Shefali Arora

Bio: Shefali Arora is an academic researcher from Netaji Subhas Institute of Technology. The author has contributed to research in topics: Deep learning & Biometrics. The author has an hindex of 5, co-authored 19 publications receiving 58 citations. Previous affiliations of Shefali Arora include Thapar University.

Fingerprint Spoofing Detection to Improve Customer Security in Mobile Financial Applications Using Deep Learning

Abstract: Online banking and financial services using mobile applications are seeing a persistent growth among customers, who are using these for their financial transactions. This rise in the use of such applications in smart devices has increased security concerns. There is need for secure mechanisms to prevent fraud and protect personal information. This paper investigates the use of biometric identification in banking and financial services, which leverage the use of smartphones and tablets. While customer engagement and brand loyalty are important concerns, these services are making use of biometric authentication to make customer interactions more secure. However, as technology is growing rapidly, spoofing attacks are becoming common. In this paper, authors have proposed a robust framework to detect spoofing attacks in fingerprint recognition. The process of spoofing detection involves contrast enhancement using histogram equalization and a deep convolutional neural network architecture. Authors have validated the results on various biometric spoofing benchmarks, each one containing real and spoofed samples of user fingerprints. The results indicate that our proposed framework performs better as evaluated against other existing pre-trained CNN models and state-of-the-art methods.

Automated Cryptocurrencies Prices Prediction Using Machine Learning

Abstract: Currently, Cryptocurrency is one of the trending areas of research among researchers. Many researchers may analyze the cryptocurrency features in several ways such as market price prediction, the impact of cryptocurrency in real life and so on. In this paper, we focus on market price prediction of the number of cryptocurrencies based on their historical trend. For our study, we tried to understand and identify the daily trends in the cryptocurrency market which analyzing the features related to the price of cryptocurrency. Our dataset consists of over nine features relating to the cryptocurrency price recorded daily over the period of 6 months. We applied some machine-learning algorithms to predict the daily price change of cryptocurrencies.

A Computer Vision System for Iris Recognition Based on Deep Learning

Abstract: Biometric systems are playing an important role in identifying a person, thus contributing to global security. There are many possible biometrics, for example height, DNA, handwriting etc., but computer vision based biometrics have found an important place in the domain of human identification. Computer vision based biometrics include identification of face, fingerprints, iris etc. and using their abilities to create efficient authentication systems. In this paper, we work on a dataset [1] of iris images and make use of deep learning to identify and verify the iris of a person. Hyperparameter tuning for deep networks and optimization techniques have been taken into account in this system. The proposed system is trained using a combination of Convolutional Neural Networks and Softmax classifier to extract features from localized regions of the input iris images. This is followed by classification into one out of 224 classes of the dataset. From the results, we conclude that the choice of hyperparameters and optimizers affects the efficiency of our proposed system. Our proposed approach outperforms existing approaches by attaining a high accuracy of 98 percent.

Handwriting recognition using Deep Learning in Keras

Abstract: Nowadays, deep learning is playing an important role in the domain of image classification. In this paper, a Python library known as Keras, is used for classification of MNIST dataset, a database with images of handwritten images. Two architectures – feed forward neural networks and convolutional neural networks are used for feature extraction and training of model, which is optimized using Stochastic Gradient Descent. This paper gives an overview of multi-class classification of these images using these models, and their performance evaluation in terms of various metrics. It is observed that convolutional neural networks achieve a greater accuracy as compared to feedforward neural networks for classification of handwritten digits.

Challenges and opportunities in biometric security: A survey

Abstract: Biometric systems identify individuals based on unique traits such as the face, fingerprints, iris etc. The main objective of the study is to understand the role of deep learning in the process of ...

MORPH: A Longitudinal Image Database of Normal Adult Age-Progression.

Abstract: This paper details MORPH a longitudinal face database developed for researchers investigating all facets of adult age-progression, e.g. face modeling, photo-realistic animation, face recognition, etc. This database contributes to several active research areas, most notably face recognition, by providing: the largest set of publicly available longitudinal images; longitudinal spans from a few months to over twenty years; and, the inclusion of key physical parameters that affect aging appearance. The direct contribution of this data corpus for face recognition is highlighted in the evaluation of a standard face recognition algorithm, which illustrates the impact that age-progression, has on recognition rates. Assessment of the efficacy of this algorithm is evaluated against the variables of gender and racial origin. This work further concludes that the problem of age-progression on face recognition (FR) is not unique to the algorithm used in this work.

A new collaborative filtering recommendation algorithm based on dimensionality reduction and clustering techniques

Abstract: With the advent and explosive growth of the Web over the past decade, recommender systems have become at the heart of the business strategies of e-commerce and Internet-based companies such as Google, YouTube, Facebook, Netflix, LinkedIn, Amazon, etc. Hence, the collaborative filtering recommendation algorithms are highly valuable and play a vital role at the success of such businesses in reaching out to new users and promoting their services and products. With the aim of improving the recommendation performance of such an algorithm, this paper proposes a new collaborative filtering recommendation algorithm based on dimensionality reduction and clustering techniques. The ?-means algorithm and Singular Value Decomposition (SVD) are both used to cluster similar users and reduce the dimensionality. It proposes and evaluates an effective two stage recommender system that can generate accurate and highly efficient recommendations. The experimental results show that this new method significantly improves the performance of the recommendation systems.

Big data machine learning using apache spark MLlib

Abstract: Artificial intelligence, and particularly machine learning, has been used in many ways by the research community to turn a variety of diverse and even heterogeneous data sources into high quality facts and knowledge, providing premier capabilities to accurate pattern discovery. However, applying machine learning strategies on big and complex datasets is computationally expensive, and it consumes a very large amount of logical and physical resources, such as data file space, CPU, and memory. A sophisticated platform for efficient big data analytics is becoming more important these days as the data amount generated in a daily basis exceeds over quintillion bytes. Apache Spark MLlib is one of the most prominent platforms for big data analysis which offers a set of excellent functionalities for different machine learning tasks ranging from regression, classification, and dimension reduction to clustering and rule extraction. In this contribution, we explore, from the computational perspective, the expanding body of the Apache Spark MLlib 2.0 as an open-source, distributed, scalable, and platform independent machine learning library. Specifically, we perform several real world machine learning experiments to examine the qualitative and quantitative attributes of the platform. Furthermore, we highlight current trends in big data machine learning research and provide insights for future work.