52results about How to "Realize automatic recommendation" patented technology

The invention discloses a data access method and device, computer equipment and a readable storage medium. At least one first field to be mapped corresponding to a first external dataset to be accessed is acquired; based on a term vector corresponding to each participle in the first fields to be mapped, a dataset vector to be matched corresponding to the first external dataset is obtained; based on the dataset vector to be matched and a dataset sorting model trained in advance, a first standard dataset matched with the first external dataset is determined; based on the similarity between standard fields in the first standard dataset and the first fields to be mapped, first field mapping relationships between the first fields to be mapped and the standard fields are built; based on the first field mapping relationships, the first external dataset is accessed into the first standard dataset. By means of the data access method and device, the computer equipment and the readable storage medium, automatic recommendation of data access and automatic mapping of field levels can be achieved.

The invention discloses a method and a system for automatically recommending a learning partner in a distance education social networks service (ESNS). The method comprises the following steps of: 1, automatically receiving learning partner recommendation application of a user side; 2, responding to the learning partner recommendation application, automatically acquiring user information, and dividing the user information into user individual information, academic information and learning partner information; 3, acquiring the tightness of a candidate learning partner and the user according to the user individual information, the academic information and the learning partner information; and 4, acquiring the candidate learning partner according to the tightness of the candidate learning partner and the user and automatically recommending and displaying the candidate learning partner to the user. By the method and the system, the learning partner in the ESNS can be automatically recommended.

The invention provides a recommendation method based on a user cluster. In order to conduct effective recommendation on users reasonably according to user interest, the interest degree is acquired according to total browsing frequency, browsing time, total browsing time, effectively browsing efficiency and effective browsing time of subject labels of the users to form a user interest characteristic vector; core users are screened according to the user interest characteristic vector to form a core user set, and a K-means cluster algorithm is utilized to cluster all the users; after all user clusters are obtained, the class interest vector of each user cluster on each subject is calculated; the interest value and the class interest vector are compared to conduct recommendation. The CCVR method is better in recommendation effect compared with other recommendation methods, and the method has good accuracy.

The embodiment of the invention discloses an interior-design method and system based on user interest preference. The method includes: carrying out editing design of 3D design schemes by a designer for interior design, and uploading the 3D design schemes and interior-design 2D effect drawings, which are associated therewith, to a server; setting initial preference by a user, automatically recommending interior-design 2D effect drawings to a user mobile-end by the server according to the user preference; uploading an interest picture by the user to acquire 2D effect drawings associated with theinterest image and corresponding 3D design schemes; clicking an interior-design 2D effect drawing by the user to automatically acquire a presentation effect of a 3D design scheme; and communicating with the designer by the user to communicate the 3D design scheme online. According to the method, user attributes, browsed-content attributes and operation behaviors of the user in the system are simultaneously investigated, the user interest preference is effectively understood, automatic recommendation of the design schemes is realized, the user and the designer can simultaneously and jointly participate in scheme design online, communication costs are reduced, and efficiency of communication between the designer and the customer is improved.

The invention discloses a node cluster implementation method, device and system in a peer-to-peer network, which are used for realizing automatic promotion of a main node during peer-to-peer network cluster deployment, reducing the complexity of the cluster deployment and improving the efficiency of the cluster deployment. The method comprises the following steps: acquiring state information of each node with same cluster identifier by any node in a cluster, wherein the state information comprises the identifier and a performance parameter of the node; determining a main node according to the acquired performance parameter of each node and the performance parameter of the node.

The invention applies to the technical field of computers, and provides a question recommendation method and a question recommendation device of a mobile terminal. The question recommendation method includes: when detecting a question recommendation request of a user, obtaining knowledge points input by the user, using a preset matching algorithm for matching questions related with the knowledge points in a knowledge point question bank constructed in advance according to the obtained knowledge points, and outputting the matched questions related with the knowledge points input by the user. Accordingly, automatic recommendation of the questions is achieved, time for selecting the questions during the learning process of the user is saved, and experience of using the mobile terminal for learning is improved for the user.

The invention discloses a data quality detection method and system based on a multi-dimensional label. Based on the known type data items and the detection rule base, a multi-dimensional label analysis algorithm is used for marking corresponding dimension labels on the known type data items, and the dimension labels are used for dynamically adjusting the quality detection process of the known typedata items; a quality detection engine is recommended for the unknown type data source by using a rule similarity evaluation algorithm based on the unknown type data item and in combination with a detection rule base, and a result of the quality detection engine is verified to obtain an effective quality detection rule set; and the quality detection process and the effective quality detection rule set of the known type of data items are stored, and the multi-dimensional label rule base is updated. According to the scheme, through a multi-dimensional label algorithm and a rule similarity evaluation algorithm, the problems of poor accuracy, weak timeliness and the like caused by a fixed detection rule template are solved, rapid and accurate detection of data quality is realized, a detectionresult is fed back in time, and the quality of a data source is improved.

The invention relates to an identification binding method and device, a terminal, a server and a storage medium, and belongs to the technical field of Internet. The method comprises the steps that based on a logged-in first user identification, an application interface of a game client side is displayed and comprises a face-to-face group building option; when a triggering operation on the face-to-face group building option is detected, an identification input field is displayed in the application interface, and a recognition identification input in the identification input field is acquired; group building requests are sent to the server and carry the recognition identification and the first user identification, the server is used for binding a game group identification generated accordingto the recognition identification and the first user identification carried in the group building requests when receiving the group building requests carrying the same recognition identification andthe corresponding first user identification. It is unnecessary to build an association relationship among user identifications, a restriction that the association relationship has to be built among the user identifications is got rid of, the flexibility is improved, and an application range is enlarged.

The invention provides a platform for application distribution. The platform for application distribution comprises an application information interaction module, an application information analysis module and an application information distribution module, wherein the application information interaction module is used for allowing users to exchange information about various applications, the application information analysis module is used for analyzing information exchanged by the users, and the application information distribution module is used for issuing the analysis result of the application information distribution module onto the application information interaction module. The invention further provides a method and system for application distribution. With the help of the platform, method and system for application distribution, the users can exchange application information on the platform and the system, automatic recommendation of application information and applications can be achieved, and wide and efficient distribution of the applications is achieved.

The invention provides an outdoor advertisement delivery resource recommendation method and device and computer equipment. The method comprises the steps: determining candidate delivery resources according to the historical delivery information of a target customer and a preset rule; acquiring the historical delivery records of all customers matched with the attribute information of the target customer, and acquiring priorities of each type of delivery resources in the historical delivery records of all the customers; and determining a target delivery resource from the candidate delivery resources according to the priorities of each type of delivery resources, and recommending the target delivery resource to the target customer. By means of the method, automatic recommendation of the outdoor advertisement delivery resources can be achieved, the tedious process that a customer searches for and inquires the delivery resources many times is avoided, inquiring time is saved, delivery resource purchasing efficiency and delivery efficiency are improved, and the technical problems that in the prior art, time consumed for purchasing the outdoor advertisement delivery resources is long, andpurchasing efficiency is low are solved.

The invention relates to the field of voice recognition, and provides a product recommending method based on voice analysis. The method comprises the following steps of playing preset initial voice associated with a product inquiry request when receiving the product inquiry request; collecting first voice information replied by a customer; recognizing the first voice information to obtain the corresponding first tendency information; inquiring a preset verbal trick forest; obtaining a target node matched with the first tendency information in the preset verbal trick forest; generating and broadcasting first recommended verbal trick corresponding to the first tendency information according to the target node; receiving second voice information relied by the customer on the basis of the first recommended verbal trick; recognizing the second voice information to obtain the corresponding second tendency information; when the second tendency information is negative, performing backtrackingreaching on the target node according to the second tendency information; obtaining and broadcasting the second recommended verbal trick. The invention also discloses a product recommending device based on voice analysis, product recommending equipment based on voice analysis, and a storage medium. The product recommending accuracy is improved.

The invention relates to the technical field of intelligent air conditioners, and discloses a method, device and equipment for air conditioner working mode recommendation. The method comprises the steps that user use data information corresponding to each air conditioner of a set model within a set time period is obtained; index data information matched with the energy consumption of the air conditioner of the set model is extracted from the user use data information; an energy-saving comfortable working mode corresponding to the set model is determined through a clustering learning algorithmaccording to each piece of energy consumption information and the corresponding index data information; and the energy-saving comfortable working mode is recommended to the air conditioner of the setmodel. Thus, automatic recommendation of the energy-saving comfortable working mode is achieved, the dual requirements for energy saving and comfort of the user are met, and the intelligence of the air conditioner and the user experience are improved.

The invention provides an electronic fence recommendation method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring coordinate data of a plurality of historical track stop points of a vehicle; dividing a map of the area into a plurality of blocks according to the coordinate data of the plurality of historical track stop points of the vehicle; determining a central point according to dispersion information of each historical track stay point in each block; an electronic fence is generated according to the coordinate data of the center point and the coordinate data of each effective stay point in a target block where the center point is located, and the electronic fence is used for identifying a sub-area in the area; and pushing the electronic fence to a user. According to the method, the electronic fence can be automatically recommended, the working efficiency and the accuracy of electronic fence recommendation can be improved, in addition, the correct electronic fence area can be objectively recommended by obtaining the historical track stop point of the vehicle as the data support, and the manual error is reduced.

The invention discloses a method for automatically recommending an internet surfing type through a gateway and the gateway. The method includes the steps of judging the recommendation of a PPPoE access mode, judging the recommendation of a DHCP access mode and recommending a fixed IP access mode. The gateway sends a PADI packet to a PPPoE server, whether the gateway receives a PADO packet sent back by the PPoE server within time-out duration and a preset number of times of sending or not is judged, if yes, the recommendation of the PPPoE access mode is output, and if not, the judgment of the DHCP access mode is executed. The gateway sends a discover packet to a DHCP server, whether the gateway receives an offer packet sent back by the DHCP server within time-out duration and a preset number of times of sending or not is judged, if yes, the recommendation of the DHCP access mode is output, and if not, the recommendation of the fixed IP is output. The internet surfing type is automatically recommended through the mode that the gateway sends the data packets, and therefore the inconvenience problem caused when a user configures a router is solved.

The invention discloses a cloud resource recommendation method and device, an electronic terminal and a storage medium. The cloud resource recommendation method comprises the steps of receiving a cloud resource recommendation instruction, analyzing a target demand parameter carried in the cloud resource recommendation instruction, determining the type of the target demand parameter, and obtainingan association relationship of demand-cloud resources corresponding to the type of the target demand parameter, wherein the association relationship of the demand-cloud resources is determined based on historical recommendation information of the cloud resources; and determining target cloud resource information matched with the target demand parameter according to the association relationship ofthe demand-cloud resources, and recommending the target cloud resource information. The target cloud resource information does not need to be determined manually, and automatic recommendation of the cloud resources is realized, so that low-resource-consumption and high-efficiency cloud resource recommendation is realized. Meanwhile, by analyzing the demand parameters of the user, the cloud demandof the user is fully understood, and high-accuracy cloud resource recommendation is realized.

The invention discloses an item handling intelligent recommendation method based on a data graph and a service graph. The method comprises the steps: S1, constructing a data graph body; S2, constructing a data graph; and S3, constructing a service map. After the S3, item recommendation can be further carried out based on the data graph and the service graph. After the S3, the method can further comprise: S4, updating the service graph data; and S5, recommending a potential item. By means of the item handling intelligent recommendation method based on the data graph and the service graph, iteminformation can be automatically recommended, the interaction process of a user and a system is reduced, and meanwhile the item recommendation accuracy is improved.

The invention provides a data classification management recommendation method and device. The method comprises: matching information identification of to-be-managed data information stored by a user with information in a constructed social information base, and if the to-be-managed data information is matched with the information in the social information base, recommending the data classificationmanagement information in the social information base to the user. The data is classified and managed according to the existing data classification management information in the social contact information base, automatic recommendation of data classification management and scientific classification management of the data are achieved, and the efficiency of data classification management is improved.

The invention discloses a program recommendation method, a program recommendation system and a computer readable storage medium, and the method comprises the following steps: when that a user watchesa program is sensed, acquiring a face image of a current user; performing feature extraction for the face image of the current user; matching the extracted face feature of the current user with the face feature prestored in a cloud database, wherein the face features of different users and corresponding program parameters are stored in the cloud database; if the corresponding face feature is matched in the cloud database, acquiring the program parameter corresponding to the face feature of the current user from the cloud database; and recommending a corresponding program to the current user according to the program parameter. The method and the system perform program recommendation for the user in combination with the face feature of the current user and the user face features and the program parameters stored in the cloud database, not only can effectively protect privacy information of the user in program recommendation, but also can efficiently and accurately realize automatic recommendation of programs, thereby improve user experience.

The invention relates to the field of artificial intelligence, and provides a course recommendation method and device based on unsupervised learning, equipment and a medium, which can adopt a cycle-GAN model to extend training data, effectively solve the problems of poor model training effect and poor precision caused by insufficient samples, and meanwhile, improve the accuracy of the training data due to the characteristics of the cycle-GAN model. In addition, the samples generated after expansion have high availability, and the recommended courses are more reasonable due to the fact that prediction results of different dimensions are integrated, so that the courses are automatically recommended. The invention also relates to a blockchain technology, and the involved model can be stored in the blockchain.

The invention relates to the field of artificial intelligence, and provides a course recommendation method and device based on root cause analysis, equipment and a medium, which can identify a label of each course in course data, construct a training sample training prediction model according to the label of each course and the course data, and according to the prediction model, an improved ID3 algorithm is adopted to calculate the information entropy of each label, the information gain of each label is calculated according to the information entropy of each label, a course recommendation list is generated according to the information gain of each label, and the ID3 algorithm is combined to perform root cause analysis on the influence of training courses on performance, so that interpretability and accuracy of an analysis result are ensured, and the training efficiency is improved. Training courses having great influence on performance are analyzed in an auxiliary manner, and then automatic recommendation of the courses is realized in combination with an artificial intelligence means, so that continuous tracking, performance improvement and retention are carried out, and training really assists team improvement and individual development. In addition, the invention also relates to a block chain technology, and the prediction model can be stored in a block chain node.

The invention relates to the technical field of big data and e-commerce, and relates to a commodity identification method based on big data and e-commerce and an e-commerce platform. The method comprises the steps of obtaining a plurality of to-be-identified commodities from a commodity information base; obtaining a first commodity information parameter and a second commodity information parameterof each to-be-identified commodity in the plurality of to-be-identified commodities, with the first commodity information parameter being used for representing a commodity selling brief introductionof each to-be-identified commodity, and the second commodity information parameter being used for representing a commodity pricing strategy of each to-be-identified commodity; identifying a target commodity to be sold in the plurality of to-be-identified commodities according to the first commodity information parameter and the second commodity information parameter; automatic recommendation of the target to-be-sold commodity can be realized so as to meet the requirement of the user for personalized purchase of the commodities and the efficiency of obtaining the required commodity by the usercan be enhanced.

The invention relates to the technical field of intelligent equipment, and discloses a clothes fitting recommendation method, device and equipment. The method comprises the following steps: acquiringa to-be-dressed picture containing one or more to-be-dressed clothes; inputting the to-be-dressed picture into a configured multi-label single model to obtain a to-be-dressed clothes attribute featurevector; determining at least one to-be-recommended picture as a first to-be-recommended picture according to the acquired recommended clothing attribute feature vector of each to-be-recommended picture and the to-be-dressed clothing attribute feature vector, and recommending the first to-be-recommended picture. In this way, automatic clothes fitting recommendation is achieved, and the intelligence of equipment related to clothes is improved.

The invention provides a method and a device for recommending service. The method comprises the following steps: obtaining an identifier of a user; sending a first request to a second server, the first request being used for indicating the second server to obtain a service recommendation file stored in association with the identity file after determining that the identity file including the identifier is stored; wherein the service recommendation file is generated by the first server according to the service use data of the user and then sent to the second server, and the service use data is obtained by a deep packet inspection (DPI) server according to a service feature library and each first internet surfing data of the user within a first preset duration; and obtaining the service recommendation file from the second server, and sending service recommendation information to target equipment, the service recommendation information comprising a to-be-recommended service included in theservice recommendation file. According to the invention, the service can be accurately recommended to the user, and the service recommendation process can be automatically triggered when the user needs to do the service.

The invention belongs to the technical field of interior design. An interior design method and system based on user interest preference are disclosed. The system comprises a user mobile terminal, a server, a designer mobile terminal, a cloud database, a picture upload module, a personal center module, a 3D solution experience module, a management center module, an algorithm processor, a data processor, a designer personal center module, a 3D object editor, and a 3D solution generator. The mobile client is wirelessly connected to the server, the designer mobile end is wirelessly connected to the server, and the cloud database is wirelessly connected to the server. The invention inspects the user attribute, the attribute of the browsed content and the operation behavior of the user in the system, effectively understands the interest preference of the user, and realizes the automatic recommendation of the design scheme, and the user and the designer can participate in the scheme design simultaneously online, reduces the communication cost, and improves the communication efficiency between the designer and the customer.

Provided are an information recommending method, an electronic terminal, and a computer readable storage medium message. The information recommending method comprises: identifying a current motion mode of a user; based on feature points of reference information, searching target information matched with the feature points, wherein the reference information is arranged in advance and is associatedwith the motion mode which is identified currently; and outputting the target information, to recommend the target information to the user. The scheme provided by the application can realize information automatic recommendation based on the motion mode.

The invention discloses a recommendation method and device of ceramic design. The method includes the steps of obtaining a request for the ceramic design, wherein the request includes a design demand; extracting the design demand from the request; generating recommendation information on the basis of the design demand and a preset neural network of the ceramic design, wherein the recommendation information comprises a ceramic design scheme corresponding to the design demand. Through the recommendation method and device of the ceramic design, automatic recommendation is achieved, efficiency is improved, and moreover, since the recommendation information is generated in the neural network of the ceramic design on the basis of the design command, the efficiency is improved, and meanwhile, the recommendation effect is ensured, labor of a user is decreased, and the usage experience of the user is ensured.

The invention discloses a related patch recommendation method based on heterogeneous data. The method comprises the following steps: crawling multivariate heterogeneous data of code review; cleaning data, splicing multivariate heterogeneous data features into patch feature vectors; pairing patch pairs, taking the correlation with the prediction patch as a positive sample; taking a negative samplewithout correlation with the prediction patch; marking binary classification labels for the positive sample and the negative sample; dividing a training set and a verification set, using the trainingset to train a logistic regression model, a random forest model and a LightGBM model respectively to obtain a corresponding probability and a prediction label, calculating a corresponding accuracy rate according to the prediction label, and finally constructing a prediction score according to weighted summation of the fusion weight and the corresponding probability to obtain an optimal predictionscore. According to the method, correlation evaluation is carried out on the data submitted to the code review system through machine learning to obtain optimal recommendation, recommendation reliability and stability are improved, and more labor cost is saved.

The embodiment of the invention discloses an AI algorithm fusion recommendation method and system, and the method comprises the steps: obtaining comment data of a historical user in advance accordingto a store, carrying out word segmentation of the comment data, and obtaining a text after word segmentation; constructing a prediction model based on deep learning on the basis of texts and categories after word segmentation; according to the label and the historical record of the target user, adopting the prediction model to score the photograph probability of the target user for each type of store; and performing store recommendation based on the score. By applying the embodiment of the invention, automatic recommendation of the store can be realized, and the user experience is improved.

The invention relates to a configuration parameter recommendation method and system, an instrument and a storage medium. The method comprises the following steps of acquiring instrument description information of a target instrument to be configured; constructing a knowledge graph of the target instrument according to the instrument description information; based on the knowledge graph of the target instrument and a stored historical instrument knowledge graph, determining a historical instrument matched with the target instrument; enabling the historical instrument and the target instrument to be the same type of instruments; and recommending configuration parameters corresponding to the historical instrument as configuration parameters of the target instrument. By adopting the method, the technical threshold and the debugging workload of technicians can be reduced.