Kory Becker primaryobjects
primaryobjects
/ gerber.R
Created
May 9, 2016 17:42
Voting habits analysis with logistic regression and CART regression tree models.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| library(ROCR) | |
| library(rpart) | |
| library(rpart.plot) | |
| gerber <- read.csv('gerber.csv') | |
| table(gerber$voting) | |
| table(gerber$voting)[2] / nrow(gerber) | |
| # Which of the four "treatment groups" had the largest percentage of people who actually voted (voting = 1)? |
primaryobjects
/ letters.R
Created
May 9, 2016 18:03
Predicting letter classification A, B, P, or R using CART classification models and random forests.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| library(caTools) | |
| library(rpart) | |
| library(randomForest) | |
| letters <- read.csv('letters_ABPR.csv') | |
| letters$isB <- as.factor(letters$letter == 'B') | |
| set.seed(1000) | |
| spl <- sample.split(letters$isB, SplitRatio = 0.5) |
primaryobjects
/ over50k.R
Created
May 9, 2016 18:47
Analysis of census for earning over50k (>= $50,000 per year) income using logistic regression, CART classification trees, k-fold cross validation, and random forests.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| library(caTools) | |
| library(ROCR) | |
| library(rpart) | |
| library(rpart.plot) | |
| library(randomForest) | |
| library(caret) | |
| data <- read.csv('census.csv') | |
| set.seed(2000) |
primaryobjects
/ plot.png
Last active
June 27, 2019 13:38
Sentiment analysis of Apple tweets, using CART, random forests, logistic regression with best accuracy of 89% from random forests.
primaryobjects
/ dplot.png
Last active
May 18, 2016 16:42
Predictive coding analysis of Enron emails to determine responsive documents about energy scandal. Using a CART model in R.
primaryobjects
/ plot.png
Last active
May 18, 2016 17:52
Wikipedia vandalism detection using CART classification and regression tree model with an accuracy of 0.7188306.
primaryobjects
/ clinical_trials.R
Created
May 18, 2016 18:33
Analysis of clinical trials using NLP with a CART model to determine if abstract and title is applicable to a study.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| packages <- c('tm', 'SnowballC', 'caTools', 'rpart', 'rpart.plot', 'randomForest', 'ROCR') | |
| if (length(setdiff(packages, rownames(installed.packages()))) > 0) { | |
| install.packages(setdiff(packages, rownames(installed.packages()))) | |
| } | |
| library(tm) | |
| library(SnowballC) | |
| library(caTools) | |
| library(rpart) | |
| library(rpart.plot) | |
| library(randomForest) |
primaryobjects
/ emails.R
Created
May 18, 2016 19:18
Detection of ham and spam emails from a data set using logistic regression, CART, and random forests. Random forests performs the best on train and test sets, while logistic regression overfits the training.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| packages <- c('tm', 'SnowballC', 'caTools', 'rpart', 'rpart.plot', 'randomForest', 'ROCR') | |
| if (length(setdiff(packages, rownames(installed.packages()))) > 0) { | |
| install.packages(setdiff(packages, rownames(installed.packages()))) | |
| } | |
| library(tm) | |
| library(SnowballC) | |
| library(caTools) | |
| library(rpart) | |
| library(rpart.plot) | |
| library(randomForest) |
primaryobjects
/ movies.R
Last active
May 23, 2016 17:43
Basic movie recommendation system, using content filtering (grouping movies with similar characteristics (ie., genre) together).
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # Basic movie recommendation system, using content filtering (grouping movies with similar characteristics (ie., genre) together). | |
| # Download dataset, if it does not exist. | |
| fileName <- 'movieLens.txt'; | |
| if (!file.exists(fileName)) { | |
| # , method="curl" | |
| download.file(paste0('http://files.grouplens.org/datasets/movielens/ml-100k/u.item', ''), fileName) | |
| } | |
| movies <- read.table('movieLens.txt', header=F, sep='|', quote='"') |
primaryobjects
/ clustering.R
Created
May 23, 2016 18:36
Clustering with hierarchical and k-means in R for image analysis.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| library(flexclust) | |
| # Download dataset, if it does not exist. | |
| fileName <- 'flower.csv'; | |
| if (!file.exists(fileName)) { | |
| # , method="curl" | |
| download.file(paste0('https://d37djvu3ytnwxt.cloudfront.net/asset-v1:MITx+15.071x_3+1T2016+type@asset+block/', fileName), fileName) | |
| } | |
| fileName <- 'healthy.csv'; |