alexhanna

1. Code and Data for the Social Sciences: A Practitioner’s Guide, Gentzkow and Shapiro.
2. Good enough practices in scientific computing, Wilson et al.
3. Best Practices for Scientific Computing, Wilson et al.
4. Principled Data Processing, Patrick Ball.
5. The Plain Person’s Guide to Plain Text Social Science, Healy.
6. Avoiding technical debt in social science research, Toor.

• For background on Technical Debt: Wikipedia; D. Sculley et al., [Machine Learning: The High-Interest Credit Card of Technical Debt](https://ai.google/resear

alexhanna

#!/bin/sh

## This is copy-pasta from the original Medialab script with some mods
## https://raw.githubusercontent.com/mediacloud/cliff-docker/master/launch.sh

echo "Getting CLIFF..."
echo "  downloading Cliff WAR file from GitHub"
wget https://github.com/mitmedialab/CLIFF/releases/download/v2.6.1/cliff-2.6.1.war
sudo mv cliff-2.6.1.war /var/lib/tomcat8/webapps/
echo "  done (copied /var/lib/tomcat8/webapps/)"

alexhanna

#!/usr/bin/env python
# encoding: utf-8
"""
split_ln.py

Created by Neal Caren on 2012-05-14.
neal.caren@unc.edu

Edited by Alex Hanna on 2015-01-29
alex.hanna@gmail.com

alexhanna

## FILE: Classifying 20 Newsgroups Dataset
## For presentation with Computational Sociology source at Duke.
## AUTHOR: Alex Hanna (ahanna@ssc.wisc.edu)
## DATE: October 14, 2015

## load the RTextTools package
## Documentation of this package is available at 
## https://cran.r-project.org/web/packages/RTextTools/RTextTools.pdf
library(RTextTools)

alexhanna

insert overwrite local directory '/scratch.1/sample2013_1'
row format delimited 
fields terminated by "\t"
select id_str, created_at, regexp_replace(text, "[ \t\r\n]+", " "), user.id_str, regexp_replace(user.name, "[ \t\r\n]+", " "), user.screen_name, retweeted_status.id_str, retweeted_status.created_at, regexp_replace(retweeted_status.text, "[ \t\r\n]+", " "), retweeted_status.user.id_str, regexp_replace(retweeted_status.user.name, "[ \t\r\n]+", " "), retweeted_status.user.screen_name 
from gh_rc TABLESAMPLE (10 PERCENT)
WHERE year = 2013 and month = 1;

insert overwrite local directory '/scratch.1/sample2013_2'
row format delimited 
fields terminated by "\t"

alexhanna

SET hive.exec.compress.output=true;
SET mapred.max.split.size=256000000;
SET mapred.output.compression.type=BLOCK;
SET mapred.output.compression.codec=org.apache.hadoop.io.compress.SnappyCodec;
SET hive.exec.dynamic.partition.mode=nonstrict;
SET hive.exec.dynamic.partition=true;

CREATE EXTERNAL TABLE gh_raw (
  id BIGINT,
  created_at STRING,

alexhanna

#!/usr/bin/env python
# encoding: utf-8
"""
    Module for parsing Proquest data. 
    Only tested on limited bits of the Proquest Ethnic Newswire.
    Based loosely off a script by Neal Caren (neal.caren@unc.edu)

    Alex Hanna, alex.hanna@gmail.com
    2017-05-04
"""

alexhanna

Call for TA: CCT490 (Social Data Analytics)

The Institute of Communication, Culture, Information and Technology at the University of Toronto Mississauga is looking for a teaching assistant for CCT 490 -- Social Data Analytics -- for Fall 2016, taught by Professor Alex Hanna. The course will cover basics of data collection, processing, and analysis for social trace data, such as Twitter and Facebook messages.

The position is for 40 hours a week for the Fall 2016 term, and will involve grading assignments, assisting in labs, and invigilating exams. The position is represented by CUPE 3902, Unit 3.

Applicants must have proficency in the Python programming language. Knowledge of other programming languages is a plus but not required. Experience with analysis of social media data is preferred. Applicants must live in the Toronto area and be able to travel to the Mississauga campus at least once a week.

To apply, please send a resume or CV to alex.hanna@utoronto.ca, with a short cover letter. The deadline fo

alexhanna

t.cox2_ph <-  coxph(t.surv ~ (Age + PlusSize + PuertoRico + Wins + Highs + Lows + Lipsyncs + CompLeft +
	Wins*CompLeft + Highs*CompLeft + Lows*CompLeft + Lipsyncs*CompLeft) + cluster(ID), df) 
t.cox3s   <- coxph(t.surv ~ (Age + PlusSize + PuertoRico + Wins + Highs + Lows + LipsyncWithoutOut + CompLeft) + cluster(ID), df) 

model.df   <- data.frame(ID = integer(0), Residuals = double(0), Model = character(0))
model.list <- list(c2 = t.cox2, c2ph = t.cox2_ph, c3 = t.cox3, c3s = t.cox3s)

for (i in 1:length(model.list)) {
  name  <- names(model.list[i])
	cMod  <- model.list[[i]]

alexhanna

# File-Name:       polClassify.R           
# Edited:          2013-03-20                          
# Orig.Author:     Drew Conway (drew.conway@nyu.edu)
# 
# Packages Used:   tm, ggplot2
# 

# All source code is copyright (c) 2012, under the Simplified BSD License.  
# For more information on FreeBSD see: http://www.opensource.org/licenses/bsd-license.php