Relativity Analytics
Relativity embeds a latent semantic indexing (LSI)-based search engine, which powers the Relativity Analytics feature set. Unlike traditional keyword search engines, which are essentially static lists of all words found in a collection, the LSI-based concept search engine converts the textual content of documents into vector mathematics and creates a multi-dimensional mathematical model. This model identifies how the documents in a database relate to each other based on the syntax and frequency of all the words in all the documents, rather than on a single keyword.
Search results depend on how and where ideas and concepts intersect with similar ideas and concepts in a document collection. The Relativity Analytics feature set allows for clustering, categorization, foreign language detection and concept search techniques, all of which are fully integrated into the software and easy to use. Use Relativity Analytics to accelerate a relevance review or traditional linear review by sorting the discovery documents into conceptually similar subsets. Verify the consistency of document review decisions and provide another level of assurance that privilege documents have been properly identified. Using Relativity Analytics throughout the review workflow accelerates traditional review processes and easily identifies documents that other search techniques routinely miss.
Benefits of Relativity Analytics include:
- Accelerated traditional linear review
- Immediate overview and context of document batches assigned to reviewers
- Powerful keyword sampling
- Quick identification of documents that have virtually the same content
- Elimination of the issue of a single keyword having multiple meanings
- Ability to make bulk decisions and assignments
- Automated quality control
- Grouping documents by their primary language