13 results about "Match rate" patented technology

A high-speed matching-engine architecture is disclosed that sustains deterministic sub-microsecond latency while processing more than 10 million order messages per second per core on commodity multi-core processors. Orders reside in cache-aligned Data Holder Nodes whose occupancy and price-level boundaries are tracked with constant-time bitmask operations, eliminating pointer-chasing penalties. Per-core huge-page pools, SIMD copy kernels, and lock-free, cache-line-aligned queues further minimize TLB misses and coherence overheads. Overflow is handled by Push Back / Push Forward cascades that relocate the least- or most-prioritized orders between adjoining nodes without violating price-time priority. Node capacities vary monotonically with book depth and are re-tuned online by a lightweight machine-learning controller that maximizes cache-hit probability under changing market micro-structure. The design tightens spreads, raises match-rate revenue, and complies with stringent regulatory latency caps using standard x86-64, Arm, or other architectures.

This invention discloses a method for mining alumni information based on a large-scale model, comprising the following steps: 1) acquiring raw alumni-related data; 2) preprocessing the acquired data; 3) performing natural language processing on the acquired data; 4) semantic parsing and information extraction using a large-scale model; 5) constructing a knowledge graph; storing the extracted alumni information and relationships in a graph database; and constructing a visualized alumni knowledge graph; 6) updating the knowledge graph data; and 7) providing intelligent retrieval and question-answering services for alumni information based on the knowledge graph data. This invention uses a large-scale model to perform deep semantic understanding and information extraction from unstructured text. Compared to traditional keyword-based or small-scale model-based methods, it can more accurately identify alumni identities, professional experiences, and social relationships, while reducing the false matching rate.

To provide a technique capable of appropriately setting a recipe of substrate processing by automatically improving setup data.SOLUTION: The image processing system processes at least one of image information of the semiconductor device and image information of the probe card. The image processing system includes a calculation unit that calculates information related to a matching rate between a comparison image which is the acquired image information and a registered image held in advance, a comparison unit that compares the calculated information related to the matching rate with a threshold value held in advance, and a registration unit that adds the comparison image to the registered image to update the registered image to a new registered image in a case where the information related to the matching rate is less than the threshold value, and does not add the comparison image to the registered image in a case where the information related to the matching rate is equal to or greater than the threshold value.SELECTED DRAWING: Figure 6

A system and method for improving identity resolution match rates using deep learning techniques that enable privacy-preserving fuzzy matching of personally identifiable information (PII) employs a deep learning model trained with transformer architecture and contrastive learning on third-party identity graph data. Custom tokenizers process different Pll types by leveraging hierarchical structures and domain-specific characteristics. The trained model generates vector embeddings that enable fuzzy matching to account for variations in spellings, typographical errors, and data inconsistencies without requiring adherence to strict data schemas. A vector database stores embeddings for nearest neighbor searches to identify potential identity matches based on distance calculations between embeddings. A match filter applies logic to determine valid matches, and the system can operate without requiring movement of raw Pll data from secure environments.

The invention discloses a MoE model intensified training method based on classification data, and relates to the technical field of MoE model intensified training, and the method comprises the following steps: collecting the correctly distributed number of various samples in each batch in the MoE model intensified training process, and carrying out the preprocessing to obtain sample correct rate data and expert matching rate data; performing sample accuracy analysis to obtain first weight feature data; performing expert matching rate analysis to obtain second weight feature data; based on the first weight feature data and the second weight feature data, dynamically adjusting the proportion of various samples in the batch of enhanced training; the method is used for solving the problem that when an existing MoE model intensive training technology carries out intensive training on a MoE model used for classifying data, the proportion of various samples in each training batch cannot be dynamically adjusted according to the number of correctly distributed samples in each training batch, the efficiency of intensive training of the MoE model is improved, and meanwhile the training efficiency of the MoE model is improved. And the classification precision of the MoE model is improved.

The application belongs to the field of roads. The purpose of the application is to solve the problems of complex calculation and low matching rate with sensor node position in the existing deployment of edge computing device nodes, obtain all sensor nodes to be covered on the road, divide all sensor nodes into layers by using KD-Tree space, draw a circle with each sensor node as the center according to a preset radius, sort the coverage potential of all circles in each layer from high to low, start searching from the first circle in each layer, set the position of the edge computing device according to whether the first circle intersects with other circles, the coverage potential and the coverage score, and complete the deployment of all edge computing devices according to the query order of the circle and the deployment method. The application is used for deploying edge computing devices.

In the present invention, a background detection unit (15b) detects a background region in a template image for image matching. A calculation unit (15d) excludes the detected background region and calculates the match rate between the template image and a retrieved image.

The invention discloses a multi-agent perception and cognition generation and API (Application Program Interface) calling multi-modal evaluation method. The method comprises the following steps of: 1, downloading and deploying a standardized data set to the local; 2, evaluating the perception ability of the multi-agent, including perceiving ten sub-tasks, and calculating scores by using the single-graph single-question accuracy and the single-graph double-question full-match rate as perception indexes; 3, calculating scores by adopting the single-graph single-question accuracy rate and the single-graph double-question full-match rate as cognitive indexes, and outputting MME basic scores in combination with the perception scores; 4, generating and evaluating a multi-agent picture, extracting a target keyword, detecting a to-be-evaluated image to obtain an actual object category set, and calculating an image element matching score in combination with matching of the target keyword and an object category; 5, executing multi-agent API calling validity evaluation, and outputting an API calling matching degree score; and 6, multi-dimensional scores are fused, a comprehensive evaluation result is output, and evaluation of the multi-modal multi-agent large model is realized.

This application discloses a method, apparatus, storage medium, and computer equipment for matching reviewers, relating to the field of office automation. This application calculates the capability scores of all reviewers corresponding to the current unit, then calculates the required capability score of the current unit, and finally calculates the matching rate between each reviewer and the current unit. The reviewer with the highest matching rate is selected as the optimal reviewer for the current unit. This method of selecting the optimal reviewer based on the matching between the capability scores of the reviewers in a unit and the required capability score of the current unit reduces the workload of selecting the optimal reviewer for each unit, greatly improving review effectiveness and efficiency.

The present invention relates to an improved method of increasing the speed of the cosine calculations. Preferably the improved methodology comprises a GPU-accelerated cosine similarity implementation for Tandem Mass Spectrometry (MS / MS) with approximately 1000-fold speedup compared to the MatchMS reference at a rate of 0.005% incorrect matches and a rate of 0.002% incorrect scores.

The present embodiments relate to a matching management server comprising: a processor; a memory for storing a program executed by the processor; and a communication interface for transmission and reception with user devices and business operator devices, wherein the communication interface receives application image information and application information of users from the user devices, receives preferred person information including artist image information and activity information of artists affiliated with business operators from the business operator devices, and transmits matching results to each of the user devices and the business operator devices, and the processor generates matching results including a business operator-specific matching rate for each user by applying the preferred person information, the application image information, and the application information to a matching model.

This invention relates to the field of computer simulation and vocational skills assessment technology, and discloses an automatic assessment method for a simulation system applicable to vocational qualification certification. The invention establishes a simulation operating environment, configuring a distributed system including multiple candidate workstations, an assessment server, and a simulation model. Simultaneously, a set of benchmark parameters for the current assessment is set in the assessment server. Based on the benchmark parameter set, a fault template is randomly selected from a question bank, and specific test question data is dynamically generated in conjunction with the current simulation conditions. The test question data is distributed to designated candidate workstations. The system collects the sequence of operational events during the candidate's examination and the trajectory of the working conditions output by the simulation model in real time, thereby scoring the candidate. The scoring dimensions are based on two dimensions: the candidate's operational step matching rate and the final working condition deviation, making the candidate's final score more objective and realistic. An assessment report is generated based on the candidate's final score.

The present invention relates to automatically and optimally setting a meeting schedule using the AI. The present invention further relates to recognizing actual meeting attendees and evaluating the attendance and behavior of meeting participants. A first step involves accessing a candidate database (DB) server that includes at least one of a list of a potential meeting's group participants or personal contact, address, current location, work schedule, team information or expertise of an individual potential meeting participant. A second step involves accessing a meeting information DB server that includes a potential meeting information including at least one of the potential meeting's expected agenda, expected number of participants, expected meeting time or expected meeting location for the potential meeting. A third step involves calculating a match rate between a first data from the candidate DB server and a second data from the meeting information DB server.