Accuracy Performance Degradation in Image Classification Models due to Concept Drift

Abstract

Big data is playing a significant role in the current computing revolution. Industries and organizations are utilizing their insights for Business Intelligence by using Deep Learning Networks (DLN). However, dynamic characteristics of BD introduce many critical issues for DLN; Concept Drift (CD) is one of them. CD issue appears frequently in Online Supervised Learning environments in which data trends change over time. The problem may even worsen in a BD environment due to the veracity and variability factors. The CD issue may render the DLN inapplicable by degrading the accuracy of classification results in DLN which is a very serious issue that needs to be addressed. Therefore, these DLN need to quickly adapt to changes for maintaining the accuracy level of the results. To overcome classification accuracy, we need some dynamical changes in the existing DLN. Therefore, in this paper, we examine some of the existing Shallow Learning and Deep Learning models and their behavior before and after the Concept Drift (in experiment 1) and validate the pre-trained Deep Learning network (ResNet-50). In future work, this experiment will examine the most recent pre-trained DLN (Alex Net, VGG16, VGG19) and identify their suitability to overcome Concept Drift using fine-tuning and transfer learning approaches.

Authors and Affiliations

Manzoor Ahmed Hashmani, Syed Muslim Jameel, Hitham Alhussain, Mobashar Rehman, Arif Budiman

Keywords

Related Articles

MOSIC: Mobility-Aware Single-Hop Clustering Scheme for Vehicular Ad hoc Networks on Highways

As a new branch of Mobile ad hoc networks, Vehicular ad hoc networks (VANETs) have significant attention in academic and industry researches. Because of high dynamic nature of VANET, the topology will be changed frequent...

AMBA Based Advanced DMA Controller for SoC

This paper describes the implementation of an AMBA Based Advanced DMA Controller for SoC. It uses AMBA Specifications, where two buses AHB and APB are defined and works for processor as system bus and peripheral bus resp...

Triple SV: A Bit Level Symmetric Block-Cipher Having High Avalanche Effect

The prolific growth of network communication system entails high risk of breach in information security. This substantiates the need for higher security for electronic information. Cryptography is one of the ways to secu...

A Multiple-Criteria Decision Making Model for Ranking Refactoring Patterns

The analytic network process (ANP) is capable of structuring decision problems and finding mathematically determined judgments built on knowledge and experience. Researches suggest that ANP can be useful in software deve...

Variability Management in Business-IT Alignment: MDA based Approach

The expansion of PAIS (Process Aware Information Systems) has created the need for reuse in business processes. In fact, companies are left with directories containing several variants of the same business processes, whi...

Download PDF file
  • EP ID EP578628
  • DOI 10.14569/IJACSA.2019.0100552
  • Views 113
  • Downloads 0

How To Cite

Manzoor Ahmed Hashmani, Syed Muslim Jameel, Hitham Alhussain, Mobashar Rehman, Arif Budiman (2019). Accuracy Performance Degradation in Image Classification Models due to Concept Drift. International Journal of Advanced Computer Science & Applications, 10(5), 422-425. https://europub.co.uk/articles/-A-578628