SlideShare uma empresa Scribd logo
1 de 59
Baixar para ler offline
How I Data Mined My Text
                 Message History
                 By: Joe Cannatti
                 Puppy Sound LLC
                 joe@puppysound.com
                 @JoeCannatti



Friday, January 18, 13
How I Data Mined My Text
                 Message History
                 By: Joe Cannatti
                 Puppy Sound LLC
                 joe@puppysound.com
                 @JoeCannatti



Friday, January 18, 13
Know thyself...




Friday, January 18, 13
Know thyself...




Friday, January 18, 13
Know thyself...

                    γνῶθι σεαυτόν




Friday, January 18, 13
Know thyself...

                    γνῶθι σεαυτόν

                    Temple of Apollo at Delphi




Friday, January 18, 13
Know thyself...

                    γνῶθι σεαυτόν

                    Temple of Apollo at Delphi

                    Plato




Friday, January 18, 13
Know thyself...

                    γνῶθι σεαυτόν

                    Temple of Apollo at Delphi

                    Plato

                    Our memories are poor




Friday, January 18, 13
Know thyself...

                    γνῶθι σεαυτόν

                    Temple of Apollo at Delphi

                    Plato

                    Our memories are poor

                    We are not good at summing up our experiences




Friday, January 18, 13
Know thyself...

                    γνῶθι σεαυτόν

                    Temple of Apollo at Delphi

                    Plato

                    Our memories are poor

                    We are not good at summing up our experiences

                    Enter.....DATA


Friday, January 18, 13
Text Messages




Friday, January 18, 13
Text Messages

                    Increased use




Friday, January 18, 13
Text Messages

                    Increased use

                    Is it a good subset of your overall communication?




Friday, January 18, 13
Text Messages

                    Increased use

                    Is it a good subset of your overall communication?

                    What can we learn from it?




Friday, January 18, 13
Text Messages

                    Increased use

                    Is it a good subset of your overall communication?

                    What can we learn from it?

                    SCIENCE!!!




Friday, January 18, 13
Tools Used




Friday, January 18, 13
Tools Used

                    R




Friday, January 18, 13
Tools Used

                    R

                    RStudio




Friday, January 18, 13
Tools Used

                    R

                    RStudio

                    Navicat for Sqlite




Friday, January 18, 13
Tools Used

                    R

                    RStudio

                    Navicat for Sqlite

                    Bash




Friday, January 18, 13
So, what are we going
                    to do?



Friday, January 18, 13
iTunes backup

                    Backup your phone to your machine (unencrypted)




Friday, January 18, 13
Where dem bitz at?


                    ~/Library/Application Support/MobileSync/
                    Backup/

                    3d0d7e5fb2ce288813306e4d4636395e047a3d2




Friday, January 18, 13
Hmmm...what else is in
                    there?

                    Bash to the rescue
                    find . -name *3d0d7e5fb2ce288813306e4d4636395e047a3d2*

                    for i in $(find . -type f); do sqlite3 $i ".databases" &> /dev/null; if [[ $? ==
                    "0" ]]; then echo $i; fi; done

                    190 sqlite DBs




Friday, January 18, 13
Bash Fu	




Friday, January 18, 13
Getting ready to open
                    the project

                    Download the RSQLite .tar.gz

                    R CMD INSTALL RSQLite-<version>.tar.gz

                    install.packages(‘DBI’)




Friday, January 18, 13
Tables	




Friday, January 18, 13
Our main view	




Friday, January 18, 13
Adding Contact data




Friday, January 18, 13
A note about R




Friday, January 18, 13
A note about R




Friday, January 18, 13
Problems with the
                    Language




Friday, January 18, 13
Problems with the
                    Language




Friday, January 18, 13
Problems with the
                    Language



                            ANY APL OR J
                         PROGRAMMERS HERE?




Friday, January 18, 13
My besties




Friday, January 18, 13
Result

                             MY BABY MOMA




Friday, January 18, 13
Result

                             MY BABY MOMA




Friday, January 18, 13
Result
                             MARK G’S GIRL
                             MY BABY MOMA




Friday, January 18, 13
Friday, January 18, 13
Friday, January 18, 13
HISTORGRAMS CAN BE PRETTY WACK




Friday, January 18, 13
HISTORGRAMS CAN BE PRETTY WACK

                           THE FREEDMAN-DIACOIS RULE




Friday, January 18, 13
HISTORGRAMS CAN BE PRETTY WACK

                           THE FREEDMAN-DIACOIS RULE




Friday, January 18, 13
Friday, January 18, 13
Demo Frequency




Friday, January 18, 13
Our main Data.Frame




Friday, January 18, 13
colnames(message_data)




Friday, January 18, 13
Codez




Friday, January 18, 13
Text Frequency By
                    Month




Friday, January 18, 13
Demo send/receive



Friday, January 18, 13
Friday, January 18, 13
Saving to a file




Friday, January 18, 13
Let Talk About Sex




Friday, January 18, 13
Let Talk About Sex
                             GET YOUR MIND OUT OF
                                 THE GUTTER!!!




Friday, January 18, 13
Friday, January 18, 13
Friday, January 18, 13
R?




Friday, January 18, 13
R




Friday, January 18, 13
Friday, January 18, 13

Mais conteúdo relacionado

Destaque

Data Warehousing and Data Mining
Data Warehousing and Data MiningData Warehousing and Data Mining
Data Warehousing and Data Miningidnats
 
1.8 discretization
1.8 discretization1.8 discretization
1.8 discretizationKrish_ver2
 
Data cube computation
Data cube computationData cube computation
Data cube computationRashmi Sheikh
 
Data Mining: Classification and analysis
Data Mining: Classification and analysisData Mining: Classification and analysis
Data Mining: Classification and analysisDataminingTools Inc
 
Support Vector Machines for Classification
Support Vector Machines for ClassificationSupport Vector Machines for Classification
Support Vector Machines for ClassificationPrakash Pimpale
 
Data mining (lecture 1 & 2) conecpts and techniques
Data mining (lecture 1 & 2) conecpts and techniquesData mining (lecture 1 & 2) conecpts and techniques
Data mining (lecture 1 & 2) conecpts and techniquesSaif Ullah
 
Data mining slides
Data mining slidesData mining slides
Data mining slidessmj
 

Destaque (10)

Data visualization
Data visualizationData visualization
Data visualization
 
Data Warehousing and Data Mining
Data Warehousing and Data MiningData Warehousing and Data Mining
Data Warehousing and Data Mining
 
1.8 discretization
1.8 discretization1.8 discretization
1.8 discretization
 
Data cube computation
Data cube computationData cube computation
Data cube computation
 
Data Mining: Classification and analysis
Data Mining: Classification and analysisData Mining: Classification and analysis
Data Mining: Classification and analysis
 
Support Vector Machines for Classification
Support Vector Machines for ClassificationSupport Vector Machines for Classification
Support Vector Machines for Classification
 
Data Mining: Association Rules Basics
Data Mining: Association Rules BasicsData Mining: Association Rules Basics
Data Mining: Association Rules Basics
 
Data mining (lecture 1 & 2) conecpts and techniques
Data mining (lecture 1 & 2) conecpts and techniquesData mining (lecture 1 & 2) conecpts and techniques
Data mining (lecture 1 & 2) conecpts and techniques
 
Data mining slides
Data mining slidesData mining slides
Data mining slides
 
Data mining
Data miningData mining
Data mining
 

How I data mined my text message history