Hi, I am Aman Bhansali, pursuing my B-Tech in the Mechanical Engineering Department at the Indian Institute of Technology Jodhpur. Applications like machine learning, data science, and natural language processing that address challenges in the real world greatly fascinate me.
I explored the field and created my projects in this area after discovering my interests during the coursework session. I work diligently and honestly, and I am passionate and enthusiastic about this industry.
I have worked on a Research Project at IIT Patna under the supervision of Prof. Sriparna Saha called "Towards Query Semantic Knowledge Guided Multi-Span Medical Question Answering" which intends to enhance the medical context summarizer with multi-spanned answers provided a question. Since a lot of text is available today for a single question, it becomes a challenge to identify the most pertinent information. Our results have surpassed the existing state-of-the-art model with around a 2.5-4% increment in terms of F1 score and exact match.
Applying ML/DL concepts to the following
Analyzing and producing information from a given image is particularly beneficial in the field of medical science. The use of the ESRGAN models to improve these photos may be crucial in this situation. We use BreakHis dataset here.
Used POD to analyze the most influencing modes in the fluid flow using a schlieren imaging dataset generated from a supersonic wind tunnel. Further applied machine learning concepts for flow edge detection and flow Mach number calculation.
Using the Google Stock Price dataset from Kaggle, and explored various techniques and methods related to Time Series Analysis. Evaluated various machine learning models and NN using LSTM models and analyzed the results for future forecasting.
Used the airline sentiment analysis dataset to train a model which could detect the sentiment of a statement which today social media companies use to detect offensive sentiments.
My Services
Contact Me