49 results about "Category specific" patented technology

A database query processing method includes receiving a natural language request for information contained within a database, categorizing the natural language request in a particular category from a plurality of categories, and selecting a large language model and one or more database structures for which a SQL request is to be directed based in part on the particular category. The selected large language model is prompted to generate a SQL request directed to the selected set of one ore more database structures and the received SQL request is validated for the particular category. The validated particular SQL request is used to access data within the database and one or more visualizations is caused to be displayed based at least in part on a result of the validated particular SQL request.

Disclosed herein are system, method, and computer program product aspects for textual data extraction, generation, and evaluation. Text is input into a first fine-tuned large language model (LLM) to generate an atom (e.g., a textual phrase in a particular category). The atom is input into a second LLM that has been fine-tuned for structured output corresponding to the particular category of information. A logical structure is generated based on a structured output of the second LLM, wherein the logical structure represents the textual phrase of the atom and contains a contextual attribute associated with the textual phrase. The embodiment then stores the logical structure into a knowledge graph as a modifier node having a time-variant attribute (e.g., a timestamp associated with the textual phrase and / or the contextual attribute).

A computer-implemented method may include receiving an image to analyze for potential objects; providing the image as input to a pretrained machine learning model, where the pretrained machine learning model is trained to detect a known object that corresponds to a predefined category within the pretrained machine learning model or a generic object; determining based on output from the pretrained machine learning model, that the image comprises an instance of a generic object; performing, based on determining that the image comprises an instance of a generic object, a similarity search between the image and a library of images; and determining, for the instance of the generic object, a specific category defined by the library of images based at least in part on the similarity search. Various other methods and systems are also disclosed.

For at least a selected class attribute of the multiple class attributes, one or more bias metrics are determined that estimate a degree to which a particular workflow (having a set of processing stages) is biased in association with the class attribute. Each user of a set of users is associated with a set of user data to be processed by the particular workflow. At least one of the set of processing stages includes executing a machine-learning model. It can be detected that a bias-mitigation option corresponding to a specific class attribute has been selected. For each of at least two of the set of processing stages: a de-biasing technique is selected; and the processing stage is modified by applying the de-biasing technique. A modified version of the particular workflow (which includes the modified processing stages) is applied to each of a set of input data sets.

Methods and systems are described herein for a mechanism for detecting model bias using a test model to compare predictions for specific categories of data. Thus, the system may detect category-specific model bias and / or present alternative predictions, for example, via overfitted and alternative machine learning models. Another mechanism for detecting model bias is to use residual data from predictions obtained from a machine learning model as the machine learning model processes different datasets (e.g., a training dataset and a compare dataset).

A database query processing method includes receiving a natural language request for information contained within a database, categorizing the natural language request in a particular category from a plurality of categories, and selecting a large language model and one or more database structures for which a SQL request is to be directed based in part on the particular category. The selected large language model is prompted to generate a SQL request directed to the selected set of one ore more database structures and the received SQL request is validated for the particular category. The validated particular SQL request is used to access data within the database and one or more visualizations is caused to be displayed based at least in part on a result of the validated particular SQL request.

A program is provided for causing a computer providing a virtual space (VS) to function as: a category presenting unit (41) that presents, on a screen, one or more categories related to one or more items used by an object (O) in the virtual space (VS) in a manner whereby a category can be specified; a category specifying unit (42) that specifies a particular category from among the categories, based on a user input; an item presenting unit (43) that presents, on the screen, one or more items associated with the specified particular category in a manner whereby an item can be specified; an item specifying unit (44) that specifies a particular item from among the one or more items, based on a user input; and a variation presenting unit (45) that presents one or more variation items for the specified particular item on the screen in association with the particular item in a manner whereby a variation item can be specified. This program provides an information processing technique that enables variation items for a particular item to be specified in an efficient manner.

An object of a system according to an embodiment is to generate video content in accordance with a specific category by effectively utilizing in-house information.SOLUTION: A system includes an information collection unit, an analysis unit, and a generation unit. The information collection unit collects necessary data from in-house information sources. The analysis unit analyzes the data collected by the information collection unit. The generation unit generates moving image content according to the specific category on the basis of the data analyzed by the analysis unit.SELECTED DRAWING: Figure 1

A method, computer program product, and computing system for identifying a portion of a transcript associated with a particular category. A segment from the transcript is aligned with a segment from a model-generated summary of the transcript. A portion of the transcript omitted from the model-generated summary is identified based upon, at least in part, the identified portion of the transcript associated with the particular category and the aligned segments from the transcript and the model-generated summary.

The present invention discloses a self-supervised model copyright protection method, electronic device and storage medium based on double-layer optimization, which are used to enhance the model copyright protection security in self-supervised learning scenarios. Internal optimization designs a watermark optimization strategy to strongly associate watermark samples with target labels, and obtains dynamic weights based on the feature similarity loss between the watermark sample and other non-target samples to optimize the effect of the watermark on each category of data, thereby avoiding interference between the watermark sample features and the features of other category samples; external optimization refers to the use of a joint loss function in the encoder feature space to retain the normal classification function of the model and the copyright protection goal of the model watermark, thereby ensuring the effectiveness, utility and concealment of the watermark. The present invention can effectively prevent the failure of the watermark on a specific category through a double-layer optimization strategy, thereby ensuring the stability and robustness of the model watermark in practical applications.

A system is provided.SOLUTION: This system is provided with a means for analyzing the contents of an inquiry and classifying it into a designated category, a means for generating a reply sentence based on the analyzed result, a means for adding an attached file to the reply sentence and a means for transmitting a mail.SELECTED DRAWING: Figure 1

A computing system and method for data augmentation. A method includes receiving first data examples, each in one of a particular category associated with a label; receiving a definition for each label, resulting in label definitions, the label definitions describing subject matter of the associated category; generating second data examples from the first data examples, each in one of the particular categories; generating third data examples from the first data examples based on a data repository; merging the second and third data examples to form a corpus; and training a language model with the corpus.

A system is provided.SOLUTION: A system comprising: means for analyzing article information in an electronic device using a generative artificial intelligence model to determine whether it is likely to contain spoiler information; means for hiding spoiler information from display to a user based on the determination; and means for enabling customization of the extent of spoiler information by filtering certain categories and content based on user settings.SELECTED DRAWING: Figure 1

Based on receiving data defining a new data item for a construction project corresponding to a particular category of data items, a computing system (1) automatically: (i) predicts that a change event for the construction project is needed by inputting the new data item into a first machine learning model trained to predict a need for a change event from data items corresponding to certain categories of data items, including the particular category of the new data item, (ii) determines initial recommended data for the predicted change event, and (iii) determines additional data for the predicted change event corresponding to a particular class of additional data by inputting the initial recommended data for the predicted change event into a second machine learning model trained to predict one or more classes of additional data for a change event, and (2) automatically create a data item representing the predicted change event.

A video processing apparatus includes a video analyzer that analyzes video data captured by a surveillance camera, detects an event belonging to a specific category, and outputs a detection result, a display controller that displays, together with a video of the video data, a category setting screen for setting a category of an event included in the video, and a learning data accumulator that accumulates, as learning data together with the video data, category information set in accordance with an operation by an operator to the category setting screen. The video analyzer performs learning processing by using the learning data accumulated in the learning data accumulator.

Methods and systems are provided for managing private data access for resolving a query. The system, e.g., accesses private data generated by a plurality of distinct applications associated with a user and inputs the data into a data classification model. The model classifies the private data into specific data categories based on data type. The classified data is then aggregated into a dataspace, which serves as a unified repository for private data corresponding to a specific category collected from the distinct applications. A unique access configuration is associated with the dataspace to define a set of authorized connection types permitted to access the repository. In response to a query requiring the private data, access to the dataspace is provided if the requesting entity matches the authorized connection types defined in the unique access configuration.

Systems and methods for using large language models (LLMs) to analyze category specific workflows for generating an optimized workflow that can be used in live conversations to provide a response are described. The methods include categorizing a plurality of projects into specific categories. The actions performed by agents for the plurality of projects are translated into workflows. The workflows are analyzed based on optimization factors and clustering options, such as including redundancies in workflow steps, using alternative solutions to a workflow step, determining whether any escalation performed is justified and if so, adopting escalation related steps. The workflows are consolidated and optimized into an optimized workflow that is tested and verified, and then used in a live conversation.

The embodiment of the invention relates to a data real-time classified storage method and device, electronic equipment and a storage medium. The method comprises the steps of obtaining real-time operation data collected from target equipment at a target moment; the distance between the real-time operation data and each class center in a preset number of class centers is determined, and each class center in the preset number of class centers corresponds to one data class; based on each obtained distance, determining a target data category to which the real-time operation data belongs; and based on the distance between the real-time operation data and the class center of the target data class, determining a target position of the real-time operation data in a preset data classification table, and storing the real-time operation data in the target position. According to the embodiment of the invention, the data collected in real time in the operation process of the equipment is classified, the data classification table is utilized to facilitate more accurate data extraction and more efficient analysis of data of a specific category, and the data recording and analysis efficiency is improved.

The present disclosure provides systems and methods for hierarchical multi-level model distillation that create specialized artificial intelligence models from large language models. A hierarchy for generating specialized generative models from a large language model is determined, including models associated with different categories and complexity levels for tasks with varying computational requirements. The specialized models are trained using the large language model based on specific categories to provide responses to category-related requests. Each model is pruned based on complexity levels to enable responses to requests of varying complexity. The models are quantized from common precision factors to corresponding precision factors associated with complexity levels. Deployment locations are determined for each pruned model based on complexity levels. Specialized models are transmitted to corresponding locations for deployment across diverse computational environments.

A video processing apparatus includes a video analyzer that analyzes video data captured by a surveillance camera, detects an event belonging to a specific category, and outputs a detection result, a display controller that displays, together with a video of the video data, a category setting screen for setting a category of an event included in the video, and a learning data accumulator that accumulates, as learning data together with the video data, category information set in accordance with an operation by an operator to the category setting screen. The video analyzer performs learning processing by using the learning data accumulated in the learning data accumulator.

This invention discloses a two-stage unsupervised adaptation method, system, device, and medium for visual language models. These are corresponding solutions. The solutions include: constructing a high-quality auxiliary dataset through adaptive retrieval, achieving both cost-effectiveness and versatility, and providing an efficient and reliable new paradigm for cross-domain knowledge transfer; furthermore, decomposing the complex adaptation task into a two-stage optimization process, controlling the distribution of auxiliary data to be similar to the pre-training or target distribution through adjustable data distribution control parameters, significantly reducing the difficulty of single-step adaptation; simultaneously, during category filtering, using weak supervision signals contained in the image-text pair data, matching frequency statistics, and predictive information entropy to filter samples of specific categories from the large-scale image-text dataset, avoiding the introduction of noisy category samples; and furthermore, based on two-stage training, the model can gradually adapt to the target task and target data distribution, improving adaptation performance and achieving higher classification accuracy.

A video processing apparatus includes a video analyzer that analyzes video data captured by a surveillance camera, detects an event belonging to a specific category, and outputs a detection result, a display controller that displays, together with a video of the video data, a category setting screen for setting a category of an event included in the video, and a learning data accumulator that accumulates, as learning data together with the video data, category information set in accordance with an operation by an operator to the category setting screen. The video analyzer performs learning processing by using the learning data accumulated in the learning data accumulator.

This disclosure describes techniques for automated data extraction, validation, and routing based on unstructured text data. In some cases, the techniques described herein include receiving text data, segmenting the text data into multiple segments, assigning each segment to a category, generating a prompt for each segment based on the segment's category, extracting field values from each segment using the generated prompt, validating or rejecting the extracted field values based on category-specific validation rules, and routing the validated field values to category-specific target databases and / or reviewer platforms based on the validation results.

A computing system may generate, using a plurality of category-specific models, a plurality of validation reports assessing a procedure associated with an organization, wherein each category-specific model of the plurality of category-specific models is associated with a corresponding business function of the organization and is trained via machine learning to produce a corresponding validation report that is a written assessment of whether the procedure is satisfactory for the corresponding business function. The computing system may generate, based on the plurality of validation reports, a combined validation report that is a written assessment of whether the procedure is satisfactory for the organization.

Information messages can be used to signal changes to graphics rendering information (GRI) with coded rendered graphics / video data in a bitstream. Syntax elements can be added to an information message to represent updated parameters in GRI categories such as rendering engine parameters, projection matrix, world to cam matrix, and / or depth parameters. Category presence indicators can be set to signal when information messages included syntax elements representing parameters in a particular GRI category. If a particular category presence indicator is set, a decoder can parse from the information message the syntax elements representing parameters in the associated GRI category. Alternatively, if a particular category presence indicator is not set, the information message will not include syntax elements representing parameters in the associated GRI category and the decoder can use previously-received and / or default parameter values when generating target graphics from the rendered graphics data.

The invention discloses a multi-category feature selection method for independent inspection of category specific conditions. The method comprises the following steps: step 1, data preprocessing and category division; 2, class specific condition independence testing is carried out; step 3, carrying out class specific Markov Blanket discovery, and carrying out class specific Markov Blanket discovery; step 4, carrying out Markov Blanket structure optimization, and carrying out Markov Blanket structure optimization; step 5, carrying out cross integration on the class specific features; 6, training and verifying the model; according to the method, a class-specific condition independence test method is adopted, a sample subset is independently constructed for each class, and the condition dependence intensity is calculated, so that statistical deviation caused by inter-class data distribution difference is effectively avoided. By accurately capturing a correlation structure only existing in a specific category, a condition independence test result is more fit with a real structure of data, and the resolution is remarkably improved.

To provide a provision device, a provision method and a provision program that can provide a transaction object considering an evaluation axis when a user selects a commodity.SOLUTION: A provision device 1 comprises: an estimation unit 132; a determination unit 133; and a provision unit 134. The estimation unit 132 estimates a plurality of evaluation axes for a target commodity to be selected and an importance degree of each evaluation axis on the basis of information of a commodity itself, user information of a user and a conversation process associated with a specific category which is performed with the user using a conversation type service. The determination unit 133 determines similarity among commodities using data similarity of each of a plurality of commodities corresponding to each of the plurality of evaluation axes, and the importance degree of each evaluation axis. The provision unit 134 provides a content in which a target commodity and a similar commodity being similar to the target commodity are arranged in an order depending on similarity among the commodities determined by the determination unit 133.SELECTED DRAWING: Figure 2

A video processing apparatus includes a video analyzer that analyzes video data captured by a surveillance camera, detects an event belonging to a specific category, and outputs a detection result, a display controller that displays, together with a video of the video data, a category setting screen for setting a category of an event included in the video, and a learning data accumulator that accumulates, as learning data together with the video data, category information set in accordance with an operation by an operator to the category setting screen. The video analyzer performs learning processing by using the learning data accumulated in the learning data accumulator.

A method of facilitating content selection includes generating benchmark queries for a particular category. Generating the benchmark queries includes applying a text prompt as input to a language model trained on a knowledge base. The text prompt requests search queries indicative of user interest in the particular category. The method also includes selecting, responsive to new search queries entered by users, content items associated with the particular category for delivery to client devices of the users. Selecting the content items includes determining whether the new search queries correspond to the particular category at least in part by comparing the new search queries to a query set that includes the benchmark queries.

A multi-class classifier (MCC) is trained using annotated data. The annotated data comprises instances of sample data and associated label data. Creation of the annotated data and subsequent active learning by the MCC uses resources. A target-aware active learning system selects sample data for addition to an annotation queue based on factors such as current accuracy of a particular class determination and priority of that class. As each instance in the sample data in the annotation queue is annotated and used for subsequent training, accuracy of particular classes is improved until a specified accuracy for that class is attained. By being selective in the ordering of instances in the annotation queue, overall resource usage and corresponding costs associated with creating annotated data and training is reduced. Overall accuracy for all classes is improved using a smaller overall set of annotated data compared to naïve approaches.