Continuing my long held interest in Machine Learning, I came across a great lectuire by Eric Xing, of Carnegie Mellon University, in which he provides the most clearest definition of Machine Learning, in my opinion, yet:
Machine Learning seeks to develop therios and computer systems for
- representing;
- classifying, clustering and recognizing;
- reasoning under uncertainty;
- predicting;
- and reacting to
- …
complex, real world data, based on the system’s own experience with data, and (hopefully) under a unified model or mathmatical framework, that
- can be formally charcterized and analyzed
- can take into account human prior knowledge
- can generalize and adapt across data and domains
- can operate automatically and autonomously
- and can be interpreted and perceived by [a] human
You can view his leacture here.
Inside Information Systems 