Hemant Pandey

This project is based on Kaggle challenge which involves a dataset by Zillow and revolves around predicting the price that a particular real estate property (usually a home) will sell for. The goal is to explore the data and uncover interesting observations about the Zillow data, model the data and submitting the results on Kaggle.

A portal to give students feedback of their coding assignments and helps professors grade students based on the test cases passed for the student's submission. Currently in the development and testing face, to be used by the university in the coming semesters.

Developed such a system, which can classify online posts and comments into ’insulting’ or ‘not
insulting’ using NLP techniques and classification systems based on machine learning in Python.
It is an application-specific projects, where we have applied the NLP and classification techniques learnt in class to solve a real world problem. Our algorithm yielded a training accuracy of and a test accuracy of 85.37%.

We modified the hyper-parameters of the basic model using Linear SVC… Show more

Developed such a system, which can classify online posts and comments into ’insulting’ or ‘not
insulting’ using NLP techniques and classification systems based on machine learning in Python.
It is an application-specific projects, where we have applied the NLP and classification techniques learnt in class to solve a real world problem. Our algorithm yielded a training accuracy of and a test accuracy of 85.37%.

We modified the hyper-parameters of the basic model using Linear SVC. We included a L2 regularization constant, and tokenized words into unigrams and bigrams. Our best accuracy obtained was 85.37% and the F1-score obtained was 0.85.
Show less

The purpose of the project is to create a new system call 'mergesort' which takes two input files and sort the contents into a output textfile. The input files and output files will be passed to the system call along with the specified flags which we will discuss lat er. A new entry named xmergesort is made in the sys call table.

A social networking site which gives recommendation to users based on a affinity score. The details can be found in the Pulications section.

Analysis of filtering algorithms used in Recommender Systems based on two similarity criteria - Cosine Similarity and Pearson's Correlation. Algorithms studied are :
1.) User based collaborative filtering with Cosine Similarity.
2.) User based collaborative filtering with Pearson's Coefficient.
3.) Item based collaborative filtering with Cosine Similarity.
4.) Item based collaborative filtering with Pearson's Coefficient.

Detecting the vulnerable insecure codes in the web applications in PHP and providing secure solutions to them in PHP platform. Damn Vulnerable Web Application us used for reference purposes.

The project aims to predict the box office gross sales of upcoming movies. We anticipate that the sales will depend on factors like genre, cast actors, actresses, director, movie budget etc. It would be exciting to validate our predictions on movies releasing in December, just before the project submission.

Processing the available dataset of last 5 years of H1B visa applicants using Pandas dataframe in Python 2.7, visualizing the available information using multidimensional plots and interactive visualizations in D3 JavaScript to recognize patterns and inferences. The main goal was to create an interactive visualization platform which enables easy recognition of important information from a high dimensional large dataset. Reasoning with facts and some hypothesis testing was done to support the… Show more

Processing the available dataset of last 5 years of H1B visa applicants using Pandas dataframe in Python 2.7, visualizing the available information using multidimensional plots and interactive visualizations in D3 JavaScript to recognize patterns and inferences. The main goal was to create an interactive visualization platform which enables easy recognition of important information from a high dimensional large dataset. Reasoning with facts and some hypothesis testing was done to support the inferences drawn. Show less

Visualizing PCA, MDS cosine, correlation and eigen values plots of the secified dataset in D3 JavaScript

Depiction of histogram and pie charts to represent the characteristics of different players using D3 Javascript

The purpose of this assignment is to develop a Linux kernel-based system to support per-process system call vectors. Since in Linux, it is prohibited to change the global system call/vector table from a loadable module. Having different syscalls per process for different tasks can be very helpful. The system calls are modified in a such a way that they perform function which is a little different from the original system call.

There is another syscall 'clone' which is modified:
When a… Show more

The purpose of this assignment is to develop a Linux kernel-based system to support per-process system call vectors. Since in Linux, it is prohibited to change the global system call/vector table from a loadable module. Having different syscalls per process for different tasks can be very helpful. The system calls are modified in a such a way that they perform function which is a little different from the original system call.

There is another syscall 'clone' which is modified:
When a process is forked, it inherits the syscall_vector from its parent. The clone(2) syscall which is provided a new flag which decides whether the child would inherit the syscall vector of its child or not. If the flag is -1 it inherits its parent vector. If it is 1 or 2 then it takes the vectors 1 or 2 respectively.

A ioctl() is implemented which changes the syscall vector used by a running process id. We have tried to handle all the race conditions and implemented the necessary locks and also passes the checkpatch.pl script which makes sure that the code is written as per the standards.

Show less

The project supplements the RapClust pipeline to visualize fragmented equivalence classes using
a web GUI interface. It includes incorporating optimizations to process thousands of nodes and
display it using multi-level clustering. Implemented using Sigma graphical library and NodeJS.