Name: Ponkoj Shill
Type: User
Company: University of Nevada, Reno
Bio: Ph.D. Student in Computer Science.
Machine Learning Engineer.
Data Science Enthusiastic.
Location: Reno, Nevada, USA
Ponkoj Shill's Projects
I have used Gaussian Naive Bayes Model to predict digits from handwritten examples. This model has been trained with Scikit-learn Load_Digit Dataset, which is built in dataset of Scikit-learn.
Course content for UNR CS4/691 - Human-Robot Interaction
I Used scipy.stats.ttest_ind which is a two-sided test for the null hypothesis that 2 independent samples have identical average (expected) values. This test assumes that the populations have identical variances by default. I used three datasets for these hypothesis testing such as Zillow research data site there is housing data for the United States, from the Wikipedia page on college towns is a list of university towns in the United States and from Bureau of Economic Analysis, US Department of Commerce, the GDP over time of the United States in current dollars , in quarterly intervals.
Several services of IBM Watson
Download images from google search using Selenium.
I used Scikit learn based library Surprise to build a movie recommendation system. This system can recommend a person suitable movies based previous interests and ratings. Data extraced from Movielens. Model: Matrix Factorization-based algorithm.
Here, I have collected an database example from internet and apply most of the important MySQL queries which is saved as a pdf file.
This is the Australian Weather dataset, been used to predict the next day rain. Three different classifier used in a hard voting system to predict.
Implemented a machine learning approach that can predict if a person has a risk of stroke or not. The data collected from the internet are basically the medical reports of the patients about their age, gender, hypertension, heart disease, ever married, work type, residence type, average glucose level, BMI and smoking habit etc. The size of the dataset is just more than 5 thousand. Tried several classical approaches and Multi Layer Perceptron gained the place for this application with 96.9% accuracy.
A prediction is created based on Titanic dataset if the passengers would have survived or not. This is the legendary Titanic ML competition on Kaggle.