The current version of the corpus is searchable via the NoSketchEngine[1]. The mediolatin texts are lemmatised thanks to the parameters developed in the framework of the Omnia project, slightly modified by us for the management of named entities.
It is thus possible to query both forms (aquarum), lemmas (aqua) and expressions (sicut aqua extinguit ignem). NoSketchEngine also allows the use of the CQL language (originally developed for the Corpus Workbench), which allows complex queries on forms/lemas, but also syntax (spacing between terms, search for a particular grammatical category: nouns, verbs, qualifiers, etc.). The software also permits investigations on the collocations of a given term or expression, as well as complex sorting on the metadata. An example of potential queries on CEMA will arrive very soon[2].
Also, an API is under development. It will allow the same queries and more.
[1] We would like to thank Krzysztof Nowak, as well as the Latin Lexicography-Historical Semantics team of the IRHT, for their help in integrating the corpus into this research tool.
[2] It should be noted, however, that the search (by forms or lemmas) is performed according to a simplified spelling: « u » for « v » and « i » for « j », in conformity with the parameters of the Omnia lemmatizer.