Programming Language:- Python

Library:- Keras, NLTK, django

Dataset:- CiteseerX

Team Size:- 2

Professor Mentor:- Dr. Sukomal Pal (spal.cse@iitbhu.ac.in)

Phd Student Mentor:- Tribikram Pradhan (tpradhan.rs.cse16@itbhu.ac.in)

Role Played:-During this Internship, I worked on a Local and Global Citation Recommendation system. In this, I worked in a team of 2 members and I was given the responsibility of developing Deep learning approach. The deep learning approach was combined parallelly with the graph-based approach. The Deep Learning model(LSTM) was trained on CiteseerX. The embedding vector was made using the word2vec for author, abstract, Title, and context.

Programming Language:- Python

Library:- Keras, opencv, pandas

Dataset:- DEAP dataset

Team Size:- 4

Mentor:- Mr. Sanjeev Kubakaddi (sanjeev@itie.in)

Role Played:- During this Internship different type of classifiers such as Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Naive Bayes classifier (NB) and Random Forest classifier are studied while changing the amount of data along with different Principal Component Analysis (PCA) technique like standard deviation, mean and variance to analyse the emotion signal for each frequency band obtained by wavelet decomposition from an EEG. The dataset is cleaned using the bandpass filter to remove artifacts and notch filter to remove noise generated using a power line.

Programming Language:- Python

Library:- Opencv, pandas

Dataset:- Self Made

Team Size:- 5

Professor Mentor:- Dr. N Kishorjit Singh (kishorjit@iiitmanipur.ac.in)

Role Played:- In this, I worked on a team and I was given the responsibility to find the distance. The idea of measuring the distance between the object and the center of two cameras placed at any distance from each other. The distance is calculated with the concept of Digital Image Processing. The input given to the algorithm is the only difference between the x coordinate of the left camera image and the right camera image. It was found that the distance is inversely proportional to the difference of x coordinates. The relation is verified using the graph plot between the distance and the difference of x coordinates. The error percentage is quite low(5%). In this Algorithm camera details and the angle made by the camera with the object for measuring the distance is not required. I am still Working on Adavce Driver-Assistance System.

Programming Language:- Python/Matlab

Library:- pandas, keras, sklearn, eeglab

Dataset:- Self Made

Team Size:- 8

Professor Mentor:- Dr. Anubha Gupta (anubha@iiitd.ac.in ), Dr. Rohit Verma (rohit.aiims@gmail.com)

Collabrating Agency:- sbilab IIITD, AIIMS New Delhi

Role Played:- In this project, we worked on makeing the dataset of the EEG signal for Subject having Hearing Aid. The dataset is made with different environment scenario and in the presence of Audiologist, ENT expert. The collected dataset was analyzed with the proper pipeline.