K-Link Asia

How K-Link works

K-Link is a service oriented to document management within an organization, and at the same time enables creation of a community of practice between organizations involved in a specific topic.

It's like having your private organization search engine that targets valuable materials from your community.

With this goal in mind, the K-Link Architecture has a distributed component called K-Link Network, including a set of services that can run inside your institutions.

The overall K-Link Architecture follows the approach of Service Oriented Architecture, where different components communicate together to offer a comprehensive service.

The Service Oriented Architecture allows for increased scalability and flexibility in the setup and deployment of services. In addition, each service runs in a sandbox to minimize the risk of a compromised service damaging the system.

Some services are deployed inside the institution/organization to preserve data ownership and establish a degree of offline access. Varying levels of user-access can be defined. Other services offered to the organization, such as the public network, are deployed externally.

K-Link Components

The K-Link Service oriented architecture is composed by four main components: a document management system, K-Core, a search engine and a geo-location extraction. These components enable the K-Link to perform searches, organize documents and respond to the K-Link goals previously identified.

Document Management System (K-DMS)

  • Guarantees secure online access to your information
  • Allows organizing your information using documents collections
  • Minimizes forest of files (sensitive to identical content)
  • Uses content-based search and filters to accelerate information retrieval
  • Allows publishing information from the system (to your website and Public Network)


  • The base service for interacting with the search engine
  • Exposes an API for performing full-text indexing and retrieval of a document

Search Engine

  • Is the heart of the retrieval capabilities of the K-Link
  • Uses Apache Solr
  • Identifies the language of a document based on probability and an index of known words

Geolocation Entity Recognition