62 results about "Base domain" patented technology

Various embodiments of the present disclosure provide model-based domain-aware autocomplete techniques for generating autocomplete suggestions in a complex search domain. Example embodiments are configured to generate, using a domain-aware autocomplete model, a label for an autocomplete suggestion based on a set of keywords within an autocomplete suggestion training dataset associated with a target domain source. Example embodiments are also configured to generate, using a weak-labeling model, an updated label for the autocomplete suggestion by decorrelating the set of keywords from the label. Example embodiments are also configured to generate, using a sentence classification model, a category for the autocomplete suggestion based on the updated label. Example embodiments are also configured to, using the domain-aware autocomplete model, generate a suggestion-category pair (SCP) based on the autocomplete suggestion and the category for the autocomplete suggestion. Example embodiments are also configured for initiating performance of a search query resolution based on the SCP.

The invention provides a field NL2SQL data automatic synthesis method and system based on a large language model, and the method comprises the steps: building an SQL template library through collecting a public data set and real business query data, carrying out the grading according to the SQL complexity, and forming an SQL template set covering different difficulties; specific SQL queries are generated based on an SQL template library and a large language model, verification is carried out through multiple mechanisms such as grammar check, execution verification and result dimension consistency check, and correctness and performability of the SQL queries are ensured. Generating a plurality of candidate natural language questions according to the verified SQL query and the large language model, calculating a semantic consistency score of each candidate question and the SQL query through a cross consistency evaluation mechanism, screening out the question with the most consistent semantics as a final question-answer pair sample, and meanwhile, eliminating low-quality samples by setting a consistency threshold value, so as to obtain a final question-answer pair sample; and the data accuracy and consistency are further improved.

The present application discloses a method, system, and computer system for providing real-time detection of malicious URLs based on a machine-learning powered domain risk scoring. The method includes (i) identifying a subset of higher risk websites, wherein the higher risk websites are at risk for potential malware injection or modification, and (ii) in response to identifying the subset of higher risk websites, performing an active measure based at least in part on the identified subset of higher risk websites.

Various embodiments of the present disclosure provide model-based domain-aware autocomplete techniques for generating autocomplete suggestions in a complex search domain. Example embodiments are configured to generate, using a domain-aware autocomplete model, a label for an autocomplete suggestion based on a set of keywords within an autocomplete suggestion training dataset associated with a target domain source. Example embodiments are also configured to generate, using a weak-labeling model, an updated label for the autocomplete suggestion by decorrelating the set of keywords from the label. Example embodiments are also configured to generate, using a sentence classification model, a category for the autocomplete suggestion based on the updated label. Example embodiments are also configured to, using the domain-aware autocomplete model, generate a suggestion-category pair (SCP) based on the autocomplete suggestion and the category for the autocomplete suggestion. Example embodiments are also configured for initiating performance of a search query resolution based on the SCP.

The invention discloses a field event element extraction method and system based on prompt learning and knowledge completion, and the method comprises the steps: converting an extraction task into a constrained generation problem through structured prompt, achieving the stable extraction under a small number of labeled samples through the existing knowledge prior of a large language model, reducing the dependence on large-scale labeled data, and improving the extraction efficiency. And the cross-domain adaptive capacity is improved. A vectorization knowledge retrieval mechanism is introduced, related evidences are obtained from an external knowledge source, elements which are not clearly expressed in a text are complemented, the integrity of information is enhanced, and the accuracy and credibility of a result are improved through knowledge verification. And through a self-adaptive fusion mechanism of prompt and knowledge, the model can flexibly balance text context and external knowledge, and the robustness of complex context and fuzzy expression is improved. And finally, standardized and structured element data are output, a convenient interface is provided for downstream event atlas construction and analysis tasks, and the automation degree of domain knowledge processing and the system integration efficiency are improved.

Various embodiments of the present disclosure provide model-based domain-aware autocomplete techniques for generating autocomplete suggestions in a complex search domain. Example embodiments are configured to generate, using a domain-aware autocomplete model, a label for an autocomplete suggestion based on a set of keywords within an autocomplete suggestion training dataset associated with a target domain source. Example embodiments are also configured to generate, using a weak-labeling model, an updated label for the autocomplete suggestion by decorrelating the set of keywords from the label. Example embodiments are also configured to generate, using a sentence classification model, a category for the autocomplete suggestion based on the updated label. Example embodiments are also configured to, using the domain-aware autocomplete model, generate a suggestion-category pair (SCP) based on the autocomplete suggestion and the category for the autocomplete suggestion. Example embodiments are also configured for initiating performance of a search query resolution based on the SCP.

The invention provides a method and device for cross-domain target detection and a related product, and relates to the technical field of artificial intelligence. The method comprises the following steps: acquiring a training sample image and a domain label thereof; acquiring image-level features of the training sample images; determining an image-based domain classification result according to the image-level feature, and determining a first loss based on the domain label and the image-based domain classification result; performing object feature extraction on the image-level features to obtain instance-level features of the training sample images; determining an object-based domain classification result according to the instance-level feature, and determining a second loss based on the domain tag and the object-based domain classification result; determining a third loss according to the image-based domain classification result and the object-based domain classification result; and performing parameter updating on the cross-domain target object detection model based on the first loss, the second loss and the third loss. According to the embodiment of the invention, the accuracy of cross-domain target detection can be improved.

The invention provides a domain knowledge supplement and SQL generation optimization method and device based on a large model, and belongs to the technical field of large language models. The method comprises the following steps: integrating multi-source heterogeneous knowledge, and carrying out information input and parameter configuration on the multi-source heterogeneous knowledge; based on a semantic segmentation model and a knowledge extraction agent, automatically analyzing unstructured document content and converting the unstructured document content into structured knowledge; performing word segmentation processing on the basis of the question input by the user, and calling a retrieval algorithm to recall related knowledge from the knowledge base so as to supplement information required by answering; and on the basis of the recalled structured knowledge, designing corresponding application paths according to different types of knowledge and generating an SQL (Structured Query Language). According to the domain knowledge supplement and SQL generation optimization method based on the large model, flexible, accurate and efficient SQL generation is realized through efficient integration of multi-source heterogeneous knowledge, intelligent analysis of unstructured documents, accurate recall of related knowledge, high-quality SQL generation and optimization and dynamic update of knowledge management.

The invention relates to the field of computer vision, in particular to a transformer-based domain adaptive target detection method and electronic equipment, and the method comprises the following steps: S1, obtaining a source domain and a target domain; s2, establishing a model; s3, establishing a domain confrontation feature alignment module; s4, an adaptive domain discriminator; according to the method, a domain confrontation feature alignment module is designed in an encoder part, a learnable domain confrontation token is used as a query, features generated by an encoder layer are used as key value pairs, and the features generated by the encoder are aligned through adversarial learning; a self-adaptive domain discriminator is designed, so that the model can better process hard cases, and the domain self-adaptive capability of the model is further enhanced; the method provided by the invention shows good detection effect improvement on a plurality of domain adaptive public data sets, and compared with the prior art, the method provided by the invention is effectively improved in the aspect of feature alignment of the features output by the encoder, has obvious advantages in domain adaptive target detection, and has excellent domain generalization ability.

Systems and methods for favicon comparison-based similar domain detection include receiving a base domain, the base domain being associated with an enterprise; receiving a domain list comprising a plurality of domains; performing a favicon comparison between the base domain and each of the plurality of domains within the domain list; and classifying each of the plurality of domains within the domain list as one of being associated with the enterprise or not being associated with the enterprise based on the favicon comparison.

The invention relates to the field of mechanical equipment fault diagnosis, and discloses a part characteristic frequency-based domain generalization fault diagnosis method and system, a medium and equipment, and the method comprises the steps: generating a part fault simulation signal based on physical knowledge of part fault characteristics, and taking the part fault simulation signal as a physical knowledge domain; adopting a deep neural network, extracting fault features from a plurality of source domains and physical knowledge domains through a feature extractor, and inputting the fault features into a full connection layer and a domain generalization adversarial network; the distribution difference between the source domain and the physical knowledge domain is minimized in the full connection layer, the distribution difference serves as a constraint condition, the constraint model learns invariant features based on the physical knowledge domain, all source domain fault feature distribution is aligned with physical knowledge domain fault feature distribution, and a fault classification result is output through a classifier; the domain to which the extracted fault features belong cannot be classified through a domain discriminator region in the domain generalization adversarial network, so that a domain classification result with common features of a plurality of domains is output. The problem caused by unbalanced domain data can be solved.

The invention relates to a domain adaptive target detection method and system based on adversarial double-teacher knowledge distillation, and the method comprises the steps: obtaining source domain data, carrying out the self-learning stage training of a Deformable DETR model through the source domain data, and obtaining a basic student model; generating a trunk teacher model and an auxiliary teacher model based on the basic student model, and setting different levels of domain discriminators in student branches in the basic student model for aligning resistance features; respectively inputting the target domain strong enhancement image into the trunk teacher model and the auxiliary teacher model, generating a pseudo label, training student branches by using the source domain strong enhancement image and the target domain strong enhancement image, and generating a target detection model; and in the process of training the student branches, supervising the student branch training by using the pseudo labels, and updating the weights of the auxiliary teacher model, the trunk teacher model and the student branches. According to the method, the detection precision is improved under the condition that target domain labeling is not needed.

Malicious homoglyphic domain name (MHDN) generation and associated cyber security applications are described. MHDN generation may be performed by, for example, generating, based on training data, a set of operations for use in generating the one or more potential MHDNs, wherein each operation of the set of operations may be configured to modify a base domain name according to a respective homoglyphic characteristic. The set of operations may be used to generate one or more candidate MHDN mutators. The candidate MHDN mutators may be tested for fitness values corresponding to respective likelihoods of generating an MHDN and the candidate MHDN mutators may be applied to one or more base domain names to generate potential MHDNs.

The application discloses a specific field pipeline task type dialogue system based on an LLM, which comprises an inquiry subsystem and an answering subsystem, wherein: the inquiry subsystem parses and guides a user to supplement key information of a question according to user input task text, the answering subsystem summarizes key information of a dialogue task, and performs vector matching through a built-in vertical field local vector database to generate a final answer to the dialogue task. The application uses a large language model to realize natural language understanding, dialogue state tracking and natural language generation modules in a task type dialogue system pipeline, simultaneously realizes a rule matching-based strategy learning module, understands and replies to user input text, and combines the content of a local knowledge base, so that the whole process is more interpretable, the illusion problem of the large language model is reduced, only a small amount of sample prompt learning is needed, a large amount of data training or fine tuning is not needed, and the workload during field migration is reduced.

A method and system to provide annotation-based domain mapping are disclosed. The method includes receiving a plurality of source domain objects and destination domain objects from a user. Further, the method includes storing the plurality of source domain objects and destination domain objects. Further, the method includes analyzing the destination domain objects to extract predefined annotations. Further, the method includes parsing the predefined annotations to extract mapping details and conditional details. Further, the method includes traversing a metadata graph generated based on the predefined annotations to identify a matching path in a source domain that correlates with a class path present in the metadata graph, and evaluating at least one condition associated with the matching path using at least one source attribute value. Further, the method includes mapping the source attribute value to at least one destination attribute value, upon satisfying the at least one condition.

Malicious homoglyphic domain name (MHDN) generation and associated cyber security applications are described. MHDN generation may be performed by, for example, generating, based on training data, a set of operations for use in generating the one or more potential MHDNs, wherein each operation of the set of operations may be configured to modify a base domain name according to a respective homoglyphic characteristic. The set of operations may be used to generate one or more candidate MHDN mutators. The candidate MHDN mutators may be tested for fitness values corresponding to respective likelihoods of generating an MHDN and the candidate MHDN mutators may be applied to one or more base domain names to generate potential MHDNs.

The invention relates to the technical field of network security management and flow analysis, and provides a method and a system for positioning a data flow with a null domain name based on a task ID, and the method comprises the steps: monitoring a synchronous message through a process A, and capturing network flow; the abnormal task ID is received through the process B, and a server IP list is determined according to the corresponding domain name of the abnormal task ID; filtering a Client Hello data stream of a target IP (Internet Protocol) from the captured message; judging whether each data stream carries a Server Name field or not; and counting the proportion of the data streams which do not carry the fields and have null domain names, and outputting an evidence message. The problems that traditional manual analysis is low in efficiency, poor in accuracy and high in personnel skill requirement are solved, and the network security management efficiency is remarkably improved.

This invention discloses a method and system for generating real-time content in a specific domain based on AI technology, belonging to the field of artificial intelligence technology. The method includes the following steps: 1. Constructing a local material knowledge base RAG_src in the form of word vector storage using AI technology, whose materials include multimodal materials such as text, voice, audio / video, or H5; 2. Constructing a local domain knowledge classification knowledge base RAG_policy in the form of file storage using AI technology, whose content includes domain specification documents and technical classification documents categorized by domain knowledge and information; 3. Constructing PageIndex by combining a Large Language Model (LLM) with RAG_policy; 4. Receiving content generation requests from real-time software, calling RAG_src and PageIndex through LLM to generate real-time content that meets the requirements of the corresponding scenario; 5. Checking the generated content and outputting compliant content. This invention solves the problems of non-standard content generation, disconnection from relevant standards, low reuse rate of proprietary materials, and low retrieval efficiency in existing technologies, achieving accurate, efficient, and real-time content generation. It is adaptable to various scenarios such as real-time content generation, content compliance checking, and real-time domain assistance in specific domains. Application scenarios include, but are not limited to, oral practice and oral examinations according to teaching syllabi.

Among other techniques, techniques for real-time cross-domain data management are described. An example method includes decomposing an enterprise into a plurality of different context-based domains, wherein each context-based domain produces a respective data product owned by the respective context-based domain; generating a first context-based domain dataset owned by a first context-based domain of the plurality of context-based domains; generating a first data product from the first context-based domain dataset; generating a second context-based domain dataset owned by a second context-based domain of the plurality of context-based domains; generating a second data product from the second context-based domain dataset; identifying a first data record of the first data product, wherein the first data record is associated with a first entity; identifying a second data record of the second data product, wherein the second data record is associated with a second entity; determining the first entity and the second entity are the same entity; and merging, in real time based on one or more global interface rules, the first data record and the second data record without changing either the first dataset or the second dataset.

Embodiments of the invention provide a computer-implemented method that includes executing a machine learning (ML) model operable to perform a ML task that includes generating a ML output responsive to a ML input. The ML output includes encoded domain information associated with a domain. The encoded domain information is encoded in a computer-code-based domain-specific data structure, and the ML task is associated with the domain.

Various embodiments of the present disclosure provide model-based domain-aware autocomplete techniques for generating autocomplete suggestions in a complex search domain. Example embodiments are configured to generate, using a domain-aware autocomplete model, a label for an autocomplete suggestion based on a set of keywords within an autocomplete suggestion training dataset associated with a target domain source. Example embodiments are also configured to generate, using a weak-labeling model, an updated label for the autocomplete suggestion by decorrelating the set of keywords from the label. Example embodiments are also configured to generate, using a sentence classification model, a category for the autocomplete suggestion based on the updated label. Example embodiments are also configured to, using the domain-aware autocomplete model, generate a suggestion-category pair (SCP) based on the autocomplete suggestion and the category for the autocomplete suggestion. Example embodiments are also configured for initiating performance of a search query resolution based on the SCP.

The application belongs to the field of gesture recognition, and provides an electromyography gesture recognition method and system based on iterative self-training domain adaptation. The method comprises the following steps: obtaining electromyography data of a historical user and electromyography data of a new user; extracting source domain time-frequency features based on the electromyography data of the historical user, and extracting target domain time-frequency features based on the electromyography data of the new user; using a difference-based domain adaptation method to align the source domain time-frequency features and the target domain time-frequency features; training a model based on the aligned target domain time-frequency features, judging whether the model meets the iteration number requirement, if yes, predicting the target domain time-frequency features of the electromyography data of the new user to obtain a gesture recognition result; otherwise, giving pseudo-labels to unlabeled data of the target domain time-frequency features, selecting label classes with a number of pseudo-labels less than a certain value for upsampling to balance all classes, repeating the process of aligning the source domain time-frequency features and the target domain time-frequency features until the iteration number judgment process until a set iteration number is reached.

A nameserver receives a request to access a webpage from a client device. The nameserver generates a blockchain query based on the request for querying a domain registrar blockchain storing domain name registration data in smart contracts. The nameserver sends the first blockchain query to the domain registrar blockchain. In response to receiving a nameserver identifier from the domain registrar blockchain, the nameserver retrieves a DNS record for the webpage using the received nameserver identifier. The nameserver then provides information from the retrieved DNS record to the client device to allow the client device to access the webpage.

The method and device for identifying DoH service, the electronic device and the storage medium are disclosed, wherein the method comprises: sending a first domain name resolution request to a domain name resolution server at a preset detection frequency, the first domain name resolution request being used to request a DoH-based domain name resolution service for a first detection domain name; updating a target network address corresponding to the first detection domain name in a database according to each first resolution result; comparing a network address in the database with a network address to be queried in a domain name resolution service type query request when the domain name resolution service type query request is received; and determining that the domain name resolution service in the domain name resolution service type query request is the DoH-based domain name resolution service when there is a network address same as the network address to be queried in the database.

The present invention relates to the field of communications, and aims to solve the problem that different users need DNS to resolve different IPs when accessing the same domain name. The present invention proposes a user-based domain name resolution method and system, and the method includes: when a terminal accesses the network, NAS sends a billing start message to AAA, AAA resolves the billing start message to obtain a user ID and a terminal IP address, and notifies a customized DNS to locally add a mapping relationship between the user ID and the terminal IP; when the customized DNS receives a new domain name management request, it adds a corresponding relationship between the user ID, domain name, IP address and control policy; when the terminal accesses the domain name, it sends a domain name resolution request to the customized DNS, and the customized DNS obtains the user ID according to the terminal IP, and then obtains and executes the corresponding control policy according to the user ID and the domain name, and returns the response result to the terminal. The present invention achieves different users accessing the same domain name to resolve different resolution results and execute corresponding responses by associating the terminal IP, user ID and domain name resolution control policy.

A system includes one or more processors to receive a base set of base domain names and a training set of training domain names; execute a plurality of similarity functions; iteratively execute a machine learning model (e.g., a neural network, an XGBoost model, a support vector machine, etc.) using the plurality of similarity values for each of the training set of training domain names; train the machine learning model; receive a candidate domain name; execute the plurality of similarity functions based on a comparison between the candidate domain name and each base domain name of the base set of base domain names; execute the trained machine learning model to generate a candidate malicious domain name prediction value for the candidate domain name; and generate a record identifying the candidate domain name and the candidate malicious domain name prediction value.

A computer-implemented method for domain adaptation of an object detection model includes obtaining a domain vector for a domain from one or more images in the domain, the domain vector representing the property of the domain. The domain vector is input into a fully connected layers in the object detection model. A domain-specific result of the object detection model is provided as output. The method can further include computing a domain tensor and inputting the domain tensor into convolutional layers in the object detection model.

The invention discloses a domain text data labeling method based on an ontology and a large model, which comprises the following steps of: 1) acquiring or constructing a domain ontology of a target domain, and determining related concepts and attributes of the target domain; 2) analyzing the domain ontology to obtain hierarchical structures and semantic association information of concepts in the domain ontology; 3) marking the text data to be marked by utilizing a marking framework according to the hierarchical structure of the domain ontology; wherein for non-leaf nodes in the hierarchical structure, text areas corresponding to the non-leaf nodes are identified and divided from the to-be-labeled text data, and segmentation results are transmitted downwards step by step till the leaf nodes; and for the leaf nodes, extracting entities and semantic relationships among the extracted entities from the text regions corresponding to the leaf nodes, and generating a structured labeling result. According to the method, multi-level and different-granularity associated information in the field text data is fully mined, and the labeling precision and semantic integrity of the field text data are improved.

Systems and methods for providing a financial / clinical data interchange are provided. The financial / clinical data interchange provides a distributed implementation to a secure hash key (i.e., a token) and value pair data derived from a medical claim (e.g., patient identification, submitter identification, payer identification, encounter identification, and the like) and enriched with submitter-based domain data. The token may be used as a data attribute in an API that unlocks a pointer to the value (e.g., a fast healthcare interoperability resources (FHIR) uniform resource identifier (URI) to the patient encounter associated with the claim) to leverage a FHIR query for all documented medical records associated with the claim the payer is authorized by the submitter to view.

The present application discloses a domain model design method based on multi-agents and related devices, the method comprising: using the knowledge of the target domain to train the model design agent, the domain expert agent, the development agent and the test agent respectively; initiating an event storming request to the model design agent, and the model design agent performs event storming with the domain expert agent, the development agent and the test agent; the model design agent generates the final result of the event storming by using the event storming recognition result and the domain knowledge base of the target domain; initiating a domain model design request to the model design agent, and the model design agent collaborates with the domain expert agent, the development agent and the test agent to generate an initial model design; initiating a domain model review request to the model design agent, and the model design agent collaborates with the domain expert agent, the development agent and the test agent to optimize the initial model design.