Big data internship plan at Contemi Vietnam
2 likes1,432 views
The document outlines the experience and outcomes of a big data internship at Contemi Vietnam from September 2013 to January 2014. It details the learning process involving R/Python, machine learning, and Hadoop preparation, alongside participation in various Kaggle competitions. Future plans include developing applications using open datasets, stock prediction tools, and contributing to GitHub and R-bloggers.
Big data internship plan at Contemi Vietnam
- 1. BIG DATA FIRST INTERNSHIP Sep 2013 – Jan 2014 Organized by Contemi Vietnam
- 2. Outcomes • Contemi • Big Data presence • Big Data know-how • Big Data experience • Interns • Linux • R / Python language • Machine Learning practice • Process • Scrum • Cross Industry Standard Process for Data Mining (CRISM-DM) • Kaggle profile • Hadoop
- 3. Preparation • Platform: Ubuntu 12.04 LTS • Process: • Scrum • Cross Industry Standard Process for Data Mining (CRISP DM) • Weekly blog • http://contanalytics.wordpress.com/
- 4. Headstart for Dung • 16/09 – 30/09 • Learn R / Python • Try Digit Recognizer competition on Kaggle.com • Join in Introduction to Recommender System and Web Intelligence and Big Data on Coursera.com
- 5. 3 month plan • 1/10 – 31/10 • Go through all typical Machine Learning algorithms, implement, demo and present to Contemi • 1/11 – 15/11 • Compete for AMS 2013-2014 Solar Energy Prediction Contest • URL: http://www.kaggle.com/c/ams-2014-solar-energy-prediction-contest • 16/11 – 22/11 • Compete for Accelerometer Biometric Competition • URL: http://www.kaggle.com/c/accelerometer-biometric-competition • 23/11 – 31/12 (end of internship) • Deploy Hadoop • Learn Java • Run Word counting and Sorting experiments with large data (> 1GB) • Compete for Facebook Recruting III – Keyword Extraction (personally) • Re-optimize built model basing on Hadoop
- 6. Next plan for next internships • App using Singapore open datasets • Stock prediction app for Vietnam market • Visualization • GitHub • R-Bloggers