##Lecture 1 Summary (Introduction on Data Science part I)

• We talked about different roles of Data Scientists
• T-Shaped Data Scientists
• Data Science Workflow
• Continuous, Discrete and Qualitative Data
• Supervised vs Unsupervised Learning
• Set up github accounts
• set ipython notebook
• Introduced Numpy

• Classification vs Clustering and Regression vs Dimentionality Reduction
• Flexibility vs Interpretability
• Different types of data (Cross-Sectional, Time-Series, Panel Data)
• Walkthrough Acquire& Parses with Pandas
• HW 1 assigned - Due date Feb 8th at 6:30PM

• Measures of central tendency (Mean, Median, Mode, Quartiles, Percentiles)
• Measures of Variability (IQR, Standard Deviation, Variance)
• Skewness Coefficient
• Kurtosis Coefficient
• Boxplots
• Bias vs Variance
• Central Limit Theorem – Standard Error of Mean
• Class/Dummy Variables
• Walkthrough describing and visualizing data in Pandas

• Linear Regression lines
• Single Variable and Multi-Variable Regression Lines
• Capture non-linearity using Linear Regression lines.
• Interpretting regression coefficients
• Dealing with dummy variables in regression lines
• intro on sklearn and searborn library
• HW 2 assigned - Due date Feb 17th 2016 at 6:30PM

• Hypothesis test - test of significance on regression coefficients
• p-value
• Capture non-linearity using Linear Regression lines.
• Different types of errors and R-squared
• Interaction Effects

• Bias-Variance Trade off
• Validation (Test vs Train set)
• Cross-Validation
• Ridge and Lasso Regression
• (Optional) Backward Selection, Forward Selection, All Subset Selection. (If you want to use these methods you need to use R)

• Types of missing data (MCAR, MAR, NMAR)
• Single imputation and their limitations
• Imuptation using regression lines and error
• Hot deck imputation
• multiple imputation

• Classification Problems
• Misclassifciation Error
• KNN algorithm for Classification
• Cross-Validation for KNN Algorithm
• Limitations of KNN Algorithm
• KNN algorithm for Regression

• Intro to Logistic Regression
• Odds vs Probability
• Using Logistic Regression to Make predictions
• How one interprets coefficients of Logistic Regression model
• Strength and weaknesses of Logistic Regression Model

• Unbalanced observations and Logistic Regression
• FP/FN/TP/TN/FPR/TPR
• The effect of chaning Threshold
• ROC Curves
• Area Under Curve
• How to compare classifciation algorithms

• Decision Tree for Regression
• Greedy Approach
• Decision Tree for Classification
• Gini Index and Entropy index
• Limitation of Simple Decision Tree

• Bagging
• Random Forest
• Boosting
• Tuning parameters for boosting and Random Forest

Additional Resources

• Decision Tree - Video - Part 1
• Decision Tree - Video - Part 2
• Decision Tree - Video - Part 3
• BootStrap - Video

• Definition of Natural Language Processing
• NLP applications
• Basic NLP practice
• Stop words, bag-of-words, IF-DIF

Additional Resources

• If you want to learn a lot more NLP, check out the excellent video lectures and slides from this Coursera course (which is no longer being offered).
• Natural Language Processing with Python is the most popular book for going in-depth with the Natural Language Toolkit (NLTK).
• A Smattering of NLP in Python provides a nice overview of NLTK, as does this notebook from DAT5.
• spaCy is a newer Python library for text processing that is focused on performance (unlike NLTK).
• If you want to get serious about NLP, Stanford CoreNLP is a suite of tools (written in Java) that is highly regarded.
• When working with a large text corpus in scikit-learn, HashingVectorizer is a useful alternative to CountVectorizer.
• Automatically Categorizing Yelp Businesses discusses how Yelp uses NLP and scikit-learn to solve the problem of uncategorized businesses.
• Modern Methods for Sentiment Analysis shows how "word vectors" can be used for more accurate sentiment analysis.
• Identifying Humorous Cartoon Captions is a readable paper about identifying funny captions submitted to the New Yorker Caption Contest.

• Principal Component Analysis
• Computation of PCAs
• Geometry of PCAs
• Proportion of Variance Explained

Additional Resources

• This tutorial on Principal Components Analysis (PCA) includes good refreshers on covariance and linear algebra
• To go deeper on Singular Value Decomposition, read Kirk Baker's excellent tutorial.
• Chapter 10 of Statistical Learning with applications in R

• AutoRegressive Models
• Moving Averages
• ARMA
• ARIMA

Additional Resources

• This is a good resource for AR models
• Seemingly easy to read book on time series.

• Talked about databases and data warehouse design.
• Introduction to SQL and learn the Fundamental Growth Query.
• Look at product engagement data of a fictional company and use FGQ to compute retention curves.
• Apply convolution to the retention curve to project future active users.
• Build a model to predict the retention likelihood of individual customers.
• Thanks to Michael

Additional Resources

• Well organized and easy to undestand tutorials - Thanks to Catherine
• More tutorials on SQL - Thanks to Randy

• Pre-work: Please review Bayes Questions.