The collocation interface makes it possible to search for sentences that contain a group of at least two predefined lexemes. This page is called by clicking the link "Collocations" on the details page. The lexeme selected on the details page is the first "seed" and determines the selection of the remaining words.
The first line on the collocation page contains the lexemes that constitute the current search path. The listbox "Neighbouring lexemes" right below the first line displays the lexemes that are found in phrases containing the current search path, and the sentences that contain the current search path are shown at the bottom of the page.
An example may help to explain how to use this page:
Go to the query page, search for rāma, and open the details page for rāma by clicking on the linked text.
At the top of the details page for rāma, click the link "Collocations", which opens the collocations page. The current search path contains only the word rāma, and the listbox contains all lexemes that are found in the same sentences as the lexeme rāma.
Select the lexeme akliṣṭa from the listbox. Now, the search path is rāma + akliṣṭa, and the listbox contains fewer entries, because fewer sentences contain rāma and akliṣṭa. In addition, all sentences that contain rāma and akliṣṭa are shown at the bottom of the page.
Continue by selecting other lexemes. Each time, the search path is increased by one lexeme, and fewer occurrences are displayed at the bottom of the page.