1411 results about "User group" patented technology

The invention belongs to the technical field of artificial intelligence, and discloses a government information consultation system based on a large language model, which comprises a user interaction module, a large language model core engine, a knowledge base integration module, a multi-level authority management module, a feedback optimization mechanism and a risk control module. The comprehensive intelligent government affair service system is constructed through the six core modules, remarkable advantages are shown in government affair service digital transformation, the system innovatively adopts multi-mode interactive design, multiple input and output modes of voice, text and images are supported, an intelligent authority management mechanism is matched, and the intelligent authority management mechanism is matched with the intelligent authority management mechanism. According to the technical scheme, the accessibility and convenience of government affair services are greatly improved, precise services for different user groups are achieved, it is guaranteed that sensitive data are safe and controllable while wide spreading of government affair information is guaranteed, and a large language model of a system core is subjected to professional government affair scene optimization training and is combined with a dynamically-updated knowledge graph technology.

A computerized system receives an original prompt that a querying user sends to a Large Language Model (LLM) that is operably connected to organizational data sources of an organization. Instead of executing the original prompt by the LLM, the system obtains user-related organizational context that pertains to characteristics of the querying user, obtains data-related organizational context that pertains to data from which the LLM is expected to obtain information for responding to the original query, and obtains pre-defined organizational policy rules, that indicate which type of users are authorized to access which type of organizational data. Based on the obtained data, the system modifies the original prompt into an adapted prompt. The system sends the adapted prompt, and not the original prompt, to the LLM for processing. The system obtains LLM-generated output from the LLM in response to the adapted prompt, and provides that LLM-generated output to the querying user.

The invention discloses an international propagation effect accurate evaluation method and system based on a large language model, and relates to the field of international propagation effect evaluation, and the method comprises the steps: obtaining a behavior log data set of a user, and dividing the user to a preset life cycle stage; extracting an interest label set of the user in a continuous time window, calculating a weight change rate of the same semantic keyword, and generating a user interest drift trend feature; extracting a propagation content feature vector of the to-be-evaluated propagation content, and calculating a basic matching degree with the interest label set; dynamically correcting the basic matching degree in combination with the life cycle stage and interest drift trend characteristics to obtain a final matching degree; extracting a path mode feature vector of the user, performing parameter matching with a propagation path mode library, and determining a propagation effect evaluation level; dividing the users into user groups according to the propagation effect evaluation levels, and implementing an intervention strategy on the user groups; according to the invention, refined grading evaluation of the propagation effect can be realized.

The invention provides a right content delivery strategy generation method and device based on AI big data, and relates to the technical field of strategy generation, and the right content delivery strategy generation method comprises the steps: obtaining user behavior data corresponding to a content platform type, so as to construct a corresponding target user portrait; dividing the to-be-put right and interest content into corresponding right and interest types according to content platform types; based on the target user portrait, analyzing preference degrees and response rates of user groups of the content platforms to different right and interest types; generating an adaptation relation matrix corresponding to the preference degree and the response rate; and distributing a corresponding right content type to each content platform according to the adaptation relation matrix, and determining a delivery frequency weight corresponding to each content platform. The cross-platform user portrait integration degree is improved, a right and platform dynamic matching mechanism is optimized, and the real-time adjustment capability of the delivery strategy is enhanced by constructing the cross-platform user portrait, dynamically generating the adaptation relation matrix and intelligently distributing the delivery strategy.

Action and / or process determination and recommendations for Robotic Process Automation (RPA) using semantic action graphs is disclosed. Semantic action graphs are graphs that store individual actions, and potentially graphical elements and / or text associated with the actions, as nodes, as well as the relationships between nodes as edges. Metadata to develop the semantic action graphs may be derived from task mining applications that can monitor the interactions of users with computing systems, workforce intelligence, etc. The semantic action graphs may be for a user, an organization, an industry, product-wide, etc. At their lowest level of granularity, the recommendations may be for mouse clicks, key presses, Application Programming Interface (API) calls, system events, etc. At higher levels of granularity, the recommendations may be for opening an order, creating a lead, approving a work item, etc.

The invention discloses an AI search recommendation method, system and device combined with commodity semantic understanding and a medium, and belongs to the technical field of commodity recommendation. The method comprises the steps that semantic vectors of commodity titles and descriptions are generated; generating a feature vector of the commodity image; fusing the semantic vector of the commodity title and description with the feature vector of the commodity image to generate a comprehensive semantic vector of the commodity; generating a user interest vector; updating the user interest vector by using a recurrent neural network based on the user interest vector; and carrying out joint modeling on the comprehensive semantic vector of the commodity and the updated user interest vector to generate a recommendation result. According to the method, a content semantic driving modeling mode is adopted, recommendation judgment can still be conducted through the deep matching relation between commodity image-text semantics and user behavior preferences even under the condition that user historical behaviors are limited or new commodities are online, good cold start adaptability is achieved, and the method is suitable for popularization and application. And meanwhile, high recommendation difference and content diversity are shown for different user groups.

The invention discloses a dynamic self-adaptive recommendation strategy optimization method for a business handling failure scene, and relates to the technical field of business handling recommendation and intelligent decision making, and the method comprises the steps: firstly collecting various types of data of a whole business handling process, and guaranteeing the integrity and real-time performance at a frequency of 100 milliseconds per time; a decision tree and Bayesian network fusion algorithm is used for attribution, and direct and indirect reasons are clarified; integrating data to construct a user portrait, and mining potential and subsequent demands; generating a recommendation scheme set based on attribution and portraits, and adjusting priorities and forms in combination with scene features; feedback data is introduced, a strategy weight is optimized by using a gradient descent algorithm, and the scheme is updated regularly; a multi-dimensional index weighted evaluation effect is set, and emergency optimization is carried out if the evaluation result does not reach the standard; and establishing a distributed strategy library, and reusing the optimal strategy of the similar scene by using a K-nearest neighbor algorithm. According to the method, failure reason accurate positioning and personalized recommendation are realized, the recommendation effect is continuously optimized along with data accumulation, and the method is adaptive to multiple service types and user groups.

A processor may gather raw data comprising a plurality of characteristic data samples of a target user group. The processor may categorize the characteristic data samples into a plurality of user-related classes and triggers. The processor may build an input property graph for each characteristic data sample. The processor may augment the input property graph by a concept of hierarchies. The processor may determine a modification vector from the augmented input property graph. The processor may train an encoder / decoder combination machine-learning system. An embedding vector and a modification vector are used as input for the decoder to build a trained machine-learning generative model.

The invention relates to the technical field of artificial intelligence and natural language processing, in particular to a law and regulation question answering system and method based on an industry large model, and the system comprises a data layer, a model layer, a service layer and an application layer. The data layer is responsible for collecting, storing and managing law and regulation data; the model layer comprises an industry large model and an RAG module; the service layer provides question and answer generation, multi-round dialogue management and security control core services; the application layer provides diversified application scenes for different user groups; the method has the beneficial effects that the general large model is finely adjusted in a targeted manner by using massive professional data in the legal field, so that the model can deeply learn logic association between connotation and extension of legal concepts and provisions and judgment goals of cases. Through the fine adjustment, the model can accurately grasp the core of the question when facing a complex legal question, provides an answer meeting a legal professional standard, and ensures the accuracy and preciseness of the answer, thereby effectively solving the problem of understanding deviation of professional knowledge.

The invention provides an intelligent message pushing method and system, and the method comprises the steps: collecting the social behavior data and personal attribute data of a user in a social network; dividing the users into a plurality of groups by using the social behavior data and the personal attribute data based on an ant colony algorithm; analyzing a social relation and an interaction mode among users in the group; collecting message resources in the social network, and classifying and labeling messages; performing collaborative filtering processing on the to-be-pushed message, determining a target user group, and generating a message recommendation list; and pushing the recommended message to users in the target user group according to a preset pushing strategy. The user social behavior data and the personal attribute data are converted into the ant feature vectors based on the ant colony algorithm, similar feature vector ants are gathered through pheromone updating and path selection mechanisms, different user groups are formed, the user group division accuracy is improved, and user requirements are more accurately grasped.

The invention relates to the field of movies, and discloses a movie recommendation method and system based on big data, and the method comprises the steps: collecting the multi-dimensional behavior data of a user to form an initial behavior data set; based on the initial behavior data set, calculating preference scores of the user for different categories of short play contents to obtain a user interest distribution matrix; dividing user groups and labeling user category labels; forming an initial short episode content feature set; constructing a content semantic feature matrix; if the content semantic feature matrix and the user category label have the matching requirement, judging the integrating degree of the short episode content and the user preference, and outputting a preliminarily recommended short episode content list; and on the basis of the preliminarily recommended episode content list, sorting and optimizing the recommended episode content by using click rate and complete playing rate indexes in the historical distribution data to obtain a final recommended episode content list. The method has the following effect that the accuracy and the individuation degree of short play recommendation are improved.

The invention relates to the technical field of machine learning, and discloses a user recharging prediction method and device, equipment and a storage medium, and the method comprises the steps: associating behavior data of multiple platforms of a user through an equipment fingerprint algorithm, obtaining a user multi-dimensional feature set, determining a clustering number based on an elbow rule, and obtaining a user recharging prediction result; and performing clustering analysis on the user multi-dimensional feature set according to the clustering number, generating a user value grouping label, and inputting the user value grouping label into a random forest model to obtain recharging prediction results of different user groups. According to the method, multi-platform user behavior information is comprehensively integrated through an equipment fingerprint algorithm, data islands are broken, user value grouping labels are generated through elbow rule clustering, then the user value grouping labels are input into a random forest model to predict a recharging result, user basic features are considered, value grouping information is integrated, feature dimensions are enriched, and the recharging efficiency is improved. And users with different values can be described more accurately, so that the accuracy of recharging prediction of user groups with different values is improved.

An ecommerce messaging system described herein comprises one or more ecommerce extension apps that interact with a messaging app. These embodiments leverage the extension core to manage session apps and smart services for using tap-to-message and tap-to-update processes to generate, update, and manage order workflows and in-app stores through branded messages communicated between customer and seller session apps to share message transcripts on devices of the same user groups via the messaging host, empower ecommerce with action-key-based communication, serialize-deserialize processes, and dynamic update propagation, manage return-refund workflows with temporary tables until one-time entity data model updates can be performed for session apps, and configure multi-store shopping systems on sellers' multiple devices using a store-as-a-service model to enable distributors to distribute production copies of assigned stores to authorized publishers to add custom promotions, which will be published together as production copies to their subscribed user groups for multi-store shopping and tracking.

Techniques for providing explicit proxy solutions for 5G Service Access Service Edge (SASE) with service provider network attach are disclosed. In some embodiments, a system, a process, and / or a computer program product for providing explicit proxy solutions for 5G SASE with service provider network attach includes receiving data plane traffic associated with a User Equipment (UE) from a mobile core network at a Secure Access Service Edge (SASE) cloud network via a service provider network attach using an interconnect between the mobile core network and the SASE cloud network; enforcing a security policy on data plane traffic associated with the UE based on contextual information associated with the UE to provide secured data plane traffic using the security policy configured per user group and / or per user; and forwarding the secured data plane traffic from the SASE cloud network to its original destination if allowed by the security policy, and blocking or dropping the data plane traffic from the SASE cloud network if not allowed by the security policy.

The invention relates to the technical field of online sales, and discloses a healthy retail supply chain optimization method and device and computer device.The method comprises the steps that multi-level semantic analysis is conducted on obtained text data through a natural language processing technology, and potential demand keywords of a user are extracted; constructing a user portrait based on the user potential demand keyword and the user behavior data; grouping the users based on a preset grouping standard and the user portraits in the health scene, screening corresponding required health products for different user groups, and optimizing and adjusting the required health products in combination with market trend data in the user behavior data to obtain an adjusted product selection pool; and calculating a matching degree between the user and a preset candidate social platform based on the user portrait and the product selection pool, and generating a social platform recommendation list based on the matching degree. According to the method, multi-source heterogeneous data are integrated, potential demands of users are mined by using a large model, a product selection pool is optimized, and accurate recommendation and personalized marketing are realized in combination with user portrait analysis.

The invention discloses an e-commerce commodity image-text generation optimization system based on generative AI, relates to the technical field of image-text optimization, and is used for solving the problems of insufficient image-text attraction and low conversion rate caused by difficulty in generating differentiated image-text content for different user groups. Classifying and marking the commodities, judging attention conditions, extracting commodity core features and user preference features after screening, and determining a user interest weight value of each marked commodity; feedback information of the commodities under the corresponding classification marks is collected, image-text information is generated in combination with user interest weights, updating time is set to collect historical optimization effect data, and the historical optimization effect data is used for updating user behavior data, so that image-text optimization of different commodities is achieved, and the matching degree and the display attraction of contents and user preferences are improved.

The invention relates to a multi-dimensional user grouping matching method and device, equipment and a medium. The method comprises the following steps: firstly, acquiring real-time behavior data, consumption data, interest data, historical conversion data and social media data of a user group, and integrating to form a user feature data set; performing multi-dimensional data fusion and cross-dimensional correlation analysis on the data set to generate a user group feature set; the value weight of the user group is updated based on the feature set, grouping levels are divided in combination with service accuracy requirements, and a dynamic grouping mechanism is constructed; and finally, receiving bidding requests of the service providers, constructing a dynamic bidding model in combination with a dynamic grouping mechanism to calculate bidding scores of the service providers, sorting the service providers according to the scores, and generating an accurate matching result with the target user group. According to the method, the limitation of single-dimensional grouping is overcome, the timeliness and flexibility of grouping are improved, and the accuracy and efficiency of matching between the service provider and the target user group are effectively improved.

The invention discloses an anomaly detection method and system for transaction flow data, and relates to the technical field of big data analysis. The method comprises the following steps: based on transaction flow data, acquiring operation behavior data of a target user in a transaction process and interaction behavior data after the transaction is completed, and sorting according to timestamps to form a user operation behavior sequence and a user interaction behavior sequence; inputting the sequence into a pre-trained abnormal risk prediction model, and outputting an initial risk coefficient, wherein the model integrates an individual behavior baseline and an adaptive weight module; acquiring a group behavior baseline of the similar user group, and correcting the initial risk coefficient in combination with a matching result of the sequence and the group baseline to obtain a final risk coefficient; and performing classification abnormity early warning operation on the final risk coefficient based on a preset risk threshold. According to the method, through whole-process behavior sequence analysis, individual and group baseline dual calibration and dynamic weight adjustment, the accuracy and timeliness of transaction flow data anomaly detection are effectively improved.

An administrator of an organization may create access policies within an access request management system which define relationships between resources and request-and-approval flows. For example, the administrator may create an access policy, and the access policy may be applied across multiple resources of the organization. The access policy may indicate rules that indicate characteristics of a request-and-approval flow for granting access to the corresponding resources. For example, the rules may indicate a duration to grant access to the users, a group of users that are eligible to request the resource, approving users that provide approval to the requesting user, and information to collect from the requesting users, among other examples.

The present application discloses a low-orbit satellite beam scheduling and resource allocation method, device, equipment and storage medium, which relates to the field of communication technology. The method includes: determining the time slot priority of the user to be scheduled; dividing the users to be scheduled into different user groups and determining the wave position interference matrix between the user groups; allocating time slots based on the time slot priority of the user to be scheduled and the wave position interference matrix between the user groups, and generating a beam hopping spectrum; within the update period of the beam hopping spectrum, determining the target user to be scheduled corresponding to the available beam in each time slot; allocating resources to each time slot based on the service priority of the target user to be scheduled for the available beam in each time slot. Through the above method, interference isolation and time slot allocation calculation are completed based on user priority and user grouping, multi-user resource allocation scheduling is performed in a single time slot, air interface resource information is fully utilized, and spectrum utilization and system throughput are maximized.

The invention provides a multi-modal emotion recognition fusion method based on a multi-head attention mechanism, and the method comprises the steps: carrying out the feature extraction and fusion of various data, capturing the internal relation of a modal through the multi-head attention mechanism, achieving the information complementation between modals through cross-modal interaction, and dynamically adjusting the weight according to the quality of the modals. And a self-built database containing a large amount of Chinese data is constructed, and a culture adaptation optimization strategy is combined, so that the generalization performance and culture adaptability of the model in Chinese user groups are improved. The system is deployed on a cloud server, optimized hardware and software configuration is adopted, efficient model reasoning and multi-user concurrent processing are achieved, and the large-scale real-time application requirement is met. The emotion of the user can be monitored in real time and early warning can be provided in scenes such as psychological counseling and group emotion monitoring, professionals are assisted in better understanding the emotion state of the user, and the service effect is improved.

The invention relates to the field of advertisement putting, and provides an advertisement putting management platform and a management terminal thereof, and the platform comprises an account management module which provides account registration and login for a user; the advertisement putting management module is used for providing advertisement putting setting for an advertiser; the advertisement creativity management module provides advertisement creativity for an advertiser and supports the user to upload or create the advertisement creativity; the audience orientation and marketing module collects and arranges user data, constructs a user portrait, and helps an advertiser to carry out audience orientation and accurately put an advertisement to a target user group according to the user portrait; and the data monitoring and effect evaluation module is used for collecting advertisement putting data for the put advertisement, creating a data report according to the advertisement putting data, and analyzing and evaluating the advertisement data. According to the technical scheme, the problem that the advertisement effect is difficult to improve due to the fact that an advertisement putting management platform cannot accurately position audiences in the prior art is solved.

The invention provides a user terminal flow prediction method, medium and system, and belongs to the technical field of flow prediction model fine tuning, and the method comprises the steps: firstly constructing a general flow prediction basic model comprising a mathematical prediction module and a convolutional neural network; acquiring historical traffic data of a plurality of user terminals, preprocessing the historical traffic data and generating a traffic time sequence vector; clustering analysis is carried out on the vectors, and the users are divided into a plurality of clusters; and on the basis of the cluster with the maximum clustering scale, calculating the clustering similarity between other clusters and the cluster with the maximum clustering scale. Next, an LORA fine tuning model is set for each cluster, and the smaller the similarity is, the larger the parameter scale of the fine tuning model is; and on the basis of the trained general basic model, performing fine tuning on each cluster to obtain a corresponding LORA model. And for the to-be-predicted user terminal, the cluster to which the to-be-predicted user terminal belongs is determined according to the traffic time sequence vector of the to-be-predicted user terminal, and the corresponding LORA fine tuning model is adopted for prediction, so that personalized prediction of a user group can be realized.

The invention relates to the field of real-time translation, and provides a real-time duplex translation method based on multichannel parallel processing and a corresponding product, and the method comprises the steps: collecting multipath voice signals of at least two user groups in real time through a group of audio collection modules; dynamically adjusting beam forming parameters of each audio acquisition module in one group of audio acquisition modules based on a sound source positioning result, and feeding back the beam forming parameters to the corresponding audio acquisition modules; monitoring the voice activity of each audio acquisition module corresponding to each audio channel; when it is monitored that the voice activity of any audio channel reaches a preset condition, automatically activating the translation processing flow of the audio channel and keeping the monitoring state of the other audio channels; a parallel processing mechanism is adopted for voice signals of the activated audio channels, and meanwhile real-time translation of the currently activated audio channels and voice activity monitoring of the other audio channels are executed; and transmitting a translation result of the current speaking user to other users participating in dialogue in the user group to realize synchronous coordination of multichannel data.

The invention discloses a social group evaluation method and system based on double-space fusion calibration, and the method comprises the steps: firstly, extracting user preference features based on user-news interaction data and news type tags, constructing a fixed user portrait through employing a statistical analysis and maximum preference strategy, and completing the division of the user portrait; then, starting from the matching space, sorting the users in the matching space through joint sorting of the behavior activeness and the behavior coupling degree; meanwhile, starting from a semantic space, generating a portrait semantic expression by utilizing a large language model, generating a semantic center point in combination with user behaviors, and sorting the users in the semantic space based on a vector distance; on this basis, the sorting information of the matching space and the semantic space is fused, the semantic consistency and the behavioral representativeness of the users are balanced, and finally the user groups with comprehensive representativeness under each type of portraits are screened out; further, based on the high-frequency behavior record and portrait features of the representative user, constructing an injection simulation behavior sequence, selecting low-interaction news items, and generating an evaluation sample with interference features; and finally, retraining a recommendation model, evaluating recommendation response change of each portrait group on low-interaction information, and quantifying anti-interference scores of the system under different portrait consistency conditions, thereby revealing robustness difference of the recommendation system, and providing support for security enhancement and portrait recognition. According to the method, the performance difference and the anti-interference capability of different user portrait groups in a recommendation system can be described.

The invention discloses a CNN and Informer fused high-risk user loss early warning and retention method and system, and belongs to the technical field of big data, and the method comprises the steps: obtaining the multi-source data of each user, and constructing the static features, daily dynamic features, and a time sequence feature matrix of a set time period of the user through the multi-source data; constructing a CNN and Informer fused deep learning model, and predicting the loss probability of the user by using the trained deep learning model based on the static features and daily dynamic features of the user and the time sequence feature matrix of the time period set by the user; constructing an RFL model, and obtaining a user value score by using the RFL model and the multi-source data; and in combination with the value scores and loss probabilities of all the users, grouping all the users by using a Kmenas model, outputting user grouping results and grouping portraits, and performing early warning and retention. According to the method, potential high-risk lost users can be accurately identified, and the retention cost is reduced.

The present invention discloses a method and system for personalized content recommendation and ROI improvement based on multidimensional user portraits. The method comprises the following steps: collecting user data and generating a multidimensional user portrait through feature extraction and weighted fusion; constructing a correlation graph using the advertising platform content feature vector and the multidimensional portrait as nodes of a graph structure, and the user's interactive behavior data as edges of the graph structure; analyzing the correlation graph and the multidimensional portrait using a causal inference model to identify invalid related variables and key variables that influence user decision-making; making recommendation decisions based on the causal inference results within a multi-objective optimization framework to generate a personalized recommendation list for each user; and calculating the potential conversion value of different user groups based on a time series prediction model to optimize the advertising budget allocation strategy. The present invention can improve the personalization of advertising recommendations and the ROI of advertising delivery.

The invention relates to the technical field of power consumption behavior analysis and intelligent auditing of power consumers, and discloses an auditing rule threshold dynamic recommendation method, system and device based on feature mapping and a medium, and the method comprises the steps: integrating multi-source data of a user, and generating a comprehensive feature vector through a multi-modal fusion network; then, performing group division on the users according to feature similarity under a federated learning framework; using a time sequence generative adversarial network to generate enhanced power consumption data for each user group so as to improve model robustness; training a threshold dynamic adjustment strategy network in combination with reinforcement learning and causal reasoning, and outputting an initial recommendation threshold; the optimal balance threshold set between the false alarm rate and the missing report rate is searched through multi-target Bayesian optimization, and diversified choices are provided for decision making. According to the method, the leap from automation to intellectualization and from single-point optimization to multi-target collaborative decision making is realized, and the accuracy of auditing efficiency and the scientificity of decision making are remarkably improved.

The invention relates to the technical field of recommendation methods, and provides a commodity recommendation method, which realizes phased high-accuracy comprehensive recommendation of commodities, has better recommendation accuracy, and comprises the following steps: firstly, collecting and storing basic information of the commodities, forming a basic attribute identifier of each commodity, and establishing an association approach between different commodities; then, basic information of the users is formed, commodities with matched attributes are screened in combination with the basic information, basic recommendation of the commodities is formed, associated crowds near the users are collected, recognition and classification of user groups are formed, and optimization and complementation of recommendation reference information are achieved based on information complementation formed among multiple users with high similarity in the user groups; according to the method, commodities with matched attributes are screened on the basis of basic information in combination with extended information, optimized recommendation of the commodities is formed, self-owned information is recorded according to use records of users, and dynamic recommendation of the commodities is achieved by combining the basic information of the users with associated crowds and forming phased information replacement and optimization.

Automatic detection and handling of security-related anomalies by utilizing machine learning and a large language model (LLM). A computerized method for detecting and handling security threats in an organizational network of an organization includes: (a) collecting event data that pertain to organizational users, organizational devices, and organizational resources of the organizational network; (b) constructing user profiles, device profiles, and resource profiles; (c) constructing Organizational Context information that pertains to organizational users, organizational devices, and organizational resources; (d) constructing a time-series of events, enriched with the Organizational Context information; (e) analyzing the time-series of events using a Machine Learning process that detects an anomalous event, and automatically generating an alert message pertaining to and describing the anomalous event.