Comparison: Conference Proceedings from ICALT in 2011 and 2012
Set A=ICALT 2011 Set B=ICALT 2012
Run Parameters: min term freq=0.0013; min docs=4; max p=0.001
Summary for ICALT 2011 :  Docs= 201  Terms= 16127
Summary for ICALT 2012 :  Docs= 236  Terms= 21276
Eliminating terms appearing in < 1  documents
Summary for ICALT 2011 :  Docs= 201  Terms= 16127
Summary for ICALT 2012 :  Docs= 236  Terms= 21276
Merging the terms lists -> 29931 terms
Will eliminate terms with joint (A+B) set freq <0.13%, which equates to 707 term occurrences (aggregated over both sets)
Will eliminate terms with freq in either A or B >2% which equates to removing 1 and 1 terms.
Now using 143 terms
Pre-sig test freqs Min.   : 707  
Pre-sig test freqs 1st Qu.: 825  
Pre-sig test freqs Median :1015  
Pre-sig test freqs Mean   :1355  
Pre-sig test freqs 3rd Qu.:1454  
Pre-sig test freqs Max.   :7922  
35 terms meet the probability criterion
Which split into two sets according to the dominant occurrence. A-terms: 22 ; B-terms 13
Term Co-occurrence calculation and Gephi data creation
Significant Term Co-occurrence Stats for Set A
Min.=12
1st Qu.=47
Median=69
Mean=67.93
3rd Qu.=86.5
Max.=147
Only exporting co-occurrence edges >= the 0.75 Quartile in weighting (wt>= 86.5 )
Term Co-occurrence calculation and Gephi data creation
Significant Term Co-occurrence Stats for Set B
Min.=29
1st Qu.=51.5
Median=68.5
Mean=68.65
3rd Qu.=78.75
Max.=113
Only exporting co-occurrence edges >= the 0.75 Quartile in weighting (wt>= 78.75 )