This research is focused on the development of an advisable agent platform that performs real-time information monitoring using machine learning techniques. This advisable agent is based on an approach which combines reinforcement learning with data flow-based analysis methods. The machine learning model proposed relies on prior knowledge, reinforcement learning and vector-based data analysis techniques. Relying on advice, a knowledge-based application is easily configured with an initial knowledge set which is then incrementally improved using rules, advice and induction. These machine learning capabilities are integrated with a real-time data analysis model which supports data filtering, extraction and monitoring for items of interest. This foundation for time-critical event processing and time series data analysis is derived from a data stream perspective that abstracts a series of discovery, delivery or learning events as a data flow. This data flow processing model may ultimately result in a number of potential benefits including; efficiency, scalability, and ease of deployment.