Un Taccuino on Un nuovo Taccuino

Please don’t build your own co‑pilot, and learn to do your job first

Thu, 16 Oct 2025 00:00:00 +0000

Chronicles from a wonderful day in Luxembourg

I’m a bit tired of writing the usual gushing posts about how amazing and empowering it was to be part of event X, Y, or Z. Instead, here’s a short, candid note on what I actually took home after sharing my pitch — “Bringing Internal Audit into the AI Age” — with a large, well‑qualified crowd of auditors at the IIA Annual Conference in Luxembourg.

3 cose che ho imparato in 3 anni da manager

Sat, 15 Mar 2025 00:00:00 +0000

Tre anni fa, circa, sono stato nominato per la prima volta Head of di un team, che oggi è composto di circa 20 persone. Ho sempre avuto una certa passione per il management, nel senso che nel tempo sono andato ragionando su quali fossero i modi migliori per far lavorare bene insieme le persone. Se lo sto facendo bene o male chiedetelo ai fantastici #ADAAPeople, nel frattempo queste sono tre cose che ho imparato e che condivido per capire che ne pensate e perché possano essere utili a chi sta iniziando quel fantastico (e faticoso e infinito) viaggio che è diventare manager.

Italy Coronavirus Outbreak: numbers and stats

Mon, 24 Feb 2020 00:00:00 +0000

Italy NCOV-19 outbreak

Italy NCOV-19 outbreak

For personal reasons I am trying to track the number of NCOV-19 confirmed cases in Italy as well as the number of deaths (since I live in Italy, it is not difficult to guess the personal reason…). I am thus regularly monitoring news from the italian official sources like “Regione Lombardia” and “Protezione Civile”.

Introducing the paletteR Gallery

Tue, 26 Nov 2019 00:00:00 +0000

celebrating beauty

PaletteR, the package that allows you to create an optimized palette from an image, has been staying around for nearly two years, and #rstats users have made a lot of great stuff with it. I had therefore took the time to collect what I have found around the web, just to celebrate this beauty. You can find them below in a slideshow.

a quick ride on pagedown: create PDFs from Rmarkdown

Sun, 27 Jan 2019 00:00:00 +0000

I recently came across the good talk by the always good Yiuhi Xie at the Rstudio conference about Pagedown. You can see it by yourself reaching out the rstudio website or clicking the image below:

Pagedown is a newly released package ( still in experimentary status) with a really promising mission: help you build beautiful PDFs documents from our beloved Rmarkdown. More precisely it takes Rmarkdown files and render them into html files already “paged” and ready to be saved/converted into PDF.

introducing vizscorer: a bot advisor to improve your ggplot plots

Mon, 12 Nov 2018 00:00:00 +0000

introducing vizscorer: a bot advisor to improve your ggplot plots

introducing vizscorer: a bot advisor to improve your ggplot plots

One of the most frustrating issues I face in my professional life is the plentitude of ineffective reports generated within my company. Wherever I look around me is plenty of junk charts, like barplot showing useless 3D effects or ambiguous and crowded pie charts. I do understand the root causes of this desperate state of the art: people have always less time to dedicate to reports crafting, and even less to dedicate to their plot. In the crazy and speedy-going working life, my colleagues have no time and for learning data visualization principles. Even so, this remains quite a big problem since a lot of time and money-wasting consequences come from poorly crafted reports and plots:

getting to know the new definition of default

Fri, 13 Jul 2018 00:00:00 +0000

The greatest part of my time at working is spent looking at those great piece of statistical machinery that credit risk models are. That is why I was recently required to prepare a short course about the new definition of default, which is goign to be applied from 01/01/2021 on.

It is quite a unexplored topic in the sector and I haven’t found a lot of teaching material online, that is why I tought to share here the brief deck of slide I produced for the occasion. Enjoy and feel free to comment.

how to use PaletteR to automagically build palettes from pictures

Tue, 08 May 2018 00:00:00 +0000

UpdateR package: update R version with a function (on MAC OSX)

Sat, 10 Mar 2018 00:00:00 +0000

I personally really appreciate the InstallR package from Tal galilli, since it lets you install a great number of tools needed for working with R just running a function.

Learning Dataviz Principles and Theory from Tufte

Sat, 10 Feb 2018 00:00:00 +0000

I have recently completed a great reading: Edward Tufte’s The visual display of quantitative information. In the dataviz realm, this is some kind of fundamental book. This book was some kind of structural break in the history of data visualization. In ’70s and ’80s graphics were considered a way to entertain less educated readers. Their ability to make available new insights and communicate them effectively was underestimated.

Why are you being so silent?

Wed, 12 Apr 2017 00:00:00 +0000

It has being nearly half an year since the last post about workflower was out, why did I stay so silent for that long?

I have three major updates to explain the silence:

📚 Good guys at Packt publishing asked me to write one more book about R and data mining, I suppose this is because the first one was well received
📦 I spend my spare time working on updateR so to get it ready to go on CRAN. We make it so shining and bright that it got noticed by our beloved Tal Galili and we are now working togheter to merge it into his great package installR.

streamline your analyses linking R to sas and more: the workfloweR

Wed, 21 Sep 2016 12:22:29 +0000

we all know R is the first choice for statistical analysis and data visualisation, but what about big data munging? tidyverse (or we’d better say hadleyverse 😏) has been doing a lot in this field, nevertheless it is often the case this kind of activities being handled from some other coding language. Moreover, sometimes you get as an input pieces of analyses performed with other kind of languages or, what is worst, piece of databases packed in proprietary format (like .dta .xpt and other). So let’s assume you are an R enthusiast like I am, and you do with R all of your work, reporting included, wouldn’t be great to have some nitty gritty way to merge together all these languages in a streamlined workflow?

ggplot2 themes examples

Tue, 09 Aug 2016 20:30:16 +0000

this short post is exactly what it seems: a showcase of all ggplot2 themes available within the ggplot2 package. I was doing such a list for myself ( you know that feeling …“how would it look like with this theme? let’s try this one…”) and at the end I thought it could have be useful for my readers. At least this post will save you the time of trying all differents themes just to have a sense of how they look like.

Euro 2016 analytics: Who's playing the toughest game?

Tue, 21 Jun 2016 15:55:38 +0000

I am really enjoying Uefa Euro 2016 Footbal Competition, even because our national team has done pretty well so far. That’s why after browsing for a while statistics section of official EURO 2016 website I decided to do some analysis on the data they share ( as at the 21th of June).

Just to be clear from the beginning: we are not talking of anything too rigourus, but just about some interesting questions with related answers gathered mainly through data visualisation.

a Checklist for your weekly review (GTD methodology)

Tue, 17 May 2016 16:00:14 +0000

I was crafting this checklist for my personal use, and then I found myself thinking: why should’nt I share this useful handful of bullets with my readers? So here we are, find below an useful checklist for your weekly review. The checklist is derived directly from the official GTD book by our great friend David Allen. The greatest quality of the checklist is the minimalist approach: just what you really need to read is written within each point, so that you get through your review as quick as possible. Enjoy!

Over 50 practical recipes for data analysis with R in one book

Wed, 11 May 2016 14:00:46 +0000

Ah, writing a blog post! This is a pleasure I was forgetting, and you can guess it looking at last post date of publication: it was around january... you may be wondering: what have you done along this long time? Well, quite a lot indeed:

changed my job ( I am now working @ Intesa Sanpaolo Banking Group on Basel III statistical models)
became dad for the third time (and if you are guessing, it’s a boy!)

2015 in review (let me boast myself a bit :))

Fri, 01 Jan 2016 19:52:47 +0000

The WordPress.com stats helper monkeys prepared a 2015 annual report for this blog.

Here’s an excerpt:

The concert hall at the Sydney Opera House holds 2,700 people. This blog was viewed about **8,800** times in 2015. If it were a concert at Sydney Opera House, it would take about 3 sold-out performances for that many people to see it.

Click here to see the complete report.

Rename a Data Frame Within a Function Passing an Argument

Mon, 14 Dec 2015 21:04:33 +0000

This is not actually a real post but rather a code snippet surrounded by text.

Nevertheless I think it is a quite useful one: have you ever found yourself writing a function where a data frame is created, wanting to name that data frame based on a custom argument passed to the function?

For instance, the output of your function is a really nice data frame name in a really trivial way, like “result”.

how to list loaded packages in R: ramazon gets cleaver

Thu, 10 Sep 2015 09:19:43 +0000

It was around midnight here in Italy: I shared the code on Github, published a post on G+, Linkedin and Twitter and then went to bed.

In the next hours things got growing by themselves, with pleasant results like the following:

https://twitter.com/DoodlingData/status/635057258888605696

The R community found ramazon a really helpful package.

And I actually think it is: Amazon AWS is nowadays one of the most common tools for online web applications and websites hosting.

ramazon: Deploy your Shiny App on AWS with a Function

Tue, 18 Aug 2015 21:47:00 +0000

Because Afraus received a good interest, last month I override shinyapps.io free plan limits.

That got me move my Shiny App on an Amazon AWS instance.

Well, it was not so straight forward: even if there is plenty of tutorials around the web, every one seems to miss a part: upgrading R version, removing shiny-server examples… And even having all info it is still quite a long, error-prone process.

All this pain is removed by ramazon, an R package that I developed to take care of everything is needed to deploy a shiny app on an AWS instance. An early disclaimer for Windows users: only Apple OS X is supported at the moment.

Introducing Afraus: an Unsupervised Fraud Detection Algorithm

Thu, 02 Jul 2015 21:06:39 +0000

The last Report to the Nation published by ACFE, stated that on average, fraud accounts for nearly the 5% of companies revenues.

on average, fraud accounts for nearly the 5% of companies revenues

Projecting this number for the whole world GDP, it results that the “fraud-country” produces something like a GDP 3 times greater than the Canadian GDP.

How to add a live chat to your Shiny app

Mon, 11 May 2015 21:08:48 +0000

As I am currently working on a Fraud Analytics Web Application based on Shiny (currently on beta version, more later on this blog) I found myself asking: wouldn’t be great to add live chat support to my Web Application visitors?

It would indeed!

[caption id=“attachment_490” align=“aligncenter” width=“200”] an ancient example of chatting - Camera degli Sposi, Andrea Mantegna 1465 -1474[/caption]

How to list file and folders within a folder ( basic file app)

Wed, 01 Apr 2015 20:29:04 +0000

I know, we are not talking about analytics and no, this is not going to set me as a great data scientist… By the way: have you ever wondered how to list all files and folders within a root folder just hitting a button**?**

I have been looking for something like that quite a lot of times, for instance when asked to write down an index of all the working papers pertaining to a specific audit ( yes, **I am an auditor, **sorry about that): really time-consuming and not really value-adding activity.

Catching Fraud with Benford's law (and another Shiny App)

Fri, 06 Feb 2015 00:00:00 +0000

In the early ‘900 Frank Benford observed that ’1’ was more frequent as first digit in his own logarithms manual.

More than one hundred years later, we can use this curious finding to look for fraud on populations of data.

just give a try to the shiny app

What ‘Benford’s Law’ stands for?

Around 1938 Frank Benford, a physicist at the General Electrics research laboratories, observed that logarithmic tables were more worn within first pages: was this casual or due to an actual prevalence of numbers near 1 as first digits?

How to use Github with Rstudio : step-by-step tutorial

Sun, 28 Dec 2014 00:10:00 +0000

Pushing to my Github repository directly from the Rstudio project, avoiding that annoying “copy & paste” job. Since it is one of Best Practices for Scientific Computing, I have been struggling for a while with this problem. Now that I managed to solve the problem, I think you may find useful the detailed tutorial that follows. I am not going to explain you the reason why you should use Github with your Rstudio project, but if you are asking this to yourself, you may find useful a **Stack Overflow discussion **on the topic.

Network Visualisation With R

Fri, 05 Dec 2014 15:10:34 +0000

The main reason why

After all, I am still an Internal Auditor. Therefore I often face one of the typical internal auditors problems: understand links between people and companies, in order to discover the existence of hidden communities that could expose the company to unknown risks.

the solution: linker

In order to address this problem I am developing Linker, a lean shiny app that take 1 to 1 links as an input and gives as output a network map:

Querying Google With R

Wed, 19 Nov 2014 04:15:30 +0000

If you have a blog you may want to discover how your website is performing for given keywords on Google Search Engine. As we all know, this topic is not a trivial one.

Problem is that the analogycal solution would be quite time-consuming, requiring you to search your website for every single keyword, on many many pages.

Feeling this way?

[caption id=“attachment_273” align=“aligncenter” width=“300”] “Pain and fear, pain and fear for me” - Oliver Twist[/caption]

Best Practices for Scientific Computing

Wed, 05 Nov 2014 04:40:02 +0000

I reproduce here below principles from the amazing paper Best Practices for Scientific Computing, published on 2012 by a group of US and UK professors. The main purpose of the paper is to “teach” good programming habits shared from professional developers to people that weren’t born developer, and became developers just for professional purposes.

Scientists spend an increasing amount of time building and using software. However, most scientists are never taught how to do this efficiently

Best Practices for Scientific Computing

Write programs for people, not computers.

1. _a program should not require its readers to hold more than a handful of facts in memory at once_


2. _names should be consistent, distinctive and meaningful_


3. _code style and formatting should be consistent_


4. _all aspects of software development should be broken down into tasks roughly an hour long<!-- more -->_

Automate repetitive tasks.

1. _rely on the computer to repeat tasks_


2. _save recent commands in a file for re-use_


3. _use a build tool to automate scientific workflows_

Use the computer to record history.

1. _software tools should be  used to track computational work automatically_

Make incremental changes.

1. _work in small steps with frequent feedback and course correction_

Use version control.

1. _use a version control system_


2. _everything that has been created manually should be put in version control_

Don’t repeat yourself (or others).

1. _every piece of data must have a single authoritative representation in the system_


2. _code should be modularized rather than copied and pasted_


3. _re-use code instead of rewriting it_

Plan for mistakes.

1. _add assertions to programs to check their operation_


2. _use an off-the-shelf unit testing library_


3. _use all available oracles when testing programs_


4. _turn bugs into test cases_


5. _use a symbolic debugger_

Optimize software only after it works correctly.

1. _use a profiler to identify bottlenecks_


2. _write code in the highest-level language possible_

Document design and purpose, not mechanics.

1. _document interfaces and reasons, not implementations_


2. _refactor code instead of explaining how it works_


3. _embed the documentation for a piece of software in that software_

Collaborate.

1. _use pre-merge code reviews_


2. _use pair programming when bringing someone new up to speed and when tackling particularly tricky problems_

if you want to discover more, you can download your copy of Best Practice Scientific Computing here below

download data to excel from web

Tue, 28 Oct 2014 13:02:25 +0000

This simple tutorial will show you how to download data into an excel spreadsheet, creating a web query.

Download data into excel

select “data” tab

select “from web”

input the desidered web URL

click “go” button

select data you want to download

click “import” button

Refresh downloaded data

select “data” tab

excel right() function in R

Mon, 27 Oct 2014 05:29:15 +0000

as part of the** excel functions in R,** I have developed this custom function, reproducing the excel right() function in th R language. Feel free to copy and use it.

[code language=“r”] right = function (string, char){ substr(string,nchar(string)-(char-1),nchar(string))} [/code]

you can find other function in the Excel functions in R post.

excel left() function in R

Mon, 27 Oct 2014 05:27:23 +0000

as part of the excel functions in R, I have developed this custom function, emulating the excel left() function in th R language. Feel free to copy and use it.

left = function (string,char){

substr(string,1,char)}

you can find other function in the** Excel functions in R post**.

excel functions in R

Sat, 25 Oct 2014 22:36:08 +0000

I have started my “data-journey” from Excel, getting excited by formulas like VLookup(), right() and left().

then datasets got bigger, and I discovered that little spreadsheets were not enough, and look for something bigger and stronger, eventually coming to R.

But as you know, ones never forget the first love.

So, for fun and for practice, I have written down some of excel functions in R.

I hope you will enjoy.

Data Visual 10/21: Plotly

Sat, 25 Oct 2014 22:30:47 +0000

Learn dplyr with RStudio and Datacamp

Thu, 23 Oct 2014 12:24:25 +0000

Mining Twitter with R

Thu, 09 Oct 2014 12:16:24 +0000

Great tutorial on text mining with twitter byPaeng Angnakoon

[youtube=http://youtu.be/mJVcANlkxU8]

Answering to Ben ( functions comparison in R)

Sat, 13 Sep 2014 15:05:34 +0000

Following the post about %in% operator, I received this tweet: https://twitter.com/benwhite21/status/510520550553165824

I gave a look to the code kindly provided by Ben and then I asked myself: I know dplyr is a really nice package, but which snippet is faster?

to answer the question I’ve put the two snippets in two functions:

#Ben snippet dplyr_snippet =function(object,column,vector){ filter(object,object[,column] %in% vector) } #AC snippet Rbase_snippet =function(object,column,vector){ object[object[,column] %in% vector,] }

Then, thanks to the great package microbenchmark, I made a comparison between those two functions, testing the time of execution of both, for 100.000 times.

How to Put Equations into Evernote

Thu, 11 Sep 2014 00:00:00 +0000

Problem

Some time ago I was looking for an easy way to put some math writing within my Evernote notes trough my Mac device. Even if there is no official solution to the problem and the feature request is still pending within Evernote dedicated forum, I finally came out with a very simple way to solve your problem out.

Code snippet: subsetting data frame in R by vector

Tue, 02 Sep 2014 08:05:36 +0000

Problem:

you haveto subset a data frame using as criteria the exact match of a vector content.

for instance:

you have a dataset with some attributes, and you have a vector with some values of one of the attributes. You want to make a filter based on the values in the vector.

Example: sales records, each record is a deal.

The vector is a list of selected customers you are interested in.

Saturation with Parallel Computation in R

Mon, 28 Jul 2014 20:14:41 +0000

I have just saturated all my PC:

full is the 4gb RAM

and so is the CPU (I7 4770 @3.4 GHZ)

Parallel Computation in R

which is my secret?

the doParallel package for R on mac

The package lets you make some very useful parallel computation, giving you the possibility to use all the potentiality of your CPU.

As a matter of fact, the standard R option is to use just on of the cores you have got on your PC.

How to Visualize Entertainment Expenditures on a Bubble Chart

Sat, 12 Jul 2014 07:27:24 +0000

I’ve been recently asked to analyze some Board entertainment expenditures in order to acquire sufficient assurance about their nature and responsible.

In response to that request I have developed a little Shiny app with an interesting reactive Bubble chart.

The plot, made using ggplot2 package, is composed by: a categorical x value, represented by the clusters identified in the expenditures population A numerical y value, representing the total amount expended Points defined by the total amount of expenditure in the given cluster for each company subject. Morover, point size is given by the ratio between amount regularly passed through Account Receivable Process and total amount of expenditure for that subject in that cluster.

Mon, 01 Jan 0001 00:00:00 +0000

a quick ride on pagedown: create PDFs from Rmarkdown

celebrating beauty

PaletteR has been staying around for nearly two years, and #rstats user have made a lot of great stuff with it. I therefore took the time to collect what I have found around the web. You can find them below in a slideshow.

Mon, 01 Jan 0001 00:00:00 +0000

I live in Italy, and more precisely in Milan, a city known for fashion and design events. During a lunch break I was visiting the Pinacoteca di Brera, a 200 centuries old museum. This museum is full of incredible paintings from the Renaissance period. During my visit I was particularly impressed from one of them: "La Vergine con il Bambino, angeli e Santi", by Piero della Francesca. If you see this painting you will find a profound of colours with a great equilibrium between different hues, the hardy usage of complementary colours and the ability expressed in the "chiaroscuro" technique. While I was looking at the painting I started, wondering how we moved from this wisdom to the ugly charts you can easily find within today's corporate reports ( find a great sample on the WTF visualization website) This is where Paletter comes from: bring the Renaissance wisdom and beauty within the plots we produce every day.

Introducing paletter

PaletteR is a lean R package which lets you draw from any custom image an optimized palette of colours. The package extracts a custom number of representative colours from the image. Let's try to apply it on the "Vergine con il Bambino, angeli e Santi" before looking into its functional specification.

Installing paletter

Since paletteR is available only trough Github we have to install it using devtools:

library(devtools)
install_github("andreacirilloac/paletter")

Creating a palette from your image

to draw our palette we now need to:

_image_path_

_number_of_colours_

_type_of_variable_

Here it is the code (you can donwload the picture from wikicommons visiting https://it.wikipedia.org/wiki/File:Piero_della_Francesca_046.jpg):

create_palette(image_path = "~/Desktop/410px-Piero_della_Francesca_046.jpg",
               number_of_colors =20,
               type_of_variable = “categorical")

and here it is the output: As you see the palette drawn contains all the most representative colours, like the red of the carpets or the wonderful blue of San Giovanni Battista on the left of the painting.

Functional specification

The main idea behind paletteR code is quite simple:

kmeans

Let 'see how all this works in brief.

Reading a picture into the RGB colourspace

This first step involves transforming the image into an abstract object on which we can apply statistical learning. To do so we read the image file and convert it into a three multidimensional matrix. Within the matrix to each image pixel three numbers are associated:

- one for the quantity of Red - one for the quantity of Green - one for the quantity of Blue All those three attributes range from 0 to 255, as requested by the rules of the RGB colourspace ( find out more on the related RGB colourspace page on wikipedia) . To perform this transformation we use the readJPEG() function from Jpeg package:

painting <- readJPEG(image_path)

this will generate an array having for each point within the image both the cartesian coordinates and the R, G and B values of the related colours. We now apply some statistical learning on the array, to select most representative colours and create an optimized palette.

Processing the RGB image trough kmeans

This processing step was actually the first developed of the package and I already described it in a previous post. Within tht post I devoted the right time to expose some theoretical reference to the kmeans algo and it application to images. Please refere to the How to build a color palette from any image with R and k-means algo post to get a proper explanation of this. You can also read more about this algo and its inner rationales within R for data mining a data mining chrime book. What we need to repeat here is that by applying the kmeans algo on the array we get a list of RGB colours, selected as the most representative of the ones available within the image. I clearly remember my feeling when the first palette came out of kmeans: it was thrilling, but the results were undeemebly poors. I came out for instance with this: What was wrong with the palette employed? We can pick at least three answers:

- there are too bright colours - there are too dark colours - there are too similar colours To summarise: my package was stupid, it was unable to reasonate about relationship among colours avaiable. To solve this problem I moved to the hsv colour space which is the perfect environment were to perform such kind of analyses. the HSV colourspace expresses every colour in terms of:

- Hue which properly expresses the colour, and gets a value from 0 to 360 - Saturation which expresses the quantity of colour (think about a pigment diluted with water to get it). This takes a value from 0 to 100% - Brightness or Value, which express the quantity of grey or white included within the colour. This also takes a value from 0 to 100% The way HSV system describes colours makes it easy to sort colours, moving from 0 to 360, and check for too bright or too dark colours, analysing distributions of saturation and brightness. You can get more on this on the really detailed Wikipedia page of HSV

Moving to the hsv colours space

To convert our RGB object into the HSV space we just need to apply rgb2hsv() to the values of R, G and B.

Removing outliers

What would you do next? After moving within the HSV realm we can now draw meaningful representations of our colour data. What paletteR does as a first step is to produce descriptive statistics for values of Saturation and Value. First of all we calculate quartiles of all of those values:

brightness_stats <- boxplot.stats(sorted_raw_palette$v)
saturation_stats <- boxplot.stats(sorted_raw_palette$s)

Once being done with that we remove the lowest and highest of both. This lets us fix the first two problems observed in the first palette: too bright and too dark colours. What about the third problem?

Optimising palette

To get this solved we have to reasonate about the visual distance of colours. Look for instance at those colours? You would definitely say the first and the second are more distant from each other than the second and the third. You would definitely be right, but how to make our PaletteR as cleaver as you? This is simply done within the HSV space leveraging the Hue attribute. As we have seen HSV hues are placed along a circle in a visually reasonable way. This means that a hue of 40 (which is some kind of orange) is way more distant from a hue of 100 (green) than a hue of 90 is ( another green). Knowing this we just have to select from the first set of colours coming from kmeans a second subset of colours selected as the most _distant_. This will let us avoid employing colours appearing too similar. How to do this? The current version of paletteR does it:

- generating a random sample of possible alternative palettes - measuring the median distance among hues within the palette - selecting the palette showing the greatest distance And here it is below the result for our dear Renaissance painting: Isn't that better than the previous one?

How to apply paletteR in ggplot2

Applying the obtained palette in ggplot is actually easy. The object you obtain from the _create_palette_ function is a vector of hex codes (another way of codify colours, more on the Wikipedia page). You therefore have to pass it to your ggplot plot employing scale_color_manual().A small side note: be sure to select a number of colours equal to the number of variables to plot. Let's apply our palette by Raffaello with an hyphotetical plot:

colours_vector <- create_palette(image_path = image_path,
number_of_colors =32,
type_of_variable = “categorical")
ggplot(data = mtcars, aes(x = rownames(mtcars),y = hp,color = rownames(mtcars),
                          fill = rownames(mtcars))) +
  geom_bar(stat = 'identity') +
  scale_color_manual(values = colours_vector) +
  scale_fill_manual(values=colours_vector)+
  theme_minimal() +
  guides(size = FALSE) +
  theme(legend.position = "bottom") +
  labs(title = "disp vs hp")+
  coord_flip()

Which will produce:

Join us

paletter is quite a young package, nevertheless it already catched some interest (I was also invited to give a speach about it, you can watch it online). This is because of:

- its simple and rather powerful application of statistical learning to the color space - the flexible code - the high number of possible use cases Since it is a young package there is still some work to do on it. I can see at least the following areas where further improvements could be introduced:

- authomatic selection of the type of variables among categorical and continous - computation of the final optimised palette, introducing more advanced measures of colour distance. - code profiling Would you like to give an help on this? Welcome on board! You can find the full code on Github and every contributions is welcome.

Mon, 01 Jan 0001 00:00:00 +0000

W3.CSS

Slideshow Caption

Add a caption text for each image slide with the w3-display-* classes (topleft, topmiddle, topright, bottomleft, bottommiddle, bottomright or middle).

French Alps

Northern Lights

Beautiful Mountains

The Rain Forest

Mountains!

about

Mon, 01 Jan 0001 00:00:00 +0000

brief bio

I am Andrea Cirillo, a Quantitative Analyst currently employed in a bank. My kind of mixed background ( HR guys tended to call it a resilient path, when we talked about it) is composed of a first relevant dose of artistic studies in the Brera High School, followed from an heavy dose of economics and mathematics at the Catholic University of Jesus Sacred Heart, finally crowned from Auditing, Statistics and Data Science Courses by Auditing, Hacking and Banking firms.

contacts

Mon, 01 Jan 0001 00:00:00 +0000

I do not have a general purposes email ( I actually used to have one but it got irredeemably full of spam) but I carefully crafted three main channels to get exposed to all relevant and promising opportunities.

You can find them below:

Proposals of Blog Post/Speaches

author at andreacirillo.com

Collaboration proposals

you can use my linkedin profile to get in touch with me for proposals of every kind of collaboration, either requesting to connect or sending inmail messages: linkedin profile

Deep Work - rules for focused success in adistracted world

Mon, 01 Jan 0001 00:00:00 +0000

a really brief summary

There are two diverging trends in our society:

by one hand in our society tend to succeed those whose can be considered high-skilled like real professionals in data science realm
on the other hand the same society force ourself to an increasingly fragmented style of life, full of shallow stimolous and activites, like reading Facebook notifications or immediately reply to emails

That is why the good author of the book calls for a regaining of the ability to perform Deep Work i.e. the kind of of activities performed in a state of distraction-free concentration that push your cognitive capabilities to their limit. This efforts are the one able to create new value, improve your skill, and are hard to replicate.

paletter speech

Mon, 01 Jan 0001 00:00:00 +0000

On november 20th 2018 I was requested to make a keynote speech at the 9th MilanoR meeting, the most relevant italian meeting of R users.

The main topic of the speech was one of my last coding fatigues: the Paletter package, a new R package able to derive an optimised colour palette from any kind of image.

You can also find a kind reportage of the evening directly on the official MilanoR blog, and the full deck of slides online.

R data mining

Mon, 01 Jan 0001 00:00:00 +0000

R data mining is a data mining crime book, a new kind of book we developed with good guys at Packt and the great statistician Enrico Pegoraro.

What is a data mining crime book? It is a book stuctured to teach you data mining fundamentals while immersing you in a crime case.

Within R data mining you will be immersed into the life of a wholesale company, where suddenly a misterious drop in profits arises. You will be then caught into the analyses performed to discover where this drop comes from and why.

Rstudio for R Statistical computing Cookbook

Mon, 01 Jan 0001 00:00:00 +0000

Rstudio for R Statistical computing Cookbook is an useful contribution to the R community, and especially for those just starting off with R.

The book is a concentrated of recipes targeted to the average R user to get him from being in the mean to the become an outlier. Aside from statistical analogies, this book provides recipes to perform relevant activities within all phases of data analysis projects, and literally:

the Seven Habits of Highly Effective People

Mon, 01 Jan 0001 00:00:00 +0000

a really brief summary

All of the book listed within the bookstack were relevant book for me, but this one was actually crucial for my personal and professional development.

The Visual Display of Quantitative Information

Mon, 01 Jan 0001 00:00:00 +0000

You can negotiate everything

Mon, 01 Jan 0001 00:00:00 +0000

The first time you come across You can negotiate everything by Herb Cohen it can appears one of those books full of techniques to manipulate people and persuade them to do what you want. But if you move beyond this first impression you can find a useful book written by someone who certainly was effective in the pragmatic art of getting someone to agree with you, but was also persuaded that the best way to do this was to build a good relationship based on trust with the other side in every negotiation.

Un Taccuino on Un nuovo Taccuino

Please don’t build your own co‑pilot, and learn to do your job first

Chronicles from a wonderful day in Luxembourg

3 cose che ho imparato in 3 anni da manager

Italy Coronavirus Outbreak: numbers and stats

Italy NCOV-19 outbreak

Introducing the paletteR Gallery

celebrating beauty

a quick ride on pagedown: create PDFs from Rmarkdown

introducing vizscorer: a bot advisor to improve your ggplot plots

introducing vizscorer: a bot advisor to improve your ggplot plots

getting to know the new definition of default

how to use PaletteR to automagically build palettes from pictures

UpdateR package: update R version with a function (on MAC OSX)

Learning Dataviz Principles and Theory from Tufte

Why are you being so silent?

streamline your analyses linking R to sas and more: the workfloweR

ggplot2 themes examples

Euro 2016 analytics: Who's playing the toughest game?

a Checklist for your weekly review (GTD methodology)

Over 50 practical recipes for data analysis with R in one book

2015 in review (let me boast myself a bit :))

Rename a Data Frame Within a Function Passing an Argument

how to list loaded packages in R: ramazon gets cleaver

ramazon: Deploy your Shiny App on AWS with a Function

Introducing Afraus: an Unsupervised Fraud Detection Algorithm

How to add a live chat to your Shiny app

How to list file and folders within a folder ( basic file app)

Catching Fraud with Benford's law (and another Shiny App)

In the early ‘900 Frank Benford observed that ’1’ was more frequent as first digit in his own logarithms manual.

More than one hundred years later, we can use this curious finding to look for fraud on populations of data.

just give a try to the shiny app

What ‘Benford’s Law’ stands for?

How to use Github with Rstudio : step-by-step tutorial

Network Visualisation With R

The main reason why

the solution: linker

Querying Google With R

Best Practices for Scientific Computing

Best Practices for Scientific Computing

Write programs for people, not computers.

Automate repetitive tasks.

Use the computer to record history.

Make incremental changes.

Use version control.

Don’t repeat yourself (or others).

Plan for mistakes.

Optimize software only after it works correctly.

Document design and purpose, not mechanics.

Collaborate.

download data to excel from web

Download data into excel

select “data” tab

select “from web”

input the desidered web URL

click “go” button

**select **data you want to download

click “import” button

Refresh downloaded data

select “data” tab

excel right() function in R

excel left() function in R

left = function (string,char){

substr(string,1,char)}

excel functions in R

Data Visual 10/21: Plotly

Learn dplyr with RStudio and Datacamp

Mining Twitter with R

Answering to Ben ( functions comparison in R)

How to Put Equations into Evernote

Problem

Code snippet: subsetting data frame in R by vector

Problem:

you have a dataset with some attributes, and you have a vector with some values of one of the attributes. You want to make a filter based on the values in the vector.

Saturation with Parallel Computation in R

Parallel Computation in R

How to Visualize Entertainment Expenditures on a Bubble Chart

a quick ride on pagedown: create PDFs from Rmarkdown

Andrea Cirillo

2019-01-27

select data you want to download