momorr317 Goto Github PK

bayesian_exploration_suicidal_rates

Suicide is a tragic event with strong emotional repercussions for survivors and for families of its victims. It is a major public health problem and a leading cause of death around the globe. Many organizations such as United Nations Development Program, World Bank, World Health Organization are making various effort to prevent suicide from people’s last minute decisions as well as from the root causes. These organizations also collect data to help researchers study potential variables that contribute to people’s suicide rates. Our project looks at the suicide data compiled by these organizations and aims to identify and alleviate causes that contribute to suicide rates across the population.

blogosphere_text_analytics

Customer analytics could be useful to many, including market campaign, sales and product team, upper management and even the board of directors. Customer analytics refers to the processes and technologies that give organizations the customer insight necessary to deliver offers that are anticipated, relevant and timely. Nowadays, customers can gain information anywhere and they are more connected. This provides great opportunities for customer data analytics since data is easier to collect from information platforms. If customer analytics could be interpreted better, then purchasing behaviour could be predicted with more accuracy and this would result in business strategy towards the right direction and increasing in revenue generating.

capstone

deep_learning_nasdaq

This project aims to predict price movement of a designated stock during a fixed time frame using neural network models. The motivation comes from the increasing usage of electronic-trading platform in day-to-day trading activities, whereas the automation of movement of mid-price and price spread crossing becomes an essential part of every-day trading mechanism. By characterising the existing features in given dataset and creating new statistical features such as moving averages, we applied feed forward neural network (FFNN), convolutional neural network (CNN) and recurrent neural network (RNN) on training dataset and tested on validation dataset. After a comparison and discussion of the accuracies and losses of all three models, we reach to a conclusion that FFNN model works the best, with a training accuracy at 0.531 and test accuracy at 0.5, as well as a training loss level at 1.038 and test loss level at 1.04.

ds5559

ewma

Exponentially Weighted Moving Average algorithms for Go.

housing-price-king-county

This report serves the purpose of conducting analysis on features which drive housing price in King County of Washington State in United States and searching for the best statistical model to interpret the relationship among features and housing price, as well as making predictions of housing price in future. The dataset used in this study is obtained from web-based online data repository Kaggle. Both discrete and continuous variables will be addressed and massaged for better model development. Statistical techniques such as stepwise- type procedures, various forms of transformations and in uential point analysis will be exploited for model development and model adequacy validation. Multiple graphs and statistical output for methods selection are attached in the Appendix section to support the analysis of this report.

kaggle3

machine_learning_fifa

Over 42 million console players played FIFA in 2017. All soccer fans are eager to build their favorite teams. Being both enthusiastic soccer fans and data scientists we start this project to analyze correlation among players attributes and to predict areas we are interested in. In the dataset we choose, it provides players attributes which include on all player style statistics like Agility, Aggression, Ball Control etc. and player demographics data like Nationality,Age, Wage, Height etc. Our primary goal is to make a better understanding of which personal data are more related to players’ attributes and which attributes are more important for evaluating players’ performance.

machine_learning_google_play

While many public datasets online provide Apple App Store data, there are not many counterpart datasets available for Google Play Store Apps anywhere on the web. The Play Store Apps data has enormous potential to drive App-making businesses to success. Actionable insights can be drawn for developers to work on and capture the Android market. So this dataset we used is scraped from the Google Play Store, which contains 10,000 entries of data of Play Store Apps for analyzing the Android market and each App has 13 features to describe them, such as category, reviews, rating etc. We will do Exploratory Data Analysis, supplemented by some empirically competent statistical models to see which features will affect the rating of Apps and build models to predict the rating for Google Play Store Apps.

open-data-lab

an initiative to provide infrastructure for reproducible workflows around open data

study

sys6018-competition-titanic

uva-datapalooza