention discloses an insurance product recommend method comprising the following steps: obtaining omnibearing data information of a client if to recommend a target insurance product to the client; using a pre-determined insurance product recommend model to analyze the obtained client omnibearing data information, thus obtaining a preference probability value of the client on the target insurance product; determining to recommend the target insurance product to the client if obtained preference probability value is greater than a preset probability threshold; sending a recommend order aiming at the insurance product to a pre-determined terminal. The method can timely and accurately dig out the client true demands, thus improving the client experience effect, and improving the businesspersonnel business success rate.

Apparatus, methods and systems for sharing and distributing electronic content using a mobile device are presented. In more specific terms, a viral marketing-based, discounted referral and recommendation model by which users may forward electronic content to other users in disclosed. In one aspect, a method of distributing electronic media content over a network includes displaying a discounted offer on a first device that has been transmitted from another device for which the electronic media content has been purchased. A notification is then received, and upon receiving an acceptance from the first device, the electronic media content is transmitted to the device.

The invention discloses a federated learning-based recommendation model training method, a terminal and a storage medium. The method comprises the steps of obtaining a user historical behavior data set recorded by a client application in multiple preset types of application projects; extracting a single characteristic user vector of each client application based on each group of user historical behavior data set; extracting a project feature vector set and a project score set from a user historical behavior data set of the target application; combining each single feature user vector, the project feature vector set and the project score set to obtain a local training sample set; and participating in federated learning based on the local training sample set to obtain a recommendation modelof the target type application project. According to the method, model training is carried out under a federal framework to protect user privacy data, meanwhile, recommendation model training is carried out on the basis of multi-scene data, the recommendation model obtained through training can more accurately locate the preference characteristics of the user, and therefore the recommendation effect of the recommendation model is improved.

The invention relates to a code recommendation method based on a long short-term memory (LSTM) network. For the problem that low recommendation accuracy rates, low recommendation efficiency and the like are ubiquitous in existing code recommendation technologies, the method firstly extracts source code to form an API sequence, utilizing the long short-term memory network to build a code recommendation model to learn relationships between API calls, and then carries out code recommendation. A dropout technology is used to prevent model overfitting. At the same time, using a ReLu function to instead a traditional saturation function is provided, the gradient vanishing problem is solved, a model convergence speed is accelerated, model performance is improved, and advantages of the neural network are fully exerted. The technical scheme of the invention has the characteristics of simpleness and quickness, and can better improve an accuracy rate and recommendation efficiency of code recommendation.

The invention discloses a cloud computing based real-time mass user behavior analyzing method and system. User behaviors and context data are collected by a client end in real time and pre-processed and clustered on the basis of a MapReduce model; ontology data are subjected to reasoning, and latest interests of users are analyzed in real time; a track recurrence algorithm on the basis of a user behavior context is provided for track filling; interest similarity of users is computed by a cosine factor method, an interest similarity matrix is established; a Markov transfer matrix and a collaborative filtering based Markov recommendation model are established to realize effective and precise recommendation. The user behaviors and context information are subjected to model establishment via ontology, and semantic-level sharing and reusing of large-scale behavior information are achieved via an Hbase (hadoop database) based ontology memory mode. Technologies of cloud-computing, the ontology, reasoning and knowledge discovering are combined, and problems in instantaneity, efficiency, large-scale memorizing and intelligentization in mass user behavior analysis are solved.

The invention discloses a recurrent neural network-based commodity question answering expert recommendation method. The method comprises the following steps of: preprocessing data; screening candidateexperts for all the users corresponding to a question answering community; constructing an expert user file corresponding to each candidate expert; carrying out training on the basis of word vectorsof specific domain knowledges so as to obtain a word vector search table; carrying out feature representation learning on the basis of a recurrent neural network model and constructing a community question answering expert recommendation model; and determining an expert recommendation sequence corresponding to a new question in a to-be-processed question answering community. According to the method, grammar and semantic information of sentences can be effectively expressed, and high-level features of sentence level can be mined, so that manual intervention is decreased and automatic training and learning can be carried out.

The embodiment of the invention provides an information recommendation method and device, an information recommendation model training method and device and a storage medium. The information recommendation method comprises the steps of receiving a recommendation request of a client corresponding to a target user; extracting different types of user characteristics corresponding to the target user,and forming a user vector of the target user according to the combination of the different types of user characteristics; extracting different types of article characteristics of the article, and combining the different types of article characteristics to form an article vector of the article; according to the distance between the article vector and the user vector, determining an article satisfying a condition with the vector distance of the user vector; and sending corresponding recommendation information to the client based on the article meeting the condition. According to the embodiment of the invention, personalized recommendation can be comprehensively and accurately carried out.

The invention discloses a dynamic recommendation method based on training set optimization for a recommendation system, which specifically includes: (1) establishing a preliminary recommendation portion: generating an original recommendation model according to original user grading data; (2) performing AdaBoost trainings: utilizing the original recommendation model as a classifying and judging basis to classify the data and adjust learning times of samples by means of multiple iterative learning training data; (3) screening incorrect samples: data of selected difficult samples are removed as the incorrect samples after multiple AdaBoost trainings so as to construct a new training data collection; (4) reconstructing a recommendation model: combining training results to regenerate the recommendation model based on the new training data; and (5) generating recommendation results: utilizing the new recommendation model to generate the recommendation results. The method is capable of removing the data without referential meaning in recommended service by the aid of great relevance of original training set data in content, so that validity of the training data and precision of the final recommendation model are improved.

The invention discloses a search recommendation method which includes: receiving the search word of a user, and acquiring the identification of the user; acquiring of the category of the search word; acquiring the recommendation model corresponding to the user according to the identification of the user; inquiring the recommendation model to obtain a recommendation result according to the search word and the category of the search word; providing the recommendation result in a search result page. The method has the advantages that the recommended contents are related to user behavior features, that is to say, different cards are recommended to users with different behavior features, personalized recommendation of each user is achieved, the potential user demands of the users can be maximally stimulated, and user experience is increased. The invention further discloses a search recommendation device.

The embodiment of the invention provides a metapath context-based recommendation model training method and device. The method comprises the following steps: acquiring a plurality of user information pairs in a sample information platform; for each user information pair, obtaining a path instance corresponding to the user information pair, and generating user characterization information, sample object characterization information and meta-path context characterization information corresponding to the user information pair; for each user information pair, splicing the above three kinds of information to obtain training samples; training the preset neural network model by using a plurality of training samples as inputs until the value of the function calculated by the loss function is smaller than the preset threshold value, and the neural network model is trained. Compared with the prior art, the scheme provided by the embodiment of the invention has the advantages that the trained recommendation model based on the metapath context can learn more optimized features, and furthermore, the accuracy of the recommendation result obtained based on the model can be improved.

The invention discloses a university library-oriented books personalized recommendation method, and solves the problems of poor large-scale data storage and query, extendibility and recommendation effect in an existing books recommendation algorithm of a university library. According to the basic thought, the method comprises the following steps of firstly, building a graph model by taking readers, books and the like in the library as nodes; secondly, converting operation log files of the readers into a reader-books category preference matrix, calculating similarity between the readers by the reader-books category preference matrix and a reader personal information matrix, and establishing an associated graph spectrum by taking operations and mined information as edges; thirdly, by combining the associated graph spectrum with spectral clustering, proposing a new books personalized recommendation model, and performing calculation to obtain class cluster distribution about the readers; and finally, when books recommendation needs to be carried out, calculating a recommended books list according to a collaborative filtering algorithm in a class cluster corresponding to a reader.

The invention discloses a social-label-based method for optimizing a personalized recommendation system. In the method, social label similarity and score similarity are adopted and applied to calculation of a user-and-project oriented K-nearest neighbor model, and then a user and a project of a K-nearest neighbor are used for calculating a prediction score of the project by the user at the same time. Because the label similarity and the score similarity are adopted in the method at the same time, so that the K-nearest neighbor calculation of the user and the project is more accurate, the recommendation accuracy is obviously higher than that obtained by singly adopting the score similarity, and a cold-start problem based on a label similarity model can be solved. Therefore, a data sparsityproblem can be solved by using a user-and-project oriented recommendation model, and the recommendation accuracy is also obviously higher than that of a conventional user-oriented recommendation model and a project-oriented recommendation model.

The invention discloses a court similar case recommendation model based on word vectors and word frequencies, namely a TF-W2V similarity calculation model. The judgment documents are divided into fivecase types of criminal affairs, civil affairs, execution, compensation and administrative affairs, and in order to process, store and query the judgment documents, the model extracts the key information from the submitted judgment, and finds out the judgment with the highest similarity in the same type of judgment in the document data by adopting a Word2Vc + TF-IDF text similarity algorithm to give out the similarity and recommend the judgment. According to the method, based on a word frequency and word vector method, the keywords and the word meaning information of the texts are integrated,and the similarity of the two texts is accurately calculated. The method is applied to the court judgment for similarity calculation, and the experimental results prove that the method is simple to apply, has no requirement for a labeling training set, can be applied to the texts in different fields, consumes the moderate time in calculation, is more accurate in obtained result compared with a traditional method, is closer to the expert evaluation results, and can calculate the similarity of the court texts accurately and effectively.

The invention discloses an e-commerce website real-time recommending method under big data. The method includes the following steps: using user's implicit behavior log information on e-commerce websites to train an offline recommendation model; online acquiring user's implicit behavior log information, and using distributed-storage technology and distributed-streaming technology to rapidly process user's bulk implicit behavior log information; combining the well-trained offline recommending model and user's latest implicit behavior log information that undergo the distributed-streaming and providing the latest commodity recommendation list to the user. According to the invention, the method can real-time analyze user's behaviors under big data and provide real-time recommendation and feedback, and increases user's satisfaction and transaction transfer rate of e-commence websites.

The invention provides a training method and device of a recommendation model, and a recommendation method and device, and the training method of a recommendation model comprises the steps: obtainingthe user characteristics of at least two sample users and the attribute characteristics of at least two sample application programs; generating a positive sample which is clicked and purchased by thesample user on the exposed sample application program and a negative sample which is clicked but not purchased or unclicked by the sample user on the exposed sample application program based on the user characteristics and the attribute characteristics; and training a recommendation model on the basis of a sample set comprising at least one positive sample and at least one negative sample to obtain the recommendation model, and outputting an exposure conversion rate obtained by each sample user on the basis of the click rate and the purchase rate of each exposed sample application program by the recommendation model.

The invention discloses a short video recommendation method, a device and a readable medium, belonging to the technical field of video recommendation. In the method and the device provided by the invention, after receiving a short video pull request, a short video sequence composed of a short video list viewed by a user history and a short video list not viewed is obtained. Determining a sequencevector for characterizing a short video feature in the short video sequence according to the short video sequence and the trained short video feature matrix for characterizing all the short video features; according to the sequence vector and the training short video recommendation model, the probability of each short video in the unwatched short video list is determined. According to the probability of each short video, the short video of interest is recommended to the user. By implementing the method, the short video of interest to the user is determined from the mass short video, and the short video is recommended to the user, which not only meets the viewing requirements of the user for the short video, but also improves the utilization rate of the short video application program by the user.

The invention discloses a mentor recommendation system, comprising a data source module, a teacher and student relation extraction module, an expert degree characteristic analysis module, an expert degree calculation module and a sequencing module; the invention also discloses a mentor recommendation method, comprising the following steps of: S1, mining teachers and students relations by means of a model of a probability factor figure with time constraint; S2, establishing a recommendation model based on expert degree; S3, establishing a personalized recommendation model; and S4, outputting recommendation results according to the recommendation model based on expert degree and personalization. The mentor recommendation system and the method provided by the invention have the advantages that the NDCG (Normalized Discounted Cumulative Gain) index of a recommendation model is improved by 5 to 10%, a personalized recommendation model is established by means of a probability model, and a personalized mentor recommendation purpose is achieved.

The invention belongs to the technical field of artificial intelligence, and discloses a diagnosis and treatment scheme recommendation method. The method comprises the following steps: constructing amedical record database which comprises a first medical record sample and a label; constructing a similar medical record model and a recommendation model; training the similar medical record model andthe recommendation model according to data in the medical record database; obtaining a second medical record sample; inputting the second medical record sample into the trained similar medical recordmodel, outputting one or more similar medical records similar to the second medical record sample, inputting the obtained one or more similar medical records into the trained recommendation model, and outputting a diagnosis and treatment path corresponding to the second medical record sample. In this way, an integral diagnosis and treatment scheme is obtained by recommending the current illness state of the patient. The dependence on the medical expert level is reduced. The influence of human factors is reduced. The situations of delayed treatment and the like are avoided. The invention further discloses an electronic device and a computer readable storage medium.

The invention discloses a failure record (FR) method for optimizing personalized recommendation results, which improves the personalized recommendation quality and precision by using social tag network filter and recommendation deviation removal. The social tag network filter method comprises the following steps of: establishing a project social network K neighbor by using a social tag network model, and constructing a social tag filter set during recommending in a recommendation model based on the project social network K neighbor, wherein the social tag filter set is used for filtering recommended projects with low social tag relevance in the user scored projects in the recommendation results of a project-orientated K neighbor model so as to combine information in user-project scoring data and social tag network data to recommend. The recommendation deviation removal comprises the following steps of: based on prediction values of the project-orientated K neighbor model on the known user-project scoring data and a turn score of the user, estimating the recommendation deviation by using a linear model; and when the recommendation is performed by using the recommendation model, removing the corresponding recommendation deviation estimation values from the scoring prediction values so as to optimize the recommendation results.

The invention discloses a search recommendation method. The search recommendation method comprises the following steps: receiving a picture input by a user; acquiring user identification; analyzing the picture to acquire search intention of the user; acquiring a corresponding recommendation model of the user according to the user identification; inquiring the recommendation model according to the search intention to acquire a recommendation result; providing the recommendation result in a search result web page. The method realizes the function of searching solid information and acquiring required information, enlarges search requirement range of the user, promotes search experience of the user, improves expansibility of a search engine, and adds the search function. The invention further discloses a search recommendation device.

Enabling intelligent navigation is described, including: performing analysis of historical user activity data with respect to a query term to generate reference data associated with the query term; selecting a navigation recommendation model for the query term based at least in part on the reference data; using the reference data and the selected navigation recommendation model to determine a set of recommendation data associated with the query term, wherein the set of recommendation data includes at least a portion of a category system to be displayed in response to a subsequently received query including the query term.

The invention discloses a method and a device for processing the reserved registration information. According to one embodiment of the invention, the method comprises the steps of receiving the reserved registration information sent from a client, wherein the reserved registration information contains the disease condition description information and the information of an appointed clinic registration department; importing the disease condition description information contained in the reserved registration information into a pre-trained department recommendation model to obtain the information of a recommended clinic registration department through the matching process, wherein the department recommendation model is used for representing the corresponding relationship between the disease condition description information and the information of clinic registration departments; matching the information of the appointed clinic registration department with the information of the recommended clinic registration department to obtain the information of an audit result; and sending the information of an appointment registration result to the client according to the information of the audit result. According to the invention, the information of the recommended clinic registration department is obtained through the matching process of the reserved registration information by means of the department recommendation model. After that, the information of the audit result can be obtained through matching the information of the appointed clinic registration department with the information of the recommended clinic registration department. In this way, the appointed registration efficiency and the appointed registration accuracy are improved.

The invention provides a website recommendation result displaying method and a device and a terminal with the device. The website recommendation result displaying method comprises the steps of providing each recommendation result related to items according to the items selected by an user; analyzing a website recommendation model and extracting multiple attribute values of each recommendation result according to the recommendation model; establishing a multidimensional coordinate system; mapping multiple attribute values of each recommendation result to the multidimensional coordinate system; drawing a multidimensional model of each recommendation result according to the position of each recommendation result in the multidimensional coordinate system; displaying the multidimensional model of each recommendation result in the multidimensional coordinate system. The embodiment provided by the invention has the advantages of richly expressive, three-dimensional, visual, vivid and clear display. The invention also provides the website recommendation result displaying device and the terminal with the device.

The invention discloses a Top-N movie recommendation method for performing weighted fusion on selected local models based on random anchor points. The method comprises the steps of obtaining eigenvectors of users and movies at the semantic level through an LDA topic model and a GBDT by utilizing movie text data; based on the eigenvectors, calculating Gaussian kernel similarity between the users and the movies; randomly selecting a plurality of (the users and the movies) anchor point pairs, and reconstructing a local training matrix for each anchor point pair in combination with the Gaussian kernel similarity between the users and the movies; performing training for each local training matrix by utilizing an SLIM as a basic recommendation model to obtain a corresponding local recommendationmodel; and finally, generating a final fusion recommendation model through weighted fusion among the local recommendation models. According to the recommendation method, the stability of the models is also maintained in a data sparseness scene, and the problem that a traditional single recommendation model is very easy to over-fit in the data sparseness scene can be effectively solved.