Abstract:
Representative selection involves selecting few informative exemplars from large datasets. Data selection methods rarely handle non-linear data structures, sample concise and non-redundant subsets, reject outliers, and produce interpretable results.
This paper introduces MOSAIC, a representative selection method for descriptive sketches of arbitrary manifold structures. MOSAIC uses a novel quadratic formulation to select samples that maximize global representation power, minimize redundancy, and reject disruptive information by detecting outliers.
Theoretical analyses show that the sampled representatives maximize data coverage in a transformed space and characterize the sketch geometrically. We also present a highly scalable randomized implementation of the proposed algorithm that accelerates.
Extensive experiments on real and synthetic data with comparisons to state-of-the-art algorithms show MOSAIC’s superiority in achieving the desired characteristics of a representative subset all at once while being robust to various outlier types.
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
