From an end-user perspective, BioClass is a platform that focuses on the application of reasoning models in text classification issues. It is designed to work with the results obtained from an information retrieval process in a text database, where documents can be relevant or not for a specific issue. BioClass takes this data as its input and offers multiple filtering and machine learning algorithms to handle the automatic classification problem.

From a developer perspective, BioClass offers an abstraction layer facing the classification process. Thanks to it, the developer can use its architecture and implement new reasoning models.

The TCBR-HM Model is a hybrid system proposed by our research group that aims to classify documents according to their content. It combines a text classifier that is based on Hidden Markov Models (HMM) with a Case Based Reasoning (CBR) approach.

TCBR-HMM is integrated into the BioClass framework. The classifier offers a simple and parametrizable structure, based on word's relevance, which allows the model to be adjusted to future documents.