Slide 1

Slide 1 text

Open Source Data Analysis Toolbox @tnoda [2016-02-02 Tue]

Slide 2

Slide 2 text

Outline Open Source Tools • Two axes Live Demo • Predict purchases of quoted insurance plans

Slide 3

Slide 3 text

Types of Analysis "EIPD 4ZTUFN%FWFMPQNFOU 3 1ZUIPO $MPKVSF 4DBMB $

Slide 4

Slide 4 text

Tool Stack -BOHVBHF -JCSBSJFT &OWJSPONFOU

Slide 5

Slide 5 text

Programming Languages "EIPD 4ZTUFN%FWFMPQNFOU 3 1ZUIPO $MPKVSF 4DBMB $

Slide 6

Slide 6 text

R -BOHVBHF -JCSBSJFT &OWJSPONFOU 3 $3"/ 3TUVEJP

Slide 7

Slide 7 text

Python -BOHVBHF -JCSBSJFT &OWJSPONFOU 1ZUIPO J1ZUIPO OPUFCPPL /VNQZ4DJQZ TDJLJUMFBSO TUBUTNPEFMT 1BOEBT UFOTPSPX $/5, DIBJOFS

Slide 8

Slide 8 text

Clojure (Java) -BOHVBHF -JCSBSJFT &OWJSPONFOU $MPKVSF &NBDT +BWB +7. *ODBOUFS 8FLB .BIPVU ,VSPNPKJ )BEPPQ 4QBSL 4UPSN *OUFMMJ+ 7JN

Slide 9

Slide 9 text

Scala -BOHVBHF -JCSBSJFT &OWJSPONFOU 4DBMB +BWB +7. 8FLB .BIPVU ,VSPNPKJ )BEPPQ 4QBSL 4UPSN &NBDT *OUFMMJ+ 7JN

Slide 10

Slide 10 text

Demo • Python • $200 PC

Slide 11

Slide 11 text

Wrap-up Open source tools • Ad-hoc / System development • Language / Library / Environment Demo • Emacs / ob-ipython • Outliner + iPython notebook • Kaggle