30results about How to "Intelligent Recommendation" patented technology

The invention discloses a method and device for recommending software upgrading. The method can comprise the steps that the use parameters and / or upgrading quantity white list of software in a user terminal are / is acquired; the use parameters comprise software use frequency and / or software upgrading frequency; a list of the software to be upgraded is generated according to the use parameters and / or the upgrading quantity white list; the list of the software to be upgraded is matched with a local software library to generate the soft upgrading recommending list of the user terminal. The upgrading intention strength of the user terminal on different software is judged through the local software use frequency of the user terminal and the frequency of upgrading the software through a software management client side, and therefore the software capable of being upgraded is actively recommended to the user terminal more precisely and more intelligently.

The present invention relates to the technical field of recommendation systems, in particular to a personalized intelligent clothing collocation recommendation method combined with knowledge graphs, which acquires clothing product information on an e-commerce platform and user clothing interaction information and constructs a knowledge graph; models the knowledge graph, and Each entity and relationship in the knowledge map is learned to obtain entity vectors and relationship vectors; the entity vector and relationship vector are combined with a hybrid recommendation system to predict the rating of single-item clothing and obtain clothing single-item recommendation results; calculate pairwise The matching index of clothing items, automatically match two clothing items with high matching index and predict the score, and make the final TOP-N suit recommendation result according to the score. This method makes online clothing suit recommendation more intelligent.

The invention relates to the technical field of intelligent decision-making, and provides a parking space navigation method, an electronic device and a storage medium. This method performs grayscale processing on the collected image of each parking space, extracts the contour image of each parking space, and extracts the Haar feature of each contour image to train a preset model to obtain a vehicle recognition classifier. Afterwards, the Haar feature corresponding to the real-time image of the parking space captured at a preset time point is input into the vehicle recognition classifier to identify the empty parking space in the parking lot, and generate a parking space according to the empty parking space and the current position of the vehicle. A route is recommended to the client of the user. The invention can make the parking lot users know the vacant parking spaces in the parking lot at a glance, find the parking spaces quickly, recommend the best parking route to the users more intelligently and accurately, and avoid parking difficulties and waste of time for the users due to unfamiliar parking conditions. situation, while improving the efficiency of parking operations in the parking lot.

The present invention provides a method for recommending clothing and a mobile terminal. The method includes: acquiring user's schedule information; target dress style label; determine N candidate dress images marked with the target dress style tag from the target dress library, where N is a positive integer greater than or equal to 2; A target dress image matching the pre-acquired dress reference factors; displaying the target dress image. The dressing recommendation method provided by the present invention can recommend dressing images to the user very intelligently, so that the user can solve the dressing problem conveniently and quickly.

The invention provides an intelligent recommendation method for review experts oriented to scientific and technological projects. The present invention comprises the following steps: 1). Segment the main text of pending scientific and technological projects and expert information into substring sequences and perform word segmentation by ICTCLAS of the Chinese Academy of Sciences, and perform stop word filtering on the word segmentation results to obtain a word set; 2). Build the words of project information Network, extracting feature words based on statistical features and aggregation features; expert information is relatively streamlined, and the word set obtained in step 1 is directly used as feature words; 3). Construct a knowledge representation model based on the fields and weights of feature words; establish relevant information indexes; 4 ). The group recommendation experts perform inter-field and inter-item feature merging operations on the knowledge representation model; 5). Based on the similarity between the semantic calculation experts and the technology projects (groups) to be reviewed, a threshold value is set to truncate to generate the final recommendation expert list. The implementation of the above method can greatly alleviate the problems of heavy workload in recommendation and lack of scientific review decision-making.

This application relates to satellite network security analysis methods, devices, systems and storage media. The method includes: obtaining satellite network equipment security data and completing the model construction of satellite network security ontology; according to the four-level model of satellite network security equipment, performing equipment security data After unified processing, the satellite network security hierarchical model and the satellite network security knowledge model are respectively formed; the entity name, attribute and mutual relationship are abstracted from the satellite network security hierarchical model and the satellite network security knowledge model, and are formally expressed by setting description language as The dynamic triples are used to obtain the dynamic graph model; then knowledge reasoning and graph dynamics analysis are carried out based on OWL+SWRL language, and the auxiliary construction of the satellite network equipment security knowledge map is completed; the security problem query statement is obtained and the knowledge reasoning and analysis are carried out through the constructed knowledge map. Query and analyze to obtain information on security weaknesses of the satellite network. The analysis efficiency of weak links in satellite network security has been greatly improved.

The invention provides a method and device for providing search suggestion words based on activity state. The method is applied to a mobile terminal, and includes: determining the user's frequently used places according to historical activity status information obtained based on the user's historical activity data recorded by the mobile terminal; screening out at least one of the frequently used places from the Frequently used places are used as search suggestion words; the search suggestion words are pushed to the specified type of application, so that when the search box in the specified type of application is triggered, the search suggestion words are displayed in the corresponding search box Suggested positions for search terms. This technical solution enables the mobile terminal to provide users with location-based search suggestions, and the filtered search suggestions are more in line with the current activity status of the user, without the need for the user to manually input search keywords, thereby enabling the mobile terminal to recommend search words More intelligent and humanized.

The embodiment of the present invention relates to the field of smart devices, and discloses a learning content recommendation method and a tutoring device. The method includes: collecting a large number of voice data samples and regional information samples of users in different regions; using the voice data samples as training data and using the regional information samples as training results to train a regional recognition model; command to obtain the voice data of the target user; input the voice data into the area recognition model to obtain the area information of the target user; recommend learning content corresponding to the area information for the target user. By implementing the embodiment of the present invention, different learning content can be recommended to users in different regions based on speech recognition, which can make the learning content recommendation method simpler and more intelligent.

The embodiment of the invention provides a data analysis method based on a housing monitoring platform, and the method comprises the steps: selecting a to-be-analyzed region and a data dimension related to a user in the region from the housing monitoring platform, and obtaining a to-be-analyzed data matrix; extracting an important dimension from the data dimension according to the feature vector in the data matrix; and screening out users whose correlation with the important dimension is greater than a set threshold from the housing monitoring platform to form a target user. Meanwhile, the invention further provides a data analysis device based on the housing monitoring platform. According to the technical scheme, statistical data is more comprehensive, customers can be intelligently recommended, marketing efficiency and transaction rate are improved, and data cost of companies is saved.

The invention discloses a real-time goods recommendation method based on customer shopping intention exploration. The method comprises the following steps: 1, constructing a tree-shaped customer shopping intention model; 2, detecting user behaviors in real time and identifying user intention; 3, detecting the change of the user intention in real time; 4, tracking the change track of the user intention and predicting the deep intention of the user: predicting the deep intention of the user based on the discrete intention of the user; and 5, determining a goods recommendation strategy: determining the goods recommendation strategy according to the shopping intention of the customer. The method disclosed by the invention automatically identifies the shopping intention of the customer, performs precise individualized recommendation, and improves the customer satisfaction; and moreover, the method automatically detects the change of the shopping intention of the customer and adjusts the goods recommendation strategy. The method adapts to the changing interest of the customer in real time, deeply explores the shopping intention of the customer based on the user intention model and provides professional recommendation service to the user.

An embodiment of the present application provides an image processing method, device, storage medium, and electronic equipment. The image processing method includes: determining the augmented reality model that is used most frequently by users of each personality type according to multiple historical images; an image of the current user; determining the personality type of the current user according to the image; adding the augmented reality model that is most frequently used by users of the personality type to the image. In the image processing method, the electronic device can determine the personality type of the current user according to the photos taken by the current user, automatically obtain the augmented reality model that the user of the personality type uses the most times, and add the augmented reality model to the photographed The final image can be generated on the image, and the augmented reality model can be intelligently and accurately recommended to the user, without the need for the user to manually select the augmented reality model, thereby improving the efficiency of taking photos.

The present application discloses a method for naming a folder, including: a first terminal device receives instruction information for creating or updating a folder, the folder includes N application programs APP, and N is a positive integer. The first terminal device determines the folder name with the largest weight value from the candidate folder names according to the weight of the folder name provided by the cloud server and the position weights of the P APPs. The candidate folder name is provided by the cloud server, and the cloud server provides The weight of the folder name is the weight of the folder name corresponding to M APPs determined by the cloud server based on the data provided by the target terminal device. P APPs are the APPs shared by M APPs and N APPs, M is a positive integer, and P is positive integer. The first terminal device names the folder according to the folder name with the largest weight value. Through the technical solution provided by the embodiment of the present application, the accuracy of folder naming is improved, and user experience is improved.

The invention provides an intelligent music recommendation method, system and device. The method comprises the following steps: collecting sounds in the surrounding environment during the music playing; analyzing the collected sounds and judging whether a user is humming the currently played music or not; adjusting a recommendation weight for the currently played music according to the judging result. The intelligent music recommendation method, system and device can be used for judging whether the user likes the currently played music or not through judging whether the user is humming the currently played music or not, and then adjusting the recommendation weight for the currently played music; compared with the manner purely depending on labels, content histories, machine learning or listening histories, the intelligent music recommendation method, system and device can be used for recommending music according with the user preferences for the users more intelligently according to the response, for the currently played music, of the users, so as to improve the music recommendation accuracy.