The Data Mining Forum                             open-source data mining software open-source data mining software data science journal data mining conferences
This forum is about data mining, data science and big data: algorithms, source code, datasets, implementations, optimizations, etc. You are welcome to post call for papers, data mining job ads, link to source code of data mining algorithms or anything else related to data mining. The forum is hosted by P. Fournier-Viger. No registration is required to use this forum!.  
FPGrowth_itemsets (Max pattern length)
Posted by: Irfan
Date: March 16, 2020 12:12AM

Thank you for the nice platform which has a lot of implemented algorithms.

In the FPGrowth_itemsets for frequent pattern mining, there is an option for max pattern length, which restricts the number of itemsets generated to be not greater than the value entered.

Are there any means of having Minimum pattern length which will restrict the number of frequent itemsets generated to be not less than the value entered?

I am interested with this feature. Kindly any suggestion(s)

Options: ReplyQuote
Re: FPGrowth_itemsets (Max pattern length)
Date: March 16, 2020 08:15AM

Hi,

Welcome to the forum.

This feature would not be difficult to implement but it will take me maybe 30 min to 1 hour do it. But tomorrow I have many things to do. I will try to implement it tomorrow evening after I finish with my work and will let you know when it is done.

Best regards,

Philippe

Options: ReplyQuote
Re: FPGrowth_itemsets (Max pattern length)
Posted by: Irfan
Date: March 16, 2020 06:48PM

Dear Prof,

Thank you very much for the quick reply.

Kindly waiting for updates.

Hopefully, the release/dowload of SPMF will also be updated with this feature.

Thank you

Options: ReplyQuote
Re: FPGrowth_itemsets (Max pattern length)
Posted by: Irfan
Date: March 16, 2020 09:54PM

Greetings,


I notice the same issue In the RPGrowth_itemsets for rare pattern mining, there is an option for max pattern length, which restricts the number of itemsets generated to be not greater than the value entered.

Also if this will be modified to restrict the number of rare itemsets generated to be not less than the value entered will be helpful. This by having an optional of Min pattern length. The same as I suggested in FPGrowth_itemsets.

Having all these two options (Maximum pattern length and Minimum pattern length) will be helpful. Users can decide which one to use depending on the problem at hand.

Thank you very much.

Options: ReplyQuote
Re: FPGrowth_itemsets (Max pattern length)
Date: March 17, 2020 09:08AM

Good evening,

I have uploaded a new version of SPMF with the minimum pattern for FPGrowth and RPGrowth. I have not updated the webpage yet (the documentation), but you can download the new version from the SPMF website, and start to use it. I will update the documentation later.

Best regards,

Philippe

Options: ReplyQuote
Re: FPGrowth_itemsets (Max pattern length)
Posted by: Irfan
Date: March 17, 2020 05:32PM

Dear Prof,

Thank you very much for adding the suggested features. I tested it with my dataset and it works fine.


Also, I have another observation on RPgrowth, this is not functionality but just to make a clear picture on the label.

On the label "Minsup(%)" i think it coulde be better if renamed as "Maxsup", because when we refer on the documentation of RPgrowth algorithm, it seems that this represent the upper boundary of what considered rare i.e not gretaer than the entered value.


If it will be meaningful, you can consider that suggestion.


Again,

Thank you very much for updated version of both RPgrowth and FPgrowth algorithm

Options: ReplyQuote


This forum is powered by Phorum and provided by P. Fournier-Viger (© 2012).
Terms of use.