Ver 2.x

All information is interconnected and regenerated as useful data.

KNOWLEDGE STUDIO is a studio for effectively constructing and managing knowledge inside of your business and company. Making semantically connected asset data more usable for consumers' objectives will increase the value of your data.

  • #Risk management
  • #Content recommendation
  • #Knowledge management
  • #Deep QA
  • #Open data
  • #Knowledge engineering

What makes KNOWLEDGE STUDIO so special?

KNOWLEDGE STUDIO is a product that enables intelligent analysis from knowledge graph data generation to management, providing functions such as knowledge extraction, knowledge conversion, data integration, open data, and large-scale inference.

  • Point 01

    Asia's first AI and Graph DB technologies convergence

    and commercialization

  • Point 02

    Proven super-capacity graph

    data processing

  • Point 03

    Powerful built-in

    inference engine

  • Point 04

    Easy and convenient built-in knowledge

    extraction and conversion

  • Point 05

    Contain the largest

    reference in Korea

System Configuration

Core Technology

Graph data conversion

Automatic extraction, transformation, and integration of graph data from structured/unstructured documents

It is necessary to transform and integrate data that is scattered inside and outside of the organization into graph data. This makes it easy for users to analyze and leverage integrated data.

Features

  • 01Data conversion function that supports various formats

    It provides procedures and method for generating graph data and the functions to test the result in advance before conversion and saving the conversion result.

  • 02Data transformation utility

    It provides various utilities for data conversion, such as DB viewer, Excel viewer, text viewer, SPARQL query, schema visualization, and configuration management through SVN integration.

  • 03Powerful data conversion rules

    The RML language provides its own data conversion rules that can convert various types of data to graph data, making it easy to convert and validate graph data.

  • 04Ultra-large data conversion

    It stores, transforms, and augments over 10 billion triples of super-capacity graph data.

Graph data extraction

Knowledge extraction based on graph neural networks

It is possible to simply construct the information that clients demand by extracting knowledge from a range of documents using powerful natural language processing engines and graph neural networks.

Features

  • 01High-quality natural language processing based on deep learning

    Deep learning (artificial neural network) and the most recent machine learning techniques are used for natural language processing in a way that is quicker, more accurate, and of greater quality than traditional methods.

  • 02Knowledge extraction based on neural network graphs

    Analyzes the interconnections and relationships of individual pieces of knowledge, identifies and extracts the most relevant knowledge. The identified knowledge information is used to build information retrieval and knowledge graphs.

Graph data storage and inference

High-speed query processing engine for storing and managing ultra-large graph data

RDF-based graph data can be inferred by structuring the meaning of the data through ontology. KNOWLEDGE STUDIO provides a variety of features, including inference (RDFS, OWL, OWL2) and query services for stored graph data.

Features

  • 01Large-volume graph data saving and inference

    The graph data saving function is the function that parses and validates large-scale graph data promptly. Graph data inference provides optimized inference rules for RDF, OWL, and OWL2.

  • 02Graph data framework

    Offers graph data storage creation and access through the graph data framework. It supports the open-source frameworks Jena and RDF4J.

  • 03SPARQL query

    It provides a graph query language for querying graph data, and SPARQL Endpoint makes it easy for users to query and acquire data.

  • 04Graph data management

    It includes setting and management of the graph data repository, data loading, query management testing, index generation management for analysis, plugins and server monitoring.

  • 05Inference based on various rules

    The graph data inference engine provides a set of rules, such as RDF, OWL-Horst, OWL2-QL, and OWL2-RL, based on a data model that users may fine-tune and utilize.

  • 06Schema update transaction

    It can modify misspelled schemas, quickly perform inferences on existing data that went through schema changes, and reflect them.

  • 07Consistency verification

    It helps ensure consistency in the configuration of graph data by verifying whether that graph data is configured to fit the axiom schema.

  • 08Re-inference

    This function deletes previously reasoned data and re-creates reasoned data in order to secure the consistency of graph data or delete incorrectly reasoned data.

Graph data analysis

Provide real-time reflection and analysis of knowledge that is collected and changed in real time

It contains built-in graph data analysis features including graph index generator, user/time zone/topic/interest-specific knowledge network analyzer, and data storage and analysis indexes and algorithms for external information analysis from social media, email, and knowledge management systems.

Features

  • 01Time series topic analysis

    Topic analysis performs topic analysis for graph data using the automatic keyword extraction and analysis function of the unstructured data analysis engine, and provide results on time series topic analysis.

  • 02Geospatial analysis

    This allows users to apply and utilize the spatial information analysis algorithm in the knowledge graph that contains the spatial data, via SPARQL.

  • 03Knowledge network analysis

    Various network analysis techniques can be used to analyze complex networks to discover new facts and phenomena.

  • 04Knowledge network analysis algorithm

    More than 10 network analysis algorithms, including centrality analysis, connectivity analysis, and cluster analysis, are available through the query language.

Deep QA

Optimized query analysis and real-time answers through knowledge-based in-depth queries and NLU

In order to determine and provide the optimal response, the system learns from a variety of information about the user's inquiry. Users may create a more sophisticated consultation system by using comprehensive and accurate knowledge graph-based Q&A.

Features

  • 01Natural Language Understanding (NLU)

    Accelerating the understanding of user queries, NLU provides the ability to learn and analyze patterns for queries based on knowledge graphs.

  • 02Natural Language Generation (NLG)

    It provides the ability to generate user answers through natural language processing, such as KBQA results and NLU for user queries.

  • 03Knowledge Base Question Answering (KBQA)

    KBQA identifies the key meaning of questions and provides accurate answers through pattern learning of knowledge graph-based user queries.

  • 04QA manager

    It serves as a window for handling user queries and providing answers, and can manage various QA engines such as KBQA and IRQA.

KNOWLEDGE STUDIO TOOL

Professional tool for one-stop processing of graph data conversion to analysis

KNOWLEDGE STUDIO provides all-in-one features among which ultra-fast large-scale graph data storage, ultra-fast queries, and analysis.

  • Graph data conversion

  • Graph data storage and inference

  • Graph data analysis

  • Add-on knowledge extraction

  • Add-on deep QA

Tool Introduction

  • 01Graph data conversion

    • Creates conversion rules quickly and easily using a variety of features such as editing and executing data conversion rules, data sources, and user functions.

      1. Plugin management
      2. Autocomplete user functions for data purification, combining, etc.
      3. Edit transformation rules for validation 능
      4. Data conversion and testing
      5. Various viewer features
      6. Project management
  • 02Graph data storage and inference

    • It supports a variety of graph data formats to store multiple graph data, and provides data inference when stored.

      1. Storage setup and management
      2. Graph data query management and testing
      3. SPARQL query
      4. Graph data management and editing
      5. User management and monitoring
      6. Project management
  • 03KENT.V, Saltlux’s professional information extraction tool

    • Automatic extraction of knowledge for users to learn about knowledge extraction from many forms of knowledge resources

      1. Automatic knowledge extraction task management
      2. Learning model-based knowledge extraction
      3. Inspection of extracted knowledge
      4. Creation, testing, and management of knowledge learning model
      5. Evaluation of automatic knowledge extraction model
  • 04Add-on Deep QA, a knowledge base-based professional Q&A tool

    • In conjunction with the Q&A system, the optimal response is presented to the user based on the knowledge graph

      1. Build, supervise, and manage deep QA knowledge data
      2. Lookup, registration, editing, and management of language resource dictionary
      3. Knowledge base list and instance management
      4. SPARQL query tool

Success Story

  • Integrated data management

    Integrated platform with graph data Intelligent KMS system

    In order to improve the efficiency and usage of knowledge content management, Nonghyup Bank's outdated knowledge management system was restructured and refreshed into an intelligent knowledge management system that adheres to the latest AI technology trends.

    • Enhanced ease of use such as knowledge generation, management, and search
    • Provided accurate answers through linkage with AI consultation QA
  • Establishment of open data

    Establishment of a linked data (LOD) service Customized IP-Biz information-sharing platform

    An information channel that can be used in linkage with various business information possessed by other government departments and public institutions based right information was established, allowing employees in small and medium enterprises and start-up companies to obtain information promptly and conveniently.

    • Opens more than five million cases of domestic and overseas patent information.
    • Provides three million cases of patent analysis DB linkage information from thirteen institutions.
  • Integrated data management

    Knowledge-based product information platform with real-time product monitoring and analysis

    Online product data from first- and second-tier financial firms were gathered, and after data extraction and purification, an integrated product database was created. This database was then transformed into a knowledge graph and archived for use as a Q&A service.

    • Through monitoring of financial product information and related product analysis, data usability and internal staff productivity were improved.
  • In-depth Q&A service

    General knowledge information provided by KT GiGA Genie equipped with deep QA

    KT GiGA Genie back-end Q&A service was applied to provide general domain, life information, general knowledge, and professional deep QA with a 94% accuracy rate through continuous learning.

    • AlaaS-based customized AI service implementation provided easy and convenient service and customized information to customers
  • NH Nonghyup Bank
    Integrated data management

    Integrated platform with graph data Intelligent KMS system

    • Enhanced ease of use such as knowledge generation, management, and search
    • Provided accurate answers through linkage with AI consultation QA
  • Patent Office
    Establishment of open data

    Establishment of a linked data (LOD) service

    • Opens more than five million cases of domestic and overseas patent information.
    • Provides three million cases of patent analysis DB linkage information from thirteen institutions.
  • KOOKMIN BANK
    Integrated data management

    Knowledge-based product information platform with real-time product monitoring and analysis

    • Through monitoring of financial product information and related product analysis, data usability and internal staff productivity were improved.
  • KT
    In-depth Q&A service

    General knowledge information provided by KT GiGA Genie equipped with deep QA

    • AlaaS-based customized AI service implementation provided easy and convenient service and customized information to customers

Reference

  • Establishment of open data

    National Library of Korea

    LOD-based valuable old newspaper platform

  • Social and knowledge data

    Busan Human Resources Development Institute

    Busan-related experts and intellectuals Integrated DB-based social and knowledge network platform

  • Establishment of open data

    ETRI

    AI open API/DATA service

  • Establishment of open data

    Korea Culture Information Service Agency

    Convergence and open DB for LOD-based cultural prosperity

  • Busan International Film Festival

    Busan Metropolitan City

    Knowledge graph of Busan's cultural information data and a LOD disclosure platform

  • Anthropogeography information and search service

    National Geographic Information Institute

    Ontology technology modeling for anthropogeographic information and complex search services

  • LOD public portal

    K-water

    Industrial ecosystem that enables private use of water resources, map data, and public LOD

  • Establishment of open data for Gyeonggi Provincial Office

    Gyeonggi-do

    LOD service for search and data utilization of Gyeonggi Provincial Office management data