WRAP-UP

Open this page in a new tab and keep the table below open.
Click Home to return.

In the table below, the first post is the oldest project; the last post is the latest project. In the Home page, the top tiles are the latest projects.

Project	Image	Parameters	Tags & topics
Modeling Credit Risk		Code: R Tool: Notebook	Logit, probit, loglog and decision trees. Descriptive statistics. Train and test sets. Predictions. Confusion Matrix and ROC. Bank loan portfolio acceptance rate, bad rate, and risk tolerance.
Titanic: Getting the Picture		Code: R, Python Tool: RStudio, Notebook	Decision trees, k-NN, and random forests. Storytelling and narrative. Data exploration: tables vs Venn diagrams vs visualization. Train and test sets. Confusion Matrix. Folds and cross-validating. Pruning and avoiding overfitting.
Exploring Pitch Data		Code: R Tool: Notebook	Multivariate analysis and visual exploration. Clean and format datasets. Pitching velocity, mix, patterns, location in the ball-strike zone. Change by month, by game, by inning. Ball-strike count, early- and late-game situations. Velocity, impact, and contact rate.
Mining Text		Code: R Tool: Notebook	Natural language processing, sentiment analysis, and topic modeling. Build a corpus of texts (documents or any tweet, email, comment, publication, status, etc.). Download data using APIs. Populate a database. Explore the statistics. Filter and extract regular expressions. Visualize words, frequencies, ngrams. Assess sentiment, draw conclusions, and provide advice.
Forage de texte		Code: R Tool: RStudio	Traitement du langage naturel. Construire un corpus de textes. Explorer les statistiques. Visualiser les mots, les fréquences, les mots communs, les mots différents, les bigrammes. Utiliser des nuages, des graphiques à barres et des dendrogrammes.
Funny!			A collection of funny illustrations.
Data Storytelling		Code: R Tool: RStudio	Present to a technical and a nontechnical audience. Storytelling. Bring arcane subjects into general use. Use econometrics techniques. Pose hypotheses, set goals, perform analyses and draw conclusions.
…and Counting		Code: R Tool: RStudio	Model consumer demand (unit sold). Predict trends. Poisson and Negative Binomial distributions for counting discrete events.
Visualization		Code: R, Python Tool: RStudio, Notebook	Show graphics and maps instead of explanation or simple data tables. Visualization. Bring opaque data into general understanding. Storytelling with numbers. Present surveys and polling data.
Interactive Visualization		Code: R Tool: RStudio	Interactive data visualization and graphics.
Pythonic Stuff		Code: Python Tool: Text Editor	A series of projects. A website using a simple web framework. Documentation websites using static site generators. A command-line game and an application to be downloaded and installed.
Thoughtful…			An anthology of thoughtful images.
Optimizing the Coffee		Code: Python Tool: Notebook	Mathematical optimization. The cooling effect of cream in the coffee. Extrapolation and interpolation.
Descriptive & Inferential Statistics		Code: R Tool: RStudio	Basic to advanced statistical methods. Analyze census data (US state population). Infer the population with sampling and bootstrapping. Simulations and Monte Carlos.
Tweet, Tweet		Code: R Tool: RStudio	Web scraping (tweets) with an API. Natural Language Processing. Select Topics and keywords to capture tweets. Get up-to-the-minute data and measure delays between tweet (tweeting speed). Text mining and word clouds. Compare two assess popularity with the Poisson distribution. Analyze and manipulate text strings.
Map Mashup & Geointelligence		Code: R Tool: RStudio	Data visualization and map mashups. Introduction to spatial analysis. How to add intelligence to maps.
Geospatial Analysis and Geostatistics		Code: R Tool: RStudio	Introduction to geospatial models. Visualization with maps. Analyze the Australian Football League audience. Spatial autocorrelation. Autoregressive, lag and error models. Spatial logit and probit models. More advanced models.
Funny 2!			Another collection of funny illustrations.
Infographic Software		Code: Tableau Tool: Tableau	Experimenting with Tableau. Infographic examples.
Sieving Data		Code: R Tool: RStudio	Data mining. Market basket analysis. Understanding consumer behaviour. Association rules or what is behind recommendation systems. data mining. Market basket analysis. Understanding consumer behaviour. Association rules or what is behind recommendation systems. Dimension reduction. Multidimensional scaling. Factorial analysis, Component analysis (principal, simple, multiple). Linear discriminant analysis. Feature selection.
Algorithms and Spam		Code: R Tool: RStudio	Analyze texts (emails) with algorithms. Differentiate spam and nonspam. Custom methods, tree-based methods, and Support Vector Machine. Train, test, and evaluate the methods.
Survival of the Fittest		Code: R Tool: RStudio	Survival analysis. Event history analysis. Failure and churn analysis. Parametric, semiparametric, and nonparametric models: proportional hazards, accelerated failure time, exponential, piecewise exponential, Weibull, lognormal and Cox regression. Customer churn analysis. Censored and truncated data. Limited dependent variable and Tobit models.
Machine Learning: Classifiers & Clusters		Code: R Tool: RStudio	Classification and clustering methods. Unsupervised techniques: k-means cluster, k-nearest neighbours, hierarchical clustering. Supervised techniques: regression, trees, random forests. Train and test sets. Prediction. Performance measures: Dunn’s index, ROC, AUC, confusion matrix. Dimension reduction: multidimensional scaling, factorial analysis, component analysis, linear discriminant analysis and feature selection.
Funny 3!			Yet again… funny illustrations.
Web-based Interactive Maps		Code: R, Python Tool: RStudio, Notebook	Responsive. True interactivity: zoom in, zoom out, markers, pop-ups, move around, etc.
À la LaTeX		Code: LaTeX Tool: LaTeX Editor	Quick reference and extensions to write in LaTeX.