prahasan21 Goto Github PK

applied-machine-learning

From the scratch implementations of Perceptron, Linear regression and Logistic Regression

autonomous-car-vision-system-

Using image segmentation and deep learning techniques.

b551-artificial-intelligence

All the assignments related to the course B551(Elements of Artificial Intelligence)

detailed-view-of-mnist-classification-using-dnn

MNIST Classification using DNN and Analysis of behaviour in various hidden layers

doctors-appointment-no-show-prediction

A no-show in the healthcare sector is an appointment, where the patient or client did not show- up or try to call the hospital to cancel the appointment or reschedule the appointment.

exploratory-data-analysis-of-unknown-data

feedback-and-query-expansion

essential tasks of IR – help users to generate better query, computationally, and to better characterize users’ information needs. Feedback is an important approach to achieve this goal. Rocchio algorithm is used for relevance feedback system.

ils_534-search-final-project

Task 1 is Yelp Based Restaurant Recommendation System and Task 2 is Predicting the star ratings from the Yelp reviews for Restaurants in Las Vegas

income-inequality

Gaps in earnings between a nation’s most affluent and the rest of the country continue to grow every year. This project explores this major research issue of uneven distribution of income in the U.S. and other countries between 1988 and 2008. We primarily used three datasets to investigate this topic: 1) Lakner-Milanovic World Panel Income Distribution (LM-WPID) 2) Ventile 3) World Income Distribution (WYD)

information-retrieval-algorithm-and-evaluation

Based on the Lucene index, we can start to design and implement efficient retrieval algorithms. Here I implement the TF-IDF algorithm to search the documents. And evaluate my search algorithm with the Lucene similarities.

machine-learning-in-computational-linguistics

Assignments

mnist---variational-auto-encoders

MNIST Poor seven detection using VAE approach.

mnist-classification-using-compressed-dnn

Network Compression using SVD for MNIST

mood-chords

Machine learning based service, deployed as a flask-based web application.

simple-tech-challenges

Holds instructions and datasets for simple tech challenges

speaker-verification-siamese-networks

Speaker verification is the process by which a speakers claim of identity is tested against a claimed speaker by his or her voice.

speech-denoising-using-deep-learning

I have implemented DNN, CNN (1D and 2D) and RNN (LSTM) models to remove noise from the given signal. Compared their performances

text-indexation-for-information-retrieval

Indexation is important for information retrieval. Search engine indexing collects, parses, and stores (text) data to facilitate fast and accurate information retrieval.

thinktank-dataset

The Thinktank Data for Progress collected survey data (in DFP WTHH release.csv) that represents the population of people registered to vote in the 2018 midterm elections. As per the mini-project, we need to address the following questions:

what-s-cooking-

Sentiment Analysis

yelp-restaurant-recommendation-system

A hybrid recommendation engine that uses both the content of the reviews and collaborative filtering approach to effectively recommend users restaurants using the Yelp Dataset.

yelp-review-rating-prediction

Predicting the star ratings of customers based on their reviews. Deep sentiment analysis using CNN - GloVe, Word Embedding models. Also compares the conventional algorithms performance with DNN.