Learning, Rule, Environment and overall planning of a learning sysstem
The task of learning from environment (what forms environment - www is one for nell) should be specified. It requires the learner (in this case, could be nell) to infer the laws of the environment in terms of its percepts (explain the percepts in details here) and actions, and use the laws to solve problems (what kind of problems?).
Based on research on problem space creation and discrimination learning, first design an approach in which exploration, rule creation and rule learning are coordinated in a single
framework. With this approach, the system Nell creates certain rules by noticing the
changes in the environment (web in our case) when actions (updating the web, linking in more info) are taken, and later refines the rules by explaining the failures of their predictions. Unlike
many other learning systems, nell treats learning and problem solving as interleaved activities, no training instance nor any concept hierarchy is necessary to start learning. Furthermore,
the approach is capable of discovering hidden features from the environment when normal discrimination process fails to make any progress.