The ALE Learnset Manager (ALM) is an application utilizing a trainable engine (Automated Learning Engine - ALE) in order to automate processing of documents. ALM is a web-based administration client that enables you to capture, prepare, and manage training documents that are to be learned by the Automatic Learning Engine. Based on what the engine learns from the training documents (learnsets), ALM improves extraction of document header fields, eliminating user intervention.
You can configure BFT to use ALM as an additional extraction resource helping improve header field extraction results for transcripts. Performing verifier actions such as, correcting fields configured for ALM learning, improves the learning process. After a configured header field is modified in Verifier, the associated transcript is added as a training document in ALM, to be learned. Once learned, BFT compares the ALM extraction candidates against the BFT candidates and selects the best candidate. This integration can help improve BFT’s extraction accuracy, while reducing repetitive verifier corrections.
You can use Solution Configuration Manager (SCM) to configure the ALM integration for BFT and define the fields which are to be configured for learning.