Kyle Hundman's repositories
MEMEX-WISDOM
Anomaly detection system/interface for tracking online weapons advertisements as part of DARPA MEMEX project
Aster_Hackathon
(Aster's MapReduce SQL) 1st place code and presentation from 10-hour Aster hackathon sponsored by Aster Teradata.
StarWars_Sentiment
(Python, Stanford Core NLP) The dialogue from Star Wars Episode IV: A New Hope is parsed into allegiance (rebel, empire, other) and run through Stanford Core NLP's Sentiment Analyzer. The sentiment of the Rebel Alliance is then visualized over time.
Web_Scraping_and_Classification
(Python) Using BeautifulSoup and Whoosh, Wikipedia text is indexed for all of the capital cities of the world and their countries.
Accenture_Chicago_Analytics_Competition
(R) 1st place presentation along with code sample for working with shape files and GGPLOT in R. Additional info can be found here: http://www.chicagotribune.com/news/local/suburbs/evanston_skokie_morton_grove/community/chi-ugc-article-accenture-supports-northwestern-university-gr-2014-06-06,0,5433739.story
cnn-text-classification-tf
Convolutional Neural Network for Text Classification in Tensorflow
counterfeit
Pilot for CE domain.
Pig_SpellingCorrection
(Apache Pig, MapReduce Java, Java UDFs) Code for correcting misspelled words by stemming words and calculating Levenshtein distance for a dictionary of words.
elasticsearch
Docker Official Image packaging for elasticsearch
Hadoop_Kmeans
(MapReduce Java, Pig, Python) Implementation of Kmeans algorithm on unstructured data stored in HDFS. Dataset is CMS medicare data released in April 2014.
Hive_MovieRecommendations
(Hive) Using user's ratings of previously watched movies, recommendations for unseen movies are provided.
MapReduce_max
(MapReduce Java) Code for calculating maximum temperature by year from two sample data sets (included).
MapReduce_ML_Classification
(MapReduce Java) Code for a machine learning classification problem.
MapReduce_StandardDeviation
(MapReduce Java) Building off the MapReduce_Wordcount problem, this MapReduce Java code finds the st. dev. rather than the average. Two separate MapReduce jobs are needed.
MapReduce_Wordcount
(MapReduce Java) Code for counting words that meet certain conditions.
Pig_SentimentAnalysis
(Apache Pig) Using dictionaries of positive and negative words, the sentiment of tweets is calculated.
R-walmartKaggle
(R) Code used in Kaggle competition for predicting Walmart department sales
tensordict
TensorDict is a pytorch dedicated tensor container.
topic_space
Topic modeling web application