How I can generate transactional synthetic data?

Question

I am working on Association Rules so I need transactional dataset which is unavailable on UCI repository so I need to generate transactional data. Transactional data is a set of transactions and each transaction have subset of items. Groceries data is a example of Transactional database. Let D be a transactional database and T be a transactions t={t1,t2,t3 ...... tn} and I be a set of items I={i1,i2,i3, ..... im} then transactional data looks like

TID Items
001 i1,i2,i5
002 i5,i6,i8,i10
003 i1,i4
004 i6,i4,i8
Thanks

I'm not understanding your question, can you try to rephrase? — winhowes, Mar 10 '15 at 05:09

score 0 · Answer 1 · answered Mar 10 '15 at 18:59

So based off of your definition what it looks like you're trying to do is generate a two dimensional array. In JavaScript you could do something like this:

var t = 5, d = [], r = 10, s = 10; 
for(var i=0; i<t; i++){
    d.push([]);
    for(var j=0; j<Math.random()*r; j++){
        d[i].push("i"+Math.floor(Math.random()*s))
    }
}

Here we let t be the number of transactions, d be the two dimensional array of transactions, r be the maximum number of transactions in row i and s be the maximum value of some number in the string (i+someNumber). Running the above and printing out d (console.log(d)) could give you something like this:

0 i3, i8
1 i5, i6, i8
2 i1, i2, i5
3 i3, i8 
4 i9, i1, i7, i3, i5

score 0 · Answer 2 · answered Jul 01 '20 at 00:04

0

Here is an opensource app that leverages the R package conjurer to generate transactional data.

answered Jul 01 '20 at 00:04

SidharthMacherla

390
1
8

How I can generate transactional synthetic data?

2 Answers2