lpsolve - solver in R

To study:
http://flovv.github.io/From_descritpive_to_prescriptive/
https://icyrock.com/blog/2013/12/linear-programming-in-r-using-lpsolve/
http://lpsolve.r-forge.r-project.org/
http://horicky.blogspot.co.uk/2013/01/optimization-in-r.html
http://lpsolve.sourceforge.net/5.5/R.htm

Marketing with R, very interesting articles.

My search started with the question, how to reduce the huge amount of information about the history of purchases, contacts of hundred thousands of customers to manageable few variables.
I came across  Recency, Frequency, Monetary value (RFM) analysis.

Review articles of these users on R bloggers.

https://www.r-bloggers.com/author/jack-han/
https://www.r-bloggers.com/author/analyze-core-r-language/

ReadingList: Data Bases in R

1. Data Table by Matt Dowle.
Cheatsheet: https://s3.amazonaws.com/assets.datacamp.com/img/blog/data+table+cheat+sheet.pdf
https://www.analyticsvidhya.com/blog/2016/05/data-table-data-frame-work-large-data-sets/

Basics: https://www.dezyre.com/data-science-in-r-programming-tutorial/r-data-table-tutorial
https://www.r-bloggers.com/a-data-table-r-tutorial-by-datacamp-intro-to-dti-j-by/
https://www.r-bloggers.com/elegant-fast-data-manipulation-with-data-table/

Adv: https://www.r-bloggers.com/advanced-tips-and-tricks-with-data-table/
https://www.r-bloggers.com/efficient-aggregation-and-more-using-data-table/
Quick union: https://www.r-bloggers.com/r-quick-tip-collapse-a-lists-of-data-frames-with-data-table/
Aggregation: https://www.r-bloggers.com/two-of-my-favorite-data-table-features/
Binning: https://www.r-bloggers.com/using-data-table-for-binning/

Joining: http://www.magesblog.com/2014/06/the-joy-of-joining-datatables.html#more
Manipulation the fast way: http://www.magesblog.com/2012/11/datatable-rocks-data-manipulation-fast.html#more

Solving common problems: https://www.r-bloggers.com/solve-common-r-problems-efficiently-with-data-table/

Matt Dowle presentations:
http://datascience.la/matt-dowles-data-table-talk-from-user-2014/

Roll:
https://www.r-bloggers.com/understanding-data-table-rolling-joins/

https://www.r-bloggers.com/advanced-tips-and-tricks-with-data-table/

All articles:
https://github.com/Rdatatable/data.table/wiki/Articles

H2O data table?
http://t.co/TYGG6KcMpx
https://www.analyticsvidhya.com/blog/2016/05/h2o-data-table-build-models-large-data-sets/


2. Dplyr by Hadley Wickham... plenty of stuff everywhere (to be elaborated on)

Top reading and exploring on my list. xgboost, ensembles, preprocessing.

1. xgboost - one of Kaggle winner packages, efficient code for home users.
http://www.r-bloggers.com/an-introduction-to-xgboost-r-package/
https://github.com/rachar1/DataAnalysis/blob/master/xgboost_Classification.R
Awesome XGBoost: https://github.com/dmlc/xgboost/blob/master/demo/README.md#features-walkthrough
https://github.com/dmlc/xgboost/blob/master/R-package/vignettes/xgboostPresentation.Rmd
http://wiselily.com/2015/07/12/xgboost-data-mining-example-1/


apply Xgboost in Kaggle with us https://m.youtube.com/watch?v=zwKFyMkvNXE

Tuning parameters:
https://rpubs.com/flyingdisc/practical-machine-learning-xgboost
http://stackoverflow.com/questions/33949735/tuning-xgboost-parameters-in-r
http://www.r-bloggers.com/r-setup-a-grid-search-for-xgboost/
https://www.kaggle.com/c/otto-group-product-classification-challenge/forums/t/14009/script-understanding-xgboost-model-on-otto-dataset
https://www.kaggle.com/tqchen/otto-group-product-classification-challenge/understanding-xgboost-model-on-otto-data/code
http://stats.stackexchange.com/questions/171043/how-to-tune-hyperparameters-of-xgboost-trees
https://www.kaggle.com/c/bnp-paribas-cardif-claims-management/forums/t/19922/how-to-tune-xgboost-using-r
https://www.kaggle.com/forums/f/15/kaggle-forum/t/17120/how-to-tuning-xgboost-in-an-efficient-way
https://www.analyticsvidhya.com/blog/2016/01/xgboost-algorithm-easy-steps/
https://github.com/topepo/caret/issues/336

Plot:
http://rpackages.ianhowson.com/cran/xgboost/man/xgb.plot.tree.html
(off topic, rearange corr) https://drsimonj.svbtle.com/rearrange-your-correlations-with-corrr

 http://www.analyticsvidhya.com/blog/2016/01/xgboost-algorithm-easy-steps/
(also RF: http://www.analyticsvidhya.com/blog/2015/09/random-forest-algorithm-multiple-challenges/)
https://cran.r-project.org/web/packages/xgboost/vignettes/xgboostPresentation.html
https://www.kaggle.com/rajivranjansingh/liberty-mutual-group-property-inspection-prediction/xgboost-in-caret/run/44015/code
https://cran.r-project.org/web/packages/xgboost/vignettes/discoverYourData.html
https://cran.r-project.org/web/packages/xgboost/vignettes/xgboostPresentation.html

2. ensemble learning
http://www.vikparuchuri.com/blog/intro-to-ensemble-learning-in-r/

3. preprocessing
http://topepo.github.io/caret/preprocess.html

4. Another free course:
https://lagunita.stanford.edu/courses/HumanitiesScience/StatLearning/Winter2014/about

5. Deep learning in R resources: https://discuss.analyticsvidhya.com/t/deep-learning-in-r/10349

 Do not read this, My first attempts:

library(xgboost)

mydata<-read.csv(file="clipboard", sep="\t", header=T)

mydata<-sapply(mydata, as.numeric)

#all columns as numeric
mydata<-as.data.frame(lapply(mydata,as.numeric)

#or
mydata<-as.data.frame(sapply(mydata[,1:3], as.numeric))

#label must be 0 or 1 in logistic regression
mydata$Gatunek<-mydata$Gatunek-1


model <- xgboost(data = as.matrix(mydata[,1:2]), label = mydata$Gatunek,
                 nrounds = 2, objective = "binary:logistic")

> install.packages("DiagrammeR")
xgb.plot.tree(model = model)

Free books: Exploring Data Science by John Mount and Nina Zumel

https://dzone.com/articles/practical-data-science-with-r

xgboost, caret, cleaning, time series - a reading list for next weeks.

Go through this to learn xgboost:
https://github.com/dmlc/xgboost/blob/master/R-package/vignettes/discoverYourData.Rmd
https://cran.r-project.org/web/packages/xgboost/vignettes/discoverYourData.html
https://github.com/dmlc/xgboost/blob/master/R-package/vignettes/xgboostPresentation.Rmd

https://github.com/dmlc/xgboost/tree/master/R-package/demo
https://rpubs.com/flyingdisc/practical-machine-learning-xgboost

http://courseprojects.souravsengupta.com/tag/xgboost/
http://xgboost.readthedocs.io/en/latest/R-package/xgboostPresentation.html
https://www.kaggle.com/jkapila/caterpillar-tube-pricing/0-24-with-xgboost-in-r/run/29813
http://www.analyticsvidhya.com/blog/2016/01/xgboost-algorithm-easy-steps/

Material from recent meetup in Warsaw:
https://github.com/mi2-warsaw/SER/blob/master/SER_XIX/xgboost.R


Cleaning and  some interesting packages (editrules, deducorrect):
https://cran.r-project.org/doc/contrib/de_Jonge+van_der_Loo-Introduction_to_data_cleaning_with_R.pdf
http://www.r-bloggers.com/three-quick-and-simple-data-cleaning-helper-functions-december-2013/
http://www.meetup.com/amst-R-dam/events/57161682/ and especially:
http://cran.r-project.org/web/packages/editrules/index.html
http://cran.r-project.org/web/packages/deducorrect/index.html
Very good examples: https://github.com/data-cleaning
 http://www.r-bloggers.com/deductive-imputation-with-the-deducorrect-package/

Time Series:
http://www.r-bloggers.com/time-series-analysis-and-mining-with-r/

Ensemble:
https://cran.r-project.org/web/packages/caretEnsemble/vignettes/caretEnsemble-intro.html

Caret: (very good stuff): https://github.com/pbiecek/DataMining/tree/master/MINI_2015


Other:  
http://www.datasciencecentral.com/m/blogpost?id=6448529%3ABlogPost%3A352821

Github new tools (also offline):
Write a book:  
https://www.gitbook.com/
Collect and share snippets of code:
https://gist.github.com/


 

Excel interfacing R via RScript. (draft)

I played a bit, building a trifle application in Excel using R to explore weather data. My purpose was to make the code portable, with relative links - all files in one folder.

Here is my excel VBA code for exchanging data with R. It starts an R script and then reads back a txt file with numerical results and three plots as svg images.

Sub RunRscript2()
Dim shell As Object: Set shell = VBA.CreateObject("WScript.Shell")
Dim waitOnReturn As Boolean: waitOnReturn = True
Dim windowStyle As Integer: windowStyle = 1
Dim errorCode As Long
Dim sFile As String


'Cleanup
ThisWorkbook.Save
Worksheets("Sheet1").Activate
On Error Resume Next
Worksheets("Sheet1").QueryTables("output1").Delete
On Error GoTo 0
Worksheets("Sheet1").Pictures.Delete
'=========

sFile = ThisWorkbook.Path & "\test_xls_connectivity.R"

errorCode = shell.Run("C:\R\R-3.3.0\bin\x64\Rscript.exe " & sFile & "", windowStyle, waitOnReturn)
'errorCode = shell.Run("Rscript.exe " & sFile & "", windowStyle, waitOnReturn)

Worksheets("Sheet1").Range("$F$3").Select
sFile = ThisWorkbook.Path & "\plots.png"

ActiveSheet.Pictures.Insert(sFile).Select

    With Selection.ShapeRange.Line
        .Visible = msoTrue
        .ForeColor.ObjectThemeColor = msoThemeColorText1
        .ForeColor.TintAndShade = 0
        .ForeColor.Brightness = 0
        .Transparency = 0
    End With
Worksheets("Sheet1").Range("$F$30").Select
sFile2 = ThisWorkbook.Path & "\plots2.png"

ActiveSheet.Pictures.Insert(sFile2).Select

    With Selection.ShapeRange.Line
        .Visible = msoTrue
        .ForeColor.ObjectThemeColor = msoThemeColorText1
        .ForeColor.TintAndShade = 0
        .ForeColor.Brightness = 0
        .Transparency = 0
    End With

Worksheets("Sheet1").Range("$N$3").Select

sFile3 = ThisWorkbook.Path & "\plots3.png"


ActiveSheet.Pictures.Insert(sFile3).Select

    With Selection.ShapeRange.Line
        .Visible = msoTrue
        .ForeColor.ObjectThemeColor = msoThemeColorText1
        .ForeColor.TintAndShade = 0
        .ForeColor.Brightness = 0
        .Transparency = 0
    End With

'''''''''''''''''''''''''
strFile = ThisWorkbook.Path & "\output.csv"

    With ActiveSheet.QueryTables.Add(Connection:= _
        "TEXT;" & strFile & "", Destination:=Range("$N$30"))
        .Name = "output1"
        .FieldNames = True
        .RowNumbers = False
        .FillAdjacentFormulas = False
        .PreserveFormatting = True
        .RefreshOnFileOpen = False
        .RefreshStyle = xlOverwriteCells
        .SavePassword = False
        .SaveData = True
        .AdjustColumnWidth = True
        .RefreshPeriod = 0
        .TextFilePromptOnRefresh = False
        .TextFilePlatform = 1252
        .TextFileStartRow = 1
        .TextFileParseType = xlDelimited
        .TextFileTextQualifier = xlTextQualifierDoubleQuote
        .TextFileConsecutiveDelimiter = False
        .TextFileTabDelimiter = True
        .TextFileSemicolonDelimiter = True
        .TextFileCommaDelimiter = False
        .TextFileSpaceDelimiter = False
        .TextFileColumnDataTypes = Array(1, 1)
        .TextFileTrailingMinusNumbers = True
        .Refresh BackgroundQuery:=False
    End With


End Sub

And here is my R file which reads data from excel and then returns plots as images and reports in text format that VBA will paste in an Excel sheet. The commented install.packages need to be run once in R (maybe it needs to be corrected so that if the packages are not installed R should install them). Note the args... and then scriptPath... piece of the code which reads the path to the current location of the script. Not my solution - I will attach the source soon, sorry and thanks to the original author.

# install.packages("xlsx")
# install.packages("rattle")
# install.packages("RColorBrewer")
# install.packages("rpart.plot")
# install.packages("scatterplot3d")
require(xlsx)
require(rpart)
# require(rattle) 
require(rpart.plot)
require(ggplot2)
require(scatterplot3d)

rm(list=ls())


args <- commandArgs(trailingOnly = F)  
scriptPath <- normalizePath(dirname(sub("^--file=", "", args[grep("^--file=", args)])))
setwd(scriptPath)
# setwd("C:\\Users\\jkotows2\\Desktop\\testR_xlsm")


#read from Sheet1
DaneBadawcze = read.xlsx("test_R.xlsm",1, encoding = "UTF-8", stringsAsFactors=T)

attach(DaneBadawcze)

fit <- rpart(Padało ~ Temperatura + Ciśnienie + Wilgotność,
             ,data=DaneBadawcze, method= "class")
#cluster<-cutree(hclust(dist(DaneBadawcze[,names(DaneBadawcze)!="Padało"]) ),k = 4)

cluster<-cutree(hclust(dist(DaneBadawcze) ),k = 4)

DaneBadawcze$Klaster<-cluster


#print a plot to a file


fn <- "plots.png"
if (file.exists(fn)) file.remove(fn)

png(fn)
#fancyRpartPlot(fit)
prp(fit,type = 2, fallen.leaves = T, extra = 3, compress = T)
dev.off()

#print another plot to a file
fn <- "plots2.png"
if (file.exists(fn)) file.remove(fn)
png(fn)
scatterplot3d(type="h", x=Temperatura, y=Ciśnienie, z=Wilgotność,color=c("green","blue") [DaneBadawcze$Padało] )
dev.off()

#print another plot to a file
fn <- "plots3.png"
if (file.exists(fn)) file.remove(fn)

png(fn)
scatterplot3d(type="h", x=Temperatura, y=Ciśnienie, z=Wilgotność,color=c("green","blue","red","brown","orange") [DaneBadawcze$Klaster] )
dev.off()


write.csv2(fit$variable.importance, file="output.csv", row.names = T)

#print a report to a file
sink(file="report.txt",  type="output")
  fit$variable.importance

  unlink("report.txt")
sink()

To be continued.

5graders learning R / R w 5tej klasie.

Read this: https://benaracreations2015.shinyapps.io/gr5project/
Implement the program in Polish schools!!!
(Found on Twitter @StatGarrett and  @hadleywickham)

RStudio Addins

I am not so fast at memorising/coding... I need parallel code/(some)wysiwyg development. And my memory fails me, I need snippets management... Maybe addins will help soon. Newest RStudio offers addins functionality. Here are some:

https://github.com/daattali/addinslist#addinslist-table

VBA: Simple and excellent example of using R from Excel

Acomed Statistik from Leipzig! Thanks for this very simple example of using R from Excel with VBA - here to calculate data for a kernel density plot. I recommend studying this simple example and will certainly try to adapt it to my needs. http://www.acomed-statistics.com/simple-interface-r-excel_engl.html

Another interesting tutorial to show how to drive R scripts from Excel. http://shashiasrblog.blogspot.com/2013/10/vba-front-end-for-r.html

The quickest setup for R available in Excel (UDF's) functions working on Excel ranges, and the easiest one: http://bert-toolkit.com/
eg. I wrote these in BERT config file (a file in R format) to get clustering functionality

KMEANS_CLUST <- function( mat, ncenters ){ kmeans(mat,centers = ncenters) }


HTREE_CLUST<- function ( mat, cutheight ) {
    myclust<-hclust(dist(mat));
    myclust<-cutree(myclust,h=cutheight);             
    return(myclust)
    }
---

A big idea for kid's math education.

Teach kids statistics? One hour that may decide about their whole life's success measured in their ability to understand, to research, to choose? And learn from them. Understand your limitations. Take an ultimate Uni exam.  Absolutely mind-blowing experience of a project. "Statistics for 8-9-yr. olds or can you weigh a dog with a ruller?" http://smarterpoland.pl/index.php/2016/03/statystyka-dla-8-9-latkow-czyli-jak-zwazyc-psa-linijka/#comment-144319

What next? Here is one of the organizers Przemysław Biecek: https://www.youtube.com/watch?v=UEIcsVhFaSs, pity that it is addressed to Polish audience.You would be enchanted.

What next? R package for kids? Data exploration/machine learning for kids? It is already thought about:
https://github.com/pbiecek/BetaBit

Go SmarterPoland!

R tutorials in Computerworld.

R in 5 lines series by Sharon Machlis: http://www.computerworld.com/tag/rin5linesorless/?nsdr=true#tk.ctw-infsb

Coffee break: Lesson ideas: R and Titanic disaster - trees and forests.

Great ideas for a beginner like me to play with data mining.
Data and challenge proposed by  kaggle.com:
https://www.kaggle.com/c/titanic/dat

Good tutorials and proposed solutins here:
http://trevorstephens.com/post/72916401642/titanic-getting-started-with-r and

https://github.com/trevorstephens/titanic

Also: https://www.kaggle.com/amoyakd/titanic/randomforest-method-v1-0

https://www.kaggle.com/c/titanic/forums

Some reading on R capabilities: 
http://www.edureka.co/blog/implementation-of-decision-tree/
http://www.r-bloggers.com/a-brief-tour-of-the-trees-and-forests/


Other tutorials:
Iris data: http://rischanlab.github.io/RandomForest.html
(R Basics worth exploring: http://rischanlab.github.io/)

http://dni-institute.in/blogs/random-forest-using-r-step-by-step-tutorial/ (might be even better for a start

http://www.edureka.co/blog/implementation-of-decision-tree/

Biostars Tutorial: Machine Learning For Cancer Classification - Part 1 - Preparing The Data Sets
https://www.biostars.org/p/85124/

http://www.tutorialspoint.com/r/r_random_forest.htm

http://www.analyticsvidhya.com/blog/2015/09/random-forest-algorithm-multiple-challenges/

Other important topics to explore:
http://www.analyticsvidhya.com/blog/2015/12/faster-data-manipulation-7-packages/

Small plots on a plot: R

Pie charts instead of points or bubbles in x-y/scatter-plot/bubble plot: Uncle google found this for me:

http://blog.revolutionanalytics.com/2012/09/visualize-complex-data-with-subplots.html
http://www.r-bloggers.com/embeding-a-subplot-in-ggplot-via-subview/
http://vita.had.co.nz/papers/embedded-plots.pdf

ggsubplot

http://stackoverflow.com/questions/26577669/tiny-pie-charts-to-represent-each-point-in-an-scatterplot-using-ggplot2
http://stackoverflow.com/questions/7714677/r-scatterplot-with-too-many-points/16122003#16122003



pieGlyph
http://stackoverflow.com/questions/20465070/barplots-on-a-map/20468836#20468836

Maybe it will be possible in ggvis??
http://ggvis.rstudio.com/ggvis-basics.html


===========
Here a solution without ggplot (p.11):
https://www.r-project.org/doc/Rnews/Rnews_2003-2.pdf

====
plotrix and floating.pie?
http://ask.programmershare.com/2269_10385068/

====
http://gis.stackexchange.com/questions/4568/effectively-displaying-demographic-data-on-a-printed-map

====
ggtree
http://stackoverflow.com/questions/10368180/plotting-pie-graphs-on-map-in-ggplot

mapplots
http://www.molecularecologist.com/2012/09/making-maps-with-r/


Rozwiazanie w Excelu:
http://www.andypope.info/charts/piedatamarkers.htm

Coffee break: imputation

From CranR:
Simple mean imputation: http://www.r-bloggers.com/example-2014-5-simple-mean-imputation/

knnImputation {DMwR} Function that fills in all NA values using the k Nearest Neighbours of each case with NA values.
centralImputation {DMwR} This function fills in any NA value in all columns of a data frame with the statistic of centrality (given by the function centralvalue()) of the respective column.

Other: kNNImpute {imputation} Imputation using k-nearest neighbors. For each record, identify missinng features. http://www.bioconductor.org/packages/release/bioc/html/impute.html
 impute {Hmisc} http://www.inside-r.org/packages/cran/hmisc/docs/impute http://www.jstatsoft.org/article/view/v023i10 http://docs.zeligproject.org/en/latest/using-Zelig-with-Amelia.html

What u play with on smartphone? I play R stats :-)

I followed this tut, thanks! http://www.r-bloggers.com/install-r-in-android-via-gnuroot-no-root-required/ 
1. Installed GnuRoot app and GnuRoot Wheezy (no real rooting required, it creates its own world!)
2. Refreshed software: apt-get update & apt-get upgrade.
3. Installed R: apt-get install R-base-core. Lots of space is needed, so delete all trash apps from your android phone!

===========
I typed R, created a dataframe to play with, run kmeans on it, with no issues. There is no graphics yet, you will not see plots (probably you can generate them as files), unless you install another heavy MB of linux GUI like lxde. But for my bus trips I use txtplot - it produces density, boxplot, scatter, acf in ASCII.
To see kmeans clusters: txtplot(x,y,pch=cluster_no)
R used phone's internet and downloaded it with no issues - it means you can also download dataframes from a repository online.

 A keyboard with arrows is recommended, like Hacker's Keyboard, so that it should be easy to go through previous commands, modify them and correct typo's inside the string..


GnuRoot autor Corbin Champion also offers a solution to run Octave on your phone. I highly recommend him.  https://play.google.com/store/apps/details?id=champion.gnuroot

More: Google: install r in android:
http://www.r-ohjelmointi.org/?p=1434
http://stackoverflow.com/questions/13047994/are-there-any-android-version-of-r-without-rooting-the-device

datascienceplus.com - just joined.

I have just joined . Free R tutorials from contributors. Great idea.

R: xts and zoo package for missing data; two plots

Dealing with incomplete data:

Useful and basic entry into the topic:
http://nicercode.github.io/intro/missing-data.html and
http://thomasleeper.com/Rcourse/Tutorials/NAhandling.html
http://www.ats.ucla.edu/stat/r/faq/missing.htm

==Fill NA==
A very interesting reading on the use of zoo/xts package here (page 1 and the following): http://publish.illinois.edu/spencer-guerrero/2014/12/11/2-dealing-with-missing-data-in-r-omit-approx-or-spline-part-1/
 

My dirty notebook:

> is.na (
> mean(Data&Variable1, na.rm=TRUE)

Remove incomplete
> na.omit(Data$Variable1)
> complete.cases(Data)
> complete.data <- complete.cases(Data)

Fill continuous/time series data
> require (xts)
> prices <-na.locf(prices)
> prices <-na.fill(prices, fill="extend") are there other fills?
check out na.approx, na.spline!!!
na.contiguous (to find longest contiguous na's?) 

For plotting
na.fill etc functions turn data frame into atomic vectors...
It means that if we plot, we cannot refer to column names but to column numbers.

Plot several plots:
> par(mfrow=(c(2,1))
> plot(prices[,1], type='l')
> plot(prices[,2], type='l')




 ====
For correlation
cor ()... use the argument use='pairwise', then only the pairs with na will be ommited and not the whole observations.
 ====
Serious approach: imputation:
http://www.stat.columbia.edu/~gelman/arm/missing.pdf
e.g.: http://stackoverflow.com/questions/11990463/match-fitted-values-from-lm-with-a-data-frame-in-case-of-na-values
https://stat.ethz.ch/pipermail/r-help/2002-January/017756.html

Good article for a study: http://www.r-bloggers.com/imputing-missing-data-with-r-mice-package/
http://research.stowers-institute.org/efg/R/Statistics/loess.htm
http://www.biostat.jhsph.edu/~iruczins/teaching/jf/ch14.pdf
http://www.ats.ucla.edu/stat/r/faq/R_pmm_mi.htm
http://www.stefvanbuuren.nl/publications/MICE%20in%20R%20-%20Draft.pdf
http://stackoverflow.com/questions/2613420/handling-missing-incomplete-data-in-r-is-there-function-to-mask-but-not-remove

Maps of Poland in R

To start with:
http://www.gauss.pl/analityk/drupal/node/43
http://www.kep.uni.lodz.pl/em/materialy/BANKI/BANKI_DANYCH_cw_10.doc

Most comprehensive:
http://quantup.pl/2015/03/13/analiza-danych/wizualizacja-danych-na-mapach-w-r/

http://www.goldenline.pl/grupy/Komputery_Internet/r/mapa-polski,1523596/
http://www.gis-support.pl/baza-wiedzy/dane/dane-do-pobrania/
http://www.wais.kamil.rzeszow.pl/pakiet-r-openpoland-tutorial/

In Excel
http://www.apocotenexcel.pl/mapa2.htm

Day 2 with R. Ideas for diving in deep waters

Start with example here: https://stat.ethz.ch/R-manual/R-devel/library/class/html/knn.html
Then go through: http://blog.datacamp.com/machine-learning-in-r/
========================
Additional reading:
https://en.wikibooks.org/wiki/Data_Mining_Algorithms_In_R/Classification/kNN
https://www3.nd.edu/~steve/computing_with_data/17_Refining_kNN/refining_knn.html
http://blog.webagesolutions.com/archives/1164
http://www.analyticsvidhya.com/blog/2015/08/learning-concept-knn-algorithms-programming/

K means vs K nearest neighbors are two different things.
https://www.quora.com/Is-the-k-Means-algorithm-related-to-the-k-Nearest-Neighbors-algorithm
https://www.quora.com/How-is-the-k-nearest-neighbor-algorithm-different-from-k-means-clustering

What next:
http://rayli.net/blog/data/top-10-data-mining-algorithms-in-plain-r/

Additional topic:
R and SQL/MS SQL
http://www.burns-stat.com/translating-r-sql-basics/
http://www.r-bloggers.com/make-r-speak-sql-with-sqldf/ - use SQL in R
https://www.simple-talk.com/sql/reporting-services/making-data-analytics-simpler-sql-server-and-r/  make R talk to MS SQL

Ideas for day 3: http://courses.had.co.nz/10-tokyo/
http://www.analyticsvidhya.com/learning-paths-data-science-business-analytics-business-intelligence-big-data/learning-path-r-data-science/