Nirali Parekh

Nirali Parekh

Masters' student at Stanford University

Posts by Collection

experiences

portfolio

A Federated Approach to Predicting Emojis in Hindi Tweets  

Machine Learning   Python   Federated Learning   NLP   Privacy  

This project introduces a new dataset for Hindi emoji prediction and proposes a modified federated learning algorithm, CausalFedGSD, to balance model performance with user privacy.

Glide: Precision Marketing for Content Creators  

Software Development   Javascript   React   Node.js   Heroku   Express.js

A platform that revolutionizes content marketing by tailoring products to niche interests, leveraging real-time social media content, complemented by our AI-driven VideoIndexing tool for actionable insights.

Parkinson’s Detection - Explainable Approach to Temporal Audio Classification  

Machine Learning   Python   Explainable AI   Interpretable ML   AI in Healthcare  

An explainable deep learning model for Parkinson’s disease detection, using vocal phonation analysis to achieve high accuracy and providing a novel interpretation algorithm that assesses the significance of audio chunks in diagnosis.

ChaDRaL: RGB Image Encryption based on 3D Chaotic Map, DNA, RSA and LSB  

Software Engineering   Algorithm Design   Cryptography   Image Encryption    

This algortihm is an RGB image encryption method combining chaos theory, DNA sequencing, RSA encryption, and LSB steganography, offering secure key transfer and resistance to statistical and differential attacks.

My React Native Portfolio  

Software Engineering   Mobile Development   React Native   portfolio    

A comprehensive collection of my react native projects, each linked to its individual GitHub repository for in-depth exploration.

Sleep Stage Scoring using Raw EEG Signals  

Machine Learning   Time Series Data   CNN   AI in Healthcare    

This project employs Convolutional Neural Networks to analyze raw, single-channel EEG signals for accurately classifying the five sleep stages, providing a benchmark solution for sleep pattern analysis in medical diagnostics.

OuRCTC: Railways Reservation Assistant: SQL-Integrated Django Application  

Software Development   Django   Database Management   SQL    

This Django-based application streamlines the Indian Railways reservation process, allowing users to easily access train details, check trains between stations, book tickets, and calculate exact fares through a user-friendly interface connected to an SQL database

publications

TaxoDaCML: Taxonomy based Divide and Conquer using machine learning approach for DDoS attack classification

International Journal of Information Management Data Insights, 2021

This paper presents TaxoDaCML, a machine learning-based approach using a divide and conquer strategy for efficiently detecting and classifying various DDoS attacks with high accuracy.

Recommended citation: Thorat, O., Parekh, N., & Mangrulkar, R. (2021). TaxoDaCML: Taxonomy based Divide and Conquer using machine learning approach for DDoS attack classification. International Journal of Information Management Data Insights, 1(2), 100048. doi:10.1016/j.jjimei.2021.100048 https://www.sciencedirect.com/science/article/pii/S2667096821000410?via%3Dihub

Automatic Sleep Stage Scoring on Raw Single-Channel EEG : A comparative analysis of CNN Architectures

2021 Fourth International Conference on Electrical, Computer and Communication Technologies (ICECCT), 2021

This research explores the use of advanced Convolutional Neural Networks (CNNs) for classifying the complex stages of sleep through EEG data analysis, aiming to enhance the diagnosis and treatment of prevalent sleep disorders.

Recommended citation: Parekh, N., Dave, B., Shah, R., & Srivastava, K. (2021). Automatic Sleep Stage Scoring on Raw Single-Channel EEG : A comparative analysis of CNN Architectures. 2021 Fourth International Conference on Electrical, Computer and Communication Technologies (ICECCT), 1–8. doi:10.1109/ICECCT52121.2021.9616895 https://ieeexplore.ieee.org/document/9616895

ChaDRaL: RGB Image Encryption based on 3D Chaotic Map, DNA, RSA and LSB

2021 International Conference on Artificial Intelligence and Machine Vision (AIMV), 2022

This paper introduces ChaDRaL, an RGB image encryption method combining chaos theory, DNA sequencing, RSA encryption, and LSB steganography, offering secure key transfer and resistance to statistical and differential attacks.

Recommended citation: Parekh, N., & D’Mello, L. (2021). ChaDRaL: RGB Image Encryption based on 3D Chaotic Map, DNA, RSA and LSB. 2021 International Conference on Artificial Intelligence and Machine Vision (AIMV), 1–6. doi:10.1109/AIMV53313.2021.9670921 https://ieeexplore.ieee.org/document/9670921

A Federated Approach to Predicting Emojis in Hindi Tweets

2022 Conference on Empirical Methods in Natural Language Processing, 2022

This paper introduces a new dataset for Hindi emoji prediction and proposes a modified federated learning algorithm, CausalFedGSD, to balance model performance with user privacy.

Recommended citation: Gandhi, D., Mehta, J., Parekh, N., Waghela, K., D’Mello, L., & Talat, Z. (2022, December). A Federated Approach to Predicting Emojis in Hindi Tweets. In Y. Goldberg, Z. Kozareva, & Y. Zhang (Eds.), Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing (pp. 11951–11961). doi:10.18653/v1/2022.emnlp-main.819 https://aclanthology.org/2022.emnlp-main.819/

Parkinson’s Disease Detection - An Interpretable Approach to Temporal Audio Classification

2022 IEEE 3rd Global Conference for Advancement in Technology (GCAT), 2022

This paper presents a novel deep learning model for Parkinson’s disease detection using vocal signals, with a focus on providing interpretable results through an innovative algorithm that assesses the significance of different audio chunks.

Recommended citation: Shah, R., Dave, B., Parekh, N., & Srivastava, K. (2022). Parkinson's Disease Detection - An Interpretable Approach to Temporal Audio Classification. 2022 IEEE 3rd Global Conference for Advancement in Technology (GCAT), 1–6. doi:10.1109/GCAT55367.2022.9971881 https://ieeexplore.ieee.org/document/9971881

Text Style Transfer: A Comprehensive Study on Methodologies and Evaluation

Big Data, Machine Learning, and Applications, 2023

This paper provides a systematic review of Text Style Transfer (TST), categorizing past works based on machine learning and deep learning algorithms, discussing various TST subtasks, compiling available datasets, summarizing evaluation practices, and highlighting current challenges and future research directions in TST.

Recommended citation: Parekh, N., Trivedi, S., & Srivastava, K. (2024). Text Style Transfer: A Comprehensive Study on Methodologies and Evaluation. In M. D. Borah, D. S. Laiphrakpam, N. Auluck, & V. E. Balas (Eds.), Big Data, Machine Learning, and Applications (pp. 269–284). Singapore: Springer Nature Singapore. https://link.springer.com/chapter/10.1007/978-981-99-3481-2_22

Enabling Blockchain Architecture for Health information exchanges (HIE)

Unleashing the Potentials of Blockchain Technology for Healthcare Industries, Elsevier, 2023

This paper proposes a blockchain-based architecture for Health Information Exchanges (HIEs), to improve security, privacy, and interoperability, while granting patients complete ownership over their electronic health records.

Recommended citation: Parekh, N., & Mangrulkar, R. (2023). Chapter 5 - Enabling blockchain architecture for health information exchanges. In M. M. Ghonge, P. N., A. Das, Y. Wu, & O. Pal (Eds.), Unleashing the Potentials of Blockchain Technology for Healthcare Industries (pp. 77–93). doi:10.1016/B978-0-323-99481-1.00011-0 https://www.sciencedirect.com/science/article/pii/B9780323994811000110

talks

teaching

CS 3040: Data Structures and Algorithms

Prof. Pranit Bari, Teaching Assistant, Fall 2022, Winter 2022
DJ Sanghvi School of Engineering, University of Mumbai

This undergraduate course provides a foundational understanding of data structures and algorithms, emphasizing efficient data organization, storage, and algorithmic techniques essential for solving complex computational problems.

MS&E 249: Corporate Financial Management

Prof. Blake Johnson, Teaching Assistant, Autumn 2022
School of Engineering, Stanford University

This course focuses on the essential functions of finance in both large and small companies, covering core concepts and analytical tools for making financing decisions, evaluating investments, and managing cash flow, profitability, and risk.

OIT 276: Data and Decisions (Accelerated)

Prof. Stefan Wager, Teaching Assistant, Winter 2023
Graduate School of Business, Stanford University

OIT 276 is an advanced first-year MBA course focusing on statistics and regression analysis, integrating a group regression project with an emphasis on using the R software programming.

CS 246: Mining Massive Data Sets

Prof. Jure Leskovec, Teaching Assistant, Spring 2023
School of Engineering, Stanford University

This course offers an in-depth exploration of data mining and machine learning techniques for analyzing extensive datasets, focusing on big data systems, link analysis, similarity search, and methods for large-scale data processing, including recommender systems and social network analysis.

MS&E 231: Social Algorithms

Prof. Johan Ugander, Teaching Assistant, Autumn 2023
School of Engineering, Stanford University

This course explores the design and impact of learning algorithms within complex social systems, covering topics like search and recommendation systems, content moderation, and feed ranking.

MS&E 135: Networks

Prof. Johan Ugander, Teaching Assistant, Winter 2024
School of Engineering, Stanford University

This course offers a high-level overview of the fundamental role networks play in our social, technological, and natural worlds, focusing on key concepts in network analysis and their applications in various domains such as social networks, information spread, and collective behaviors.