138results about "Unstructured textual data retrieval" patented technology

A method, an apparatus, and a computer-readable storage medium for generating a document graph. A plurality of electronic documents is received. Each electronic document has a predetermined document type. A machine learning model is selected from the plurality of machine learning models based on the predetermined document type. The selected machine learning model is instructed to extract a plurality of document portions from each electronic document in the plurality of electronic documents in accordance with the predetermined document type. A relationship between two or more document portions is defined based on the content of each document portion, and the document portions are associated based on the relationship. A graph structure having a plurality of nodes is generated. Each node includes at least one document portion. Each node is connected to another node in accordance with the relationship between document portions included in the nodes. The graph structure is stored.

The invention relates to the technical field of passenger service, and discloses a construction method and device of a dynamic extension word bank, equipment and a medium. The method comprises the following steps: acquiring multi-source heterogeneous original knowledge data related to civil aviation passenger service, and converting the multi-source heterogeneous original knowledge data into a standard format text; processing the unstructured data in the standard format text based on the adaptive word segmentation model and the stop word list to obtain a plurality of candidate words; performing multi-dimensional corpus feature quantitative analysis on each candidate word to determine candidate keywords; and performing similarity calculation based on the entity word segmentation and the candidate keywords, determining effective candidate keywords, and adding the effective candidate keywords into the dynamic expansion word bank. By means of the method and device, the technical problems that in the prior art, a word bank used by a civil aviation service system is usually based on a static vocabulary or depends on manual compiling and updating, the static word bank cannot be updated in real time, the characteristics of the civil aviation field cannot be flexibly handled, and the maintenance cost is high due to manual updating and maintenance are solved.

The invention provides a self-adaptive transformation and migration method and system for an internal website, and relates to the technical field of website migration. Comprising the following steps: constructing a hierarchical credential adaptation architecture, establishing a dynamic authority-data mapping mechanism, realizing knowledge base-data migration linkage and deploying an adaptive security protection chain. The method is used for realizing collaborative improvement of compatibility, security and knowledge service efficiency of an internal website in a credential environment through hierarchical credential adaptation, dynamic permission and data association, knowledge base linkage and self-adaptive security protection; besides, through cooperative work of the hardware driving adaptation module and the software interface conversion module, automatic identification of hardware features and interface features, precise analysis of adaptation requirements and dynamic generation of configuration files are realized. The compatibility of an internal website to different models of servers, autonomous controllable components and software interfaces in a credential environment is remarkably improved, the adaptation period is shortened, and stable operation of the system is guaranteed.

The invention discloses a data error correction method, a data error correction device and a storage medium, and relates to the technical field of data processing. The method comprises the following steps: firstly, converting first text data in a first document to be subjected to error correction into structured second text data retaining a content association relationship, and identifying a target error type of the second text data through semantic analysis; and then, calling a target error correction engine integrated with the industry knowledge base to realize accurate identification and correction of industry exclusive errors. In this way, the problems that a general error correction tool is low in precision, not suitable for industry characteristics and free of compliance basis are solved. And finally, converting the error correction result into a revision mark supported by the first document, and mapping the revision mark to the first document, so that accurate positioning and display of the error correction result in the original first document can be realized, a user can be helped to quickly focus on a core risk, and the document checking efficiency can be improved while the error correction precision of the document is ensured.

A method is provided. The method is executed by an extraction engine implemented as a computer program within a computing environment. The extraction engine executes action and task mining on unstructured data. The method includes receiving a communication including unstructured data defining an action and automatically processing the communication by utilizing at least one generative artificial intelligence (AI) model to extract details of the action being performed in the unstructured data. The method includes automatically converting the action into an activity or a task of a process associated with the communication.

A Multimodal Search System (MSS) is designed to enable accurate search and retrieval across different modalities of data, including images, videos, and text. The MSS leverages multimodal and domain-specific models to extract and represent features and information from multimodal data, enabling seamless querying using text-based search queries. The MSS integrates multiple pipelines, including feature vector-based and tag-based approaches, to ensure robust and scalable search capabilities. The MSS addresses current multimodal-search challenges by leveraging a hybrid approach that combines multimodal embeddings, text embeddings, and semantic tagging to enable cross-modal searching. Additionally, for visual data, both tagging and embedding techniques are integrated within a single modality, ensuring a comprehensive and accurate representation. By assembling multiple retrieval strategies, the solution harnesses the strengths of each, resulting in a balanced and effective search system.

The invention discloses a slope stability evaluation method based on a large language model and an intelligent prediction model. The method specifically comprises the steps of establishing a slope training database and a large language model training database; constructing a slope stability prediction artificial intelligence model and a slope field large language model; slope parameter extraction is carried out on the data uploaded by the user through a slope field big language model, and whether the extracted slope parameters meet the input requirements of a slope stability prediction artificial intelligence model or not is judged; if the slope parameters in the data uploaded by the user are incomplete, the slope field big language model feeds back the slope parameters needing to be supplemented to the user; after the users are supplemented, the slope field big language model is extracted again and judged again until it is monitored that the users have uploaded all slope parameters; the slope field big language model sends the complete slope parameters into the trained slope stability prediction artificial intelligence model, and the slope stability prediction artificial intelligence model outputs the stability coefficient of the slope.

A method, according to one approach, includes: causing a received query to be translated from a first language to a second language. The method also includes generating potential answers for the translated query, and extracting features from the translated query. The method also includes causing the extracted features to be converted into feature vectors. The method also includes causing the feature vectors to be compared against existing vectors in a knowledge base that correspond to past question-answer pairs. The method also includes causing the potential answers to be ranked based at least in part on an outcome of comparing the feature vectors against the existing vectors in the knowledge base. Furthermore, the method includes causing a final answer to be generated based at least in part on the ranked potential answers.

The application discloses a data error correction method and device and a storage medium, and relates to the technical field of data processing. First, the first text data in the first document to be corrected is converted into structured second text data that retains content association, and the target error type of the second text data is identified through semantic analysis. Then, a target error correction engine integrated with an industry knowledge base is called to realize accurate identification and correction of industry-specific errors. In this way, the problem of low accuracy of general error correction tools, inadaptation to industry characteristics and lack of compliance basis is solved. Finally, the error correction result is converted into revision marks supported by the first document, and the revision marks are mapped to the first document, accurate positioning and display of the error correction result in the original first document can be realized, and the user can be quickly focused on the core risk, the error correction accuracy of the document is ensured, and the auditing efficiency of the document is improved.

To provide an information processing device configured to improve the efficiency in confirming whether the content of a target document matches required conditions or not.SOLUTION: An information processing device comprises: acquisition means for acquiring a first document and a second document; analysis means for extracting, from the first document, contents to be confirmed and generating a task for each content; information extraction means for extracting attribute information indicating a feature of the first document; selection means for selecting candidate documents from among one or more second documents on the basis of the attribute information; search means for searching the candidate documents for a condition pertaining to the content to be confirmed by the tasks; processing means for sequentially inputting, to a trained model, the tasks generated by the analysis means, and causing the trained model to answer whether the content to be confirmed by the input tasks matches the condition searched for from the candidate document; and output means for outputting the result processed by the processing means.SELECTED DRAWING: Figure 3

A system for searching contents within a database of textual contents is described. Said contents and the corresponding databases may be of any kind such as movie titles, music titles, song titles, scientific terms, medical titles / terms, titles formed of one or more sequences of symbols, a list of telephone numbers, a contact list, etc. Upon providing one or more keyword, the search system provides a list one or more of corresponding contents to a user. According to one aspect, after selecting a presented content, by the user, a process corresponding to selected content is executed by a processor.

The invention discloses an intelligent test paper composition method and device, and belongs to the technical field of image or video recognition or understanding. The intelligent test paper composition method comprises the following steps: collecting original data of various questions, and carrying out standardization processing on the collected data to form a standard question bank; wherein the standardization processing comprises adding applicable teaching materials, applicable grades, question types, first-level knowledge point labels, second-level knowledge point labels, difficulty coefficients and score information to the original data; pre-de-duplication processing is carried out on questions in a standard question bank, then the standard question bank is divided into a plurality of first-level question banks according to first-level knowledge point labels, and the first-level question banks are divided into a plurality of second-level question banks according to second-level knowledge point labels; and screening questions from the standard question bank according to set test paper composition information. And when the repeated questions belong to the second-level question bank related to the core knowledge point, replacing the repeated questions with another question in the same second-level question bank, thereby ensuring the accuracy of the test paper composition again.

To output in an easily understandable manner the result of determination of whether or not accept restrictions by a prescribed regulation or guideline.SOLUTION: Provided is a processing device including at least one processor, with the at least one processor constituted so as to carry out the process of: accepting input of a content in which one or a plurality of texts are included that include at least one word which is the object of determination; acquiring the one or plurality of texts from the accepted content; determining, for each of the acquired one or plurality of texts, whether or not each text in the content can be used, on the basis of restrictions by a prescribed regulation or guideline; and outputting determination result information in such a way that determination result information associated with the result of determination is displayed together with the content.SELECTED DRAWING: Figure 15

A method includes receiving, via a processing system, unstructured data associated with one or more operations performed in a production system, extracting, via the processing system, information from the unstructured data based on one or more domain specific prompts associated with the production system. The method also includes integrating, via the processing system, the information with structured data to generate updated structured data, receiving, via the processing system, a request associated with the one or more operations, and generating, via the processing system, a response based on the updated structured data, wherein the response includes a visualization representative of the response.

Systems and methods access, from one or more data storage locations, a dataset; perform data analysis on the dataset to detect one or more data quality characteristics each corresponding to at least one data quality dimension including timeliness, uniqueness, accuracy, completeness, validity, or consistency; evaluate the one or more data quality characteristics present in the dataset to identify one or more common patterns; and generate one or more data quality rule recommendations based on the identified one or more common patterns.

A computing system identifies a heterogenous multi-entity graph user graph for a user based upon an identifier for the user. The user graph includes nodes and edges connecting the nodes. The nodes include topic nodes representing topics and entity nodes. The entity nodes represent people associated with the user, documents of the user, or derived information that is derived from the documents. The computing system identifies a cluster of the topic nodes corresponding to a productivity area of the user and performs a walk of the user graph based upon the subset of the topic nodes to identify a subset of the people, the documents, and the derived information. The computing system causes a graphical user interface (GUI) to be presented on a display, where the GUI includes identifiers for the subset of the people, the documents, and the derived information.

Some implementations of the disclosed systems, apparatus, methods and computer program products may provide for determination of resource usage by tenants in a multi-tenant server system. Tenants may provide resource requests to a database of the multi-tenant server system and such resource requests may include context data. Periodic snapshots of the database may be performed to determine the pending resource requests received by the various tenants and, based on the snapshots and the context data, the resource usage of the various tenants, as well as the system as a whole, may be determined and forecasted for the future.

The invention discloses a panoramic view construction method, system and device for an aviation journey and a medium, and the method comprises the steps: obtaining aviation data from an aviation order system and business data from a third-party service system, and forming heterogeneous aviation journey data; identifying a communication protocol of the heterogeneous air travel data, and analyzing the heterogeneous air travel data according to the communication protocol and a current service analysis rule to obtain a first service field; when the heterogeneous air travel data comprises change data, updating the service analysis rule according to a semantic association relationship between the first service field and the change data, and obtaining a second service field from the change data; and constructing a panoramic view of the aviation journey according to the first business field and the second business field. By adopting the embodiment of the invention, the business analysis rule can be automatically updated, a comprehensive panoramic view with high timeliness is obtained, and the industrial collaboration efficiency and the passenger experience are further improved.

Provided herein are methods, systems, and computer program products for automatically generating datasets by processing federated forums with artificial intelligence techniques. The computer-implemented method includes obtaining conversational data from a federated forum source, categorizing at least one portion of the conversational data based on a designated application using a first set of artificial intelligence techniques, extracting information from at least a portion of the categorized data related to test case-related issues, validating at least a portion of the extracted information by analyzing a portion of the conversational data belonging to a plurality of entities and related to the extracted information using a second set of artificial intelligence techniques, generating one or more datasets related to at least one of the test case-related issues for at least one of the designated applications using the validated information, and performing at least one automated action based on the one or more generated datasets.

A method (300) includes processing a first query (116) to classify the first query as being related to an existing topic (251) that corresponds to a respective topic summary (250) stored in a datastore (198). The respective topic summary associated with a respective summary of past query-response interactions between a user and an assistant interface (150). The method also includes retrieving the respective topic summary from the topic summary datastore that corresponds to the existing topic, processing the first query conditioned on the respective topic summary retrieved from the topic summary datastore to generate a first response (118), and providing presentation content (180) based on the first response for output.

A system and method of processing a search request is provided. Identification of a desired content item is based on comparing a topic of the search request to previous user input. The method includes providing access to a set of content items with metadata that describes the corresponding content items and providing information about previous searches. The method further includes receiving a present input from the user and determining a relatedness measure between the information about the previous searches and an element of the present input. If the relatedness measure is high, the method also includes selecting a subset of content items based on comparing the present input and information about the previous searches with the metadata that describes the subset of content items. Otherwise, the method includes selecting a subset of content items based on comparing the present input with the metadata that describes the subset of content items.

Intelligent mapping from created item information to sustainability reference content from a variety of sources can be implemented to facilitate created item footprint management and other sustainability applications. The difficult task of finding appropriate emission factors across a portfolio can be automated. Assisted search can be implemented using enhanced search techniques. Fallback mappings can be implemented to accommodate different levels of granularity during search. A machine learning model can be trained based on a variety of input data, including confirmed mappings, mapping history, and rules. The process of mapping to emission datasets can thus be simplified, enabling footprint calculations to proceed.

The distributed database can implement unstructured data replication using an internal or external storage location. Metadata, such as a directory table that lists the unstructured files, can be replicated across different deployments, followed by replication of the staged data. Replicating the staged data can be implemented by replication of only the stage metadata or replication of the database files between the deployments.

A system and method of processing a search request is provided. Identification of a desired content item is based on comparing a topic of the search request to previous user input. The method includes providing access to a set of content items with metadata that describes the corresponding content items and providing information about previous searches. The method further includes receiving a present input from the user and determining a relatedness measure between the information about the previous searches and an element of the present input. If the relatedness measure is high, the method also includes selecting a subset of content items based on comparing the present input and information about the previous searches with the metadata that describes the subset of content items. Otherwise, the method includes selecting a subset of content items based on comparing the present input with the metadata that describes the subset of content items.

Various examples are provided related to identification and management of compliance-related information associated with data breach events. In one example, a method includes receiving a first data file collection associated with a first data breach event; generating information associated with presence or absence of protected information elements of all or part of the first data file collection and incorporating data files including the protected information elements in a second data file collection; analyzing data files selected from the second data file collection; and incorporating the information associated with the analysis into machine learning information that may be used for subsequent analysis of data file collections.

The invention discloses an index calculation method and device based on model fusion and a medium. The method comprises the steps that mixed data used for calculating a target index and user calculation requirements are acquired; determining a data proportion corresponding to the mixed data, and determining a structural data calculation model and a non-structural data calculation model in a preset model library based on the data proportion and / or a user calculation demand; determining result weights corresponding to the structural data calculation model and the non-structural data calculation model based on the data proportion and / or user calculation requirements; and determining an index calculation result based on the result weight, the structural data calculation model and the non-structural data calculation model. Based on the proportion of the structured data to the unstructured data in the mixed data, the combination of a traditional algorithm and large model calculation can be optimized, so that the advantages of the traditional algorithm in the aspects of structured data processing and interpretability are fully utilized.

Automatically detects a performance degradation in document retrieval. When document data is appended to the database (300), the learning unit (260) updates the language model (61) using machine learning. The performance evaluation unit (270) calculates a first statistic related to the position of each specific document data in the multiple specific document data based on the document retrieval results where each of the first tags attached to the multiple specific document data is set as a retrieval query (TQRY). The performance evaluation unit (270) calculates a second statistic related to the position of each specific document data in the multiple specific document data based on the document retrieval results where each of the second tags attached to the multiple specific document data is set as a retrieval query (TQRY). If the change in the first statistic caused by the update of the language model (61) accompanying the appending of at least one document data is greater than a first threshold, and the change in the second statistic caused by the update is greater than a second threshold (Th2), the performance evaluation unit (270) detects a performance degradation in document retrieval.

This invention relates to the medical field. It provides a method, apparatus, device, and medium for structured processing of medical materials based on OCR. The method includes: acquiring an image of a medical material to be identified; performing text recognition on the image using OCR technology to obtain multiple target texts; sorting the multiple target texts to obtain a target text set; determining the medical material type corresponding to the target text set based on a pre-trained medical material model; determining multiple structured field names corresponding to the medical material type of the target text set, and multiple keywords corresponding to each structured field name, based on a pre-constructed structured dictionary; and performing structured processing on the target text set. This invention can greatly eliminate the differences between medical materials in hospitals across the country, has high support for medical materials in hospitals nationwide, provides comprehensive coverage, has high efficiency in extracting structured medical information, and can make reasonable use of photographed medical materials.

Generally discussed herein are devices, systems, and methods for clustering based on context triggered piecewise hashing (CTPH). A method can include determining a first index of a first CTPH string of the file. The first index can include contiguous bits of the CTPH string. The first index can be smaller than the CTPH string, such as to be a proper subset of the CTPH string. The method can include determining the first index matches a second index of a cluster of files and in response to determining the first index matches the second index of the cluster, associating the file with the cluster. The method can include determining that the file includes malware based on the cluster.

The invention relates to the technical field of contract processing, and particularly provides an intelligent purchase contract comparison method and device in a cloud environment, and the method comprises the steps: firstly carrying out the data acquisition in a data storage layer, carrying out the format conversion, text preprocessing, feature extraction and similarity calculation in a data processing layer, and finally carrying out the similar contract screening in an application layer. Compared with the prior art, the method can help enterprises to optimize purchasing strategies and reduce purchasing risks, and is of great significance to digital transformation and fine management of enterprise purchasing services.