13 results about "Temporal expressions" patented technology

The invention provides a financial risk early warning model construction method based on an attention mechanism, and belongs to the technical field of financial risk early warning models.The financial risk early warning model construction method comprises the steps that financial data of 256 time steps are preprocessed through a sliding window mechanism, a multi-head sparse attention layer containing 8 attention heads is constructed, a long sequence is decomposed into 64 sub-sequence segments, and the sub-sequence segments are divided into 64 sub-sequence segments; a fixed window and a random sampling sparse mask are adopted to reduce the calculation complexity, a layered attention structure is designed to process local and global dependency relationships respectively, 512-dimensional learnable position coding is introduced to enhance time sequence expression, a combined loss function including cross entropy loss and attention regularization is established, and the cross entropy loss and the attention regularization are combined. And a dynamic hierarchical fusion weight adjustment function is used to adaptively adjust two layers of attention weight distribution through a game mechanism, and finally a financial risk early warning model for outputting five risk levels is constructed, so that efficient and real-time long-time series financial data analysis is realized.

The invention relates to the technical field of intelligent traffic, and discloses a semantic information-based vehicle spatial-temporal trajectory similarity calculation method and device, and the method comprises the steps: obtaining spatial-temporal trajectory data, dividing the spatial-temporal trajectory data with one hour as an interval, and constructing a Voronoi graph to generate a Thiessen polygon; performing multi-dimensional feature extraction on the Thiessen polygon region containing the trajectory to obtain spatial features, time features and POI semantic features of the region; constructing a space-time relation graph, and learning high-quality representation of nodes by adopting a space-time integrated random walk strategy with track perception and a GAT algorithm in sequence; constructing a model T-LSTM for enhancing the time expression ability to learn the final expression of the trajectory sequence; measuring similarities between nodes and between trajectories, and designing a graph-based double-level contrast loss function training model; according to the method, the capturing capability of the track representation on the functional semantic information can be improved.

The invention discloses a Chinese electronic medical record data processing method and system based on a large language model. The method comprises the following steps: acquiring a Chinese electronic medical record, and performing data preprocessing to obtain a preprocessed Chinese electronic medical record; inputting the preprocessed Chinese electronic medical record into a multi-agent based on the fine-tuned large language model, and combining the multi-agent with the input data according to different processing tasks of the multi-agent to obtain a first cue word; retrieving a medical extraction knowledge base according to the first cue word to obtain corresponding medical knowledge, and combining the first cue word to obtain a second cue word; each agent calls the fine-tuned large language model according to the corresponding second cue word, information extraction and standardization processing are carried out, and extracted medical information is output. According to the method, the accuracy of medical entity and attribute extraction can be improved, combined extraction of complex medical entities and attributes thereof is achieved, standardized mapping of medical terms and time expressions is completed, and the term ambiguity and fuzzy time analysis problems are solved.

Techniques are disclosed herein for resolving date / time expressions while transforming natural language to a logical form such as a meaning representation language. A class label for a token in a natural language utterance and a meaning representation for the natural language utterance can be predicted. The class label can be associated with a date / time expression. The meaning representation can include an operator and a value. When the value associated with the class label matches a predetermined value type or the operator matches a predetermined operator, the value and / or the operator can be modified, and an executable statement can be generated for the meaning representation. A query on a computing system can be executed using the executable statement.

Contextual routing and / or management of communications is described. Server(s) of service provider can receive, from a computing device of a customer, a first communication that is associated with an appointment or a reservation. The server(s) can determine context data associated with the first communication, and determine, based on the context data, that the first communication is associated with a temporal expression. The server(s) can negotiate a time for scheduling or rescheduling the appointment or the reservation based on the temporal expression, and generate a second communication to be sent to the customer. The second communication includes an actuation mechanism for the customer to perform an action.

The application discloses a Web-based disaster monitoring and early warning spatio-temporal process dynamic evolution visualization method, and specifically comprises the following steps: constructing an expression model to describe disaster component elements and disaster processes as the basis for constructing a disaster spatio-temporal expression ontology, then taking the expression ontology as the top-level guide for constructing a knowledge graph, extracting specific knowledge from a large number of real three-dimensional models to construct a disaster knowledge graph, and designing a disaster data self-adaptive organization mode and a visualization data scheduling optimization mechanism to improve the compactness and relevance of data in a Web partition. The expression model and the disaster knowledge graph enable the computer to understand disaster elements and the multidimensional relationship between them, thereby realizing dynamic adjustment and optimization of three-dimensional data, effectively reducing the overhead of complex disaster scene loading and rendering, and realizing high-fluency and low-time-consumption three-dimensional display of disaster spatio-temporal process dynamic evolution.

The invention discloses a time-space association-considered news story context generation method, which comprises the following steps of: firstly, designing a two-stage unsupervised story discovery method, namely, embedding a preliminarily aggregated news article according to document-level semantics of a news stream; the semantic related news is distributed to the same news story more finely through keyword distribution in the candidate story; then, respectively analyzing the time expression and the place name entity extracted from the news article into standard format time and position coordinates by utilizing regular matching and Wikii data, so as to mine space-time information in the news article; and finally, proposing a space-time distance calculation method based on optimal transmission, introducing a distance attenuation function to model an attenuation rule of space-time association, and constructing a story context considering the space-time association by using a maximum spanning tree. According to the method provided by the invention, the development and evolution process of the news event can be more accurately expressed in the space-time dimension.

This invention discloses a method for constructing, querying, and maintaining a time-bucket-based dual-temporal RDF block list index, belonging to the field of semantic web temporal data management technology. The invention first performs dictionary encoding, timestamp standardization conversion, and differential encoding compression on the dual-temporal RDF quintuples; then, it dynamically determines the time bucket granularity through a sliding window to achieve balanced partitioning of temporal data and eliminate empty buckets; it constructs a two-layer auxiliary index of existing bitmaps and changing bitmaps to achieve fast query pruning and block acceleration; based on the bitmap, it constructs a three-layer temporal block list core index adapted to SPO / POS / OSP modes; it uses a localization strategy to complete index addition, deletion, modification, and maintenance, and during queries, it achieves precise retrieval through time bucket pruning, bitmap filtering, and block list. This invention supports queries in all scenarios including no temporality, single temporality, and dual temporality, with dual-temporal path query time consistently below 5ms. It has advantages such as efficient index construction, low maintenance cost, strong data adaptability, and standardized temporal expression, making it suitable for efficient storage and query management of large-scale dual-temporal RDF data.

The invention discloses a deep learning auxiliary rule-based time expression extraction method, which comprises the following steps of: forming a rule set by constructing time words, time modifiers and conjunctions, matching a specific format by using a regular expression, and identifying a time expression in a text; a pre-training model and Bi-LSTM are utilized to fuse context information, position coding is utilized, lexical features and part-of-speech features are introduced to calculate a feature function, the weight is learned through maximum likelihood estimation of a CRF layer, and finally a Viterbi algorithm is used for decoding, so that optimization of a time expression and result extraction are achieved. The extracted time expressions are integrated, and different processing strategies are adopted according to whether the extraction results are consistent or not. According to the method, the defect of specific time expression recognition by purely depending on rules is overcome, and the extraction efficiency and accuracy of the time expression are remarkably improved by adding the deep learning part.

A method and device for processing temporal expressions from unstructured texts for filling a knowledge database. A temporal expression in a text is determined. A type of the temporal expression is determined as a function of the text. The temporal expression and the type are mapped on a prediction of a value of the temporal expression in a context-free representation of the temporal expression.

The present disclosure relates to a system for interpreting and managing imprecise temporal expressions. Techniques for extracting, identifying, and consuming imprecise temporal elements ("ITE") are disclosed. User input may be received from a client device. A prediction for one or more time intervals indicated by the user input may be generated based on an ITE model. The user input may be associated with the prediction and provided to the client device.

This invention discloses a method and system for generating medical ward round records based on a large language model. Specifically, it includes: transcribing the speech of a doctor-patient ward round to generate initial speech text containing speaker identifiers; inputting the text into a first processing agent for preprocessing, including terminology standardization, temporal expression standardization, and preliminary semantic normalization; inputting the preprocessing results along with patient medical data into a second generation agent; constructing structured prompts through the synergistic effect of long-term and short-term memory; and calling a large language model fine-tuned from the medical corpus for semantic reconstruction to generate a structured ward round record; establishing a cross-stage consistency constraint mechanism between terminology standardization, temporal expression standardization, and semantic reconstruction; verifying entity association paths in the generated results using a knowledge graph; and generating a consistency feedback signal. The feedback signal is then sent back to the first processing agent and the second generation agent for iterative correction.

The invention discloses a correction method and system for time type generation accuracy in a Text2SQL task, and the method comprises the steps: creating the Text2SQL task based on a received query problem inputted by a user, and executing the Text2SQL task to convert the query problem into a semantic SQL; extracting a time expression in the semantic SQL, and calling a time inference model to carry out time correction so as to generate real time based on the time expression; the semantic SQL is corrected on the basis of the real time, and the corrected semantic SQL is obtained; and correcting the information of the corrected semantic SQL, converting the corrected semantic SQL into an executable query SQL, carrying out query based on the query SQL, and returning a query result. According to the invention, reasoning of a time expression is realized by using a time conversion algorithm, real time is output, and the accuracy of time identification is improved; time expression extraction, SQL generation and time correction can be completed by calling the large model at a time, the process is simple, and the efficiency is high.