Abstract:
Human concept learning is semi-supervised learning (SSL) because people naturally combine a few labeled data with lots of unlabeled data to make classification decisions. Human concept learning is both static and dynamic.
To accommodate new data, classical SSL algorithms must be redesigned. Concept-cognitive learning, which mimics human cognition, can implement dynamic processes. Many SSL methods were designed using instance feature vector information, ignoring concept structural information, which is crucial to human knowledge organization.
This idea inspired the semi-supervised concept learning method (S2CL) for dynamic SSL, which uses concept spaces to represent knowledge as hierarchical concept structures. We also propose S2CLα for concept learning to maximize global and local conceptual information.
To effectively exploit unlabeled data, this paper first presents some new related theories for S2CL (or S2CLα) based on a regular formal decision context, then designs a new SSL framework and its algorithm. Finally, we test our methods, including concept classification and incremental learning on large amounts of unlabeled data, on various datasets.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
