Federated search is an approach to information retrieval that aggregates query results from multiple information sources. Federated search allows a user to submit a single query and receive results from multiple systems, whether the data resources are stored on-premises or with a cloud service provider, without having to query each of the systems individually. The word federate means “to join together” or “to unite.”
A federated search engine goes out to multiple data sources that belongs to the federation simultaneously, providing the end user with real-time query results taken directly from the required source of information. Query results can be integrated so they appear as if they are from one source, or they can be displayed in separate sections of the same search results page (SERP). Typically, federated search engines use application programming interfaces (APIs) to access multiple data sources and connector code to return query results in a consistent manner. It can be difficult, however, to rank the results from disparate sources in real time. This requires the application of multiple predictive algorithms, a concept known as deep learning.
Some federated search programs are able to integrate search from federated and non-federated sources. In such a scenario, the federated search application creates indexes for non-federated sources and searches them along with the native indexes of the federated information sources. Most often this type of hybrid federated search is carried out by querying indexes that are actually snapshots of the data sources that are taken on a periodic basis.
Federated search may also be referred to as distributed search, integrated search cross-database search or universal search.
Continue Reading About federated search (universal search)
- Perceptive's federated search features allow users to retrieve results from Twitter, LinkedIn, Facebook and other social media platforms