Silk DataSemantic Framework

Silk Data Semantic Framework is an innovative set of methods and tools for next-gen intelligent text processing. It doesn’t require any human involvement and prior knowledge of the area to be explored and simplifies the work of a domain expert. The process is fully automated, making text processing fast, predictable and risk-free.

Text Challenge

We are surrounded by a vast amount of unstructured text data, such as books, documents, news, which are increasingly difficult to navigate.

We need tools to analyze and classify texts automatically, find important parts in them and extract knowledge. In this case, the Silk Data Semantic Framework helps address these main challenges.

How can Semantic Framework help you?

The framework takes over the routine work of preliminary semantic analysis of documents.

How can Semantic Framework help you?

Text Challenge

Components of Semantic Framework

  • Semantic Classification – a special method of document categorization according to the content
  • Document Similarity – a search tool that finds documents that are contextually close in their meaning to the preset document
  • Semantic Segmentation – a tool that divides a document into parts (or segments) according to content meaning
  • Semantic search is an effective text search method that factors in the content of documents
  • Domain Visualization is a visual map that helps explore the structure of knowledge hidden in texts

How it Works

Use cases

Due Diligence and eDiscovery​

Retrieving documents that need to be transferred to an authorized employee for further analysis from a large array of data by using semantic classification.

Analysis of document priority.An employee should first look at the documents that are likely to contain the requisite information. The framework is able to find those documents that are closest in their meaning to the documents selected by the expert by using semantic similarity.​

Business and Finance​

Contract Intelligence. Highlighting text sections that are important for the user and quickly navigating through them by using document segmentation. It can be used for documents such as contracts, insurance products, companies' annual reports, audit reports etc.​


Semantic search. Traditional search methods become ineffective when a large number of documents are under analysis. Search efficiency can be increased through preliminary semantic analysis of texts: this enables a user to apply additional limitations (context filters) that cannot be used with traditional algorithms.

Research and Development​

Exploratory Search with Domain Visualization. Bird’s eye view of knowledge domain helps to find unexpected relationships by providing previously hidden insights and speeds up discovery.

Contact us