Abstract
Learning networks have become extremely powerful tools for data classification. One drawback in many of these applications, however, is that they require a large dataset for efficient performance. Medical imaging, for example, cannot make large enough databases available due to strong data privacy concerns and high data collection costs. Fortunately, data augmentation can provide a solution to improve the classification performance of these systems. This paper proposes a novel method for data augmentation based on the compressive sensing technique. Our approach uses sensing matrices to manipulate data features into different spaces in a compressed form, which are then combined and used as extensions of the original limited dataset. This new feature representation is allowing for a more robust learning network that thereby enables improved data classification. When applying our approach to binary-class and multi-class datasets, results showed that our approach can significantly improve the classifier performance and enhance classification results for both datasets.
Highlights
We propose a novel method for data augmentation (DA) that can be applied to various forms of data for many applications including image and text classification
The dataset was divided into four groups based on Prostate Specific Antigen (PSA) level: 62 samples with no evidence of prostate cancer, 189 samples with benign presence, 25 samples with presence at a PSA level of 4-10 ng/mL, and 42 samples with presence of prostate cancer at a PSA level of >10 ng/mL
In this study we proposed a powerful and flexible data augmentation model based on sensing data features in a compressing domain, which improves the performance and the efficiency of the learning network when compared to the original data and other scenarios for binary and multiclass datasets
Summary
ANNs have been used in a vast array of applications, including pattern recognition, prediction problems, and classification. This study focuses on the application of ANNs to an example situation where achievement performance is low due to a shortage of samples that are required for learning. A. Awedat: Novel Robust Augmentation Approach Based on Sensing Features. Low dimensions can be sensed by using an generated sensing matrix. For a signal x Rn and the sensing matrix φ Rm×n, this signal can be sensed in low dimension as in (3) [9]. The reconstruction compressed data problem is to solve non-deterministic polynomial (NP-hard) problem to find the sparse solution. We will apply CS only to manipulate data features in low dimensions and use them for classification.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
