Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A Dynamically Scalable Real-time Stream Labeling System

A real-time streaming and labeling technology, which is applied in unstructured text data retrieval, special data processing applications, text database clustering/classification, etc., can solve the problem that the label system cannot be updated and expanded online, and supports horizontal expansion and configuration Manage concise effects

Active Publication Date: 2020-10-30
HUNAN ANTVISION SOFTWARE
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problem that the above-mentioned labeling system cannot be updated and expanded online, the purpose of the present invention is to provide a dynamically scalable real-time streaming labeling system, which can support online dynamic updating and expansion of the labeling model, so as to meet the real-time streaming service requirements of the SaaS system 7×24 hours demand

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A Dynamically Scalable Real-time Stream Labeling System
  • A Dynamically Scalable Real-time Stream Labeling System
  • A Dynamically Scalable Real-time Stream Labeling System

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment 1

[0039] refer to image 3 As shown, specifically explain the process of language extension:

[0040] In the first step, the user configures new language features to the language model through the configuration manager; after the model loader detects the update of the language model, it dynamically loads the updated information into the runtime;

[0041] Step 2, create a new tag chain for the new language;

[0042] Step 3, select a tag model to add to the tag chain;

[0043] Step 4, specify model parameters and other information for the newly added label model;

[0044] Step 5, if the user needs to add more tag models to the tag chain, return to step 3; otherwise, save the tag chain configuration information; after the model loader detects that the tag chain configuration information is updated, it dynamically loads the updated information to the runtime.

specific Embodiment 2

[0046] refer to Figure 4 As shown, specifically explain the process of label model update:

[0047] In the first step, the user selects a label model through the configuration manager and updates the label model information; after the model loader detects the update of the label model, it dynamically loads the updated information to the runtime;

[0048] Step 2, the configuration manager takes out the tag chains related to the tag model (that is, the tag chain containing the tag model), and adds these tag chains to the queue;

[0049] Step 3, judge whether the tag chain queue is empty, if it is empty, it will end, and complete the process of updating the tag model this time; if it is not empty, take out the first tag chain in the tag chain queue;

[0050]Step 4: Ask the user whether to update the tag chain information. If no update is required, delete the tag chain from the above tag chain queue and return to step 3; otherwise, wait for the user to update the corresponding t...

specific Embodiment 3

[0052] refer to Figure 5 As shown, specifically explain the process of tag chain update:

[0053] Step 1: The user selects a label chain through the configuration manager and updates the configuration information of the label chain (the update operation includes adding or deleting the label model of the label chain, or modifying the configuration information of one or more label models of the label chain) ;

[0054] Step 2: After the model loader detects that the tag chain configuration information is updated, it dynamically loads the updated information into the runtime.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a dynamically expandable real-time stream label frame. The dynamically expandable real-time stream label frame comprises three logic layers which include a model layer, an operation layer and a configuration management layer. The model layer is used for describing and storing model information related to labels; the operation layer is used for detecting updating of configuration information of models and label chairs when the real-time stream label frame is operated, and the detected updating information is dynamically loaded to the related labels and the related labelchains; the configuration management layer is used for providing a configuration management interface for the models and the label chains for a user, and therefore the model and the label chains are updated and expanded. With the dynamically expandable real-time stream label frame, dynamical configuration, management, loading and updating of the labels are realized, it supports dynamical expandingof language types, the labels and the label chains and transverse expanding of a label system, and therefore the availability, expandability and maintenance of real-time stream label service are substantially improved.

Description

technical field [0001] The invention relates to the technical field of distributed computing, in particular to a dynamically scalable real-time stream labeling system. Background technique [0002] With the rapid development of the Internet and mobile Internet, people face a large amount of various text information every day, including news, Weibo, posts and so on. In order to improve people's reading efficiency, it is necessary to label the text information appropriately, which can effectively reflect the theme and core content of the text information, so as to help people quickly identify and filter out the content they are interested in. [0003] At present, the commonly used automatic labeling methods for text information are mainly implemented based on machine learning. These methods need to train appropriate label models on the basis of a large number of sample data, and realize labeling of text information based on these label models. [0004] Due to the ever-changin...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/35
CPCG06F16/35
Inventor 舒琦王晓斌黄三伟
Owner HUNAN ANTVISION SOFTWARE
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com