Finding Frequently occuring items

r
data_science

#1

I have a dataset with 40,000 rows and it has columns say timer name and user name and event date . we have 335 unique items of timer names. The problem is to find the sequence of pattern of timername column for each user name and patterns are of 2 ,3,4… ab,abc,abcf…Can someone please post R program for above scenario.

Answer :I want to list the output with the timernames involving Ab --> countera 1 ac --> counterb --> 1 af --> counterc -->1 Ab --> countera --> 2… Counter values keep incrementing for the next same pattern occurrence . Final output should be : An user followed (A,b ) for four times which is maximum.


#2

Hi ramya_sai,
Can you please ask this question again in more clear way. From your question what I understand is that you want to find out the frequently occurring items. Please do read about association mining algorithm.

Regards,
Ankit Gupta


#3

Hey Thanks for the reply :slight_smile: @facebook_user_4

I have found the most frequent item in dataset using éclat but I want to find the sets of frequent items.
If we consider A then in how many sequences A has involved like AC AD AF AG AC … in subset 2 and AC occurred 2 times so it is the highest, and ABC , ABD, ACD… of range 3 so on .
In my Dataset , Timer name column has 335 unique names .I need to find out frequent timer names of value 2 or 3 r 4 that will occur . for example

The output should be as below , this is most frequent timernames of size 2 that are occurring maximum times in dataset (say 40 times)
USR:MPG.SOCIAL_HISTORY.O1 - LOAD COMPONENT
USR: DYNDOC EXECUTE AUTOTEXT DOCUMENT

Hope you understood Kindly let me know incase of queries .


#4

@kunal Can you help in solving this ?