Shagufta Anjum

Hi, my name is

Shagufta.

I am passionate about artifificial intelligence.

I love exploring new use cases of machine learning and deep learning to solve real world problems. I am interested in AI research, particularly in computer vision and natural language processing.

Resume

About Me

I am a computer science and engineering undergrad student from Hyderabad, India. Over the years, I have developed a strong interest in machine learning and artificial intelligence, and I am currently working on a couple of research projects in this field. I aspire to pursue a masters degree in computer science with a specialization in AI. My dream is to conduct cutting-edge research at an organization devoted to leveraging the power of AI for social good. Here are a few topics that excite me:

Real world applications of ML
Natural Language Processing
Computer Vision
Ethical implications of AI
Music Information Retrival
Social network analysis

Experiences

Cognida.ai
Dell Technologies
Hitachi Vantara

Machine Learning Intern - Cognida.ai

August 2022 - present

Building a writing assistant tool for academic papers using natural language processing techniques to review errors, suggest appropriate edits, detect plagiarism, paraphrase, and generate titles.

Software Engineer Intern - Dell Technologies

June 2020 - August 2021

Worked in the global supply chain operations team to develop a system for the detection of errors in files being uploaded to AWS ECS instances. The system could also predict the file upload time using regression techniques.

Software Developer Intern - Hitachi Vantara

Jan 2020 - March 2020

Designed database schemas, implemented basic UI features, developed REST APIs, unit-tested features and deployed the application on an AWS environment.

Education

2019 - 2023

Bachelor of Technology in Computer Science & Engineering

Mahindra University

GPA: 8.9 out of 10.0, Class Rank: 5

2016 - 2019

High School

Chirec International

GPA: 95.4 out of 100

Projects

Multi-objective Evolutionary Neural Architecture Search for Autoencoders

Developed an algorithm to perform neural architecture search for autoencoder architectures using differential evolution to simultaneously minimize the number of latent variables, total parameters, and the error in recovery of the original inputs at the output layer. Implemented concurrent task apportioning on hybrid CPU-GPU architectures for efficient training of the neural networks.

Code

Leveraging Network Similarity Measures for Recommendation Systems

Designed a recommendation system for e-commerce products using memory-based and model-based collaborative filtering techniques. Proposed a novel method of graph-based recommendation based on local node similarity metrics from the product-user network. Published in International Conference on Emerging Techniques in Computational Intelligence (ICETCI), 2022.

Code

Low-level Fuzzy similarity Metrics for Music Recommendation

Extracted low-level acoustic features from raw music audio data using music information retrieval techniques for signal processing. Explored both time and frequency domains of the digitized audio signal and derived the corresponding spectrograms. Conceptualized a song recommendation system based on fuzzy clustering of user preferences using inter and intra genre song similarities.

Code

Local Community Member Detection in Social Networks

Developed a method to classify a pair of nodes in a complex network as belonging to the same community or not based solely on local node features such as centralities, distances and degree distribution. Applied machine learning classification algorithms such as SVMs, K-nearest neighbors and Random forests with Boosting.

Code

Eyes for the Elderly - Object Localization for Assisted Living

Designed a solution to assist elderly people suffering from dementia and vision loss to find items around their home. The hardware prototype includes a wall-mounted camera integrated wirelessly with a voice assistant enabled app and object detection software. Trained a custom CNN object detection model to locate the relative position of objects in a confined environment under varying lighting and occlusion conditions.

Code

Studying COVID spread using the Twitter Network

Gathered a dataset of over 500,000 tweets related to coronavirus from the Twitter API based on content and hashtags. Performed complex network-based analysis to investigate public perception, user sentiments and patterns of information dissemination.sing ML regression techniques to model the event-based propagation of information on Twitter and flow of tweets between different geographical locations and social groups.

Code

Achievements

Academic Scholaships - Rs. 1,00,000 each

Awarded three times (2019, 2021, 2022) for outstanding academic performance.

Winner - Smart India Hackathon by Government of India

Built a mobile app that allows users to try on 3D models of turbans virtually.

Winner - Hack-week at Mahindra University

Developed a 2D platform type game similar to Super Mario Bros.

President - Alumni Relations Centre at Mahindra University

Led a team of 30 students. Organized several alumni events. Set up the official alumni portal for my university.

Vice President - Street Cause NGO

Provided assistance to orphanages and old-age homes in Hyderabad.

Head of organizing team - Mahindra University's Annual Research Symposium

Led a team of 8 to organize the large-scale event attended by reputed researchers from across India.

Get in Touch

My inbox is always open. Whether you have a question or just want to say hi, I’ll try my best to get back to you!

E-mail me