This forum is about

A Multi-Objective Evolutionary Approach for Mining Frequent and High Utility Itemsets

Posted by:
**
tianikowa
**

Date: October 07, 2018 11:09AM

Hi, Is there a documentation and java source code for this article?

Title: A Multi-Objective Evolutionary Approach for Mining Frequent and High Utility Itemsets

Author: Lei Zhang Guanglong Fu Fan Cheng Jianfeng Qiu Yansen Su

DOI: http://dx.doi.org/doi:10.1016/j.asoc.2017.09.033

thanks, best regards

Title: A Multi-Objective Evolutionary Approach for Mining Frequent and High Utility Itemsets

Author: Lei Zhang Guanglong Fu Fan Cheng Jianfeng Qiu Yansen Su

DOI: http://dx.doi.org/doi:10.1016/j.asoc.2017.09.033

thanks, best regards

Posted by:
**
webmasterphilfv
**

Date: October 07, 2018 09:41PM

Have you tried to contact the corresponding author of that paper?

Best,

Philippe

Best,

Philippe

Posted by:
**
tianikowa
**

Date: November 29, 2018 07:57AM

Yes, I tried, but I did not get an answer during this time. Is it possible to provide just a numerical example of the MOEA/FHUI algorithm? Please

best regards

best regards

Posted by:
**
webmasterphilfv
**

Date: November 30, 2018 06:37PM

Hi,

I am not the author of this paper so I do not have examples for these algorithms, and I do not have time to read that paper, understand it and write an example for that. But if you are interested, I can tell you that in the next version of SPMF, we will release the code for another paper:

Wei Song, Chaomin Huang. Mining High Utility Itemsets Using Bio-Inspired Algorithms: A Diverse Optimal Value Framework. IEEE Access, 2018, 6(1): 19568-19582

The next release of SPMF containing these new algorithms will be in about 2 weeks.

Regards,

Edited 1 time(s). Last edit at 11/30/2018 06:38PM by webmasterphilfv.

I am not the author of this paper so I do not have examples for these algorithms, and I do not have time to read that paper, understand it and write an example for that. But if you are interested, I can tell you that in the next version of SPMF, we will release the code for another paper:

Wei Song, Chaomin Huang. Mining High Utility Itemsets Using Bio-Inspired Algorithms: A Diverse Optimal Value Framework. IEEE Access, 2018, 6(1): 19568-19582

The next release of SPMF containing these new algorithms will be in about 2 weeks.

Regards,

Edited 1 time(s). Last edit at 11/30/2018 06:38PM by webmasterphilfv.

Posted by:
**
tianikowa
**

Date: November 30, 2018 11:44PM

Thank you for your attention, Yes, I'll definitely be looking for spmf.

But just one question of "get the initial population P with the proposed problem-specific initialize strategy" :

What is the concept of the following sentence?

"the child individuals generated from parent individuals by using cross-mutation operator should be valid and diverse as many as possible so that the convergence of the proposed algorithm could be speeded up."

And why meta-itemset and transaction-itemset are combined together? I have already seen that meta-itemset and vertical-transaction are combined, but the combination of meta-itemset and transaction-itemset is incomprehensible

But just one question of "get the initial population P with the proposed problem-specific initialize strategy" :

What is the concept of the following sentence?

"the child individuals generated from parent individuals by using cross-mutation operator should be valid and diverse as many as possible so that the convergence of the proposed algorithm could be speeded up."

And why meta-itemset and transaction-itemset are combined together? I have already seen that meta-itemset and vertical-transaction are combined, but the combination of meta-itemset and transaction-itemset is incomprehensible

Posted by:
**
webmasterphilfv
**

Date: December 01, 2018 04:34AM

tianikowa Wrote:

-------------------------------------------------------

> Thank you for your attention, Yes, I'll definitely

> be looking for spmf.

> But just one question of "get the initial

> population P with the proposed problem-specific

> initialize strategy" :

>

> What is the concept of the following sentence?

>

> "the child individuals generated from parent

> individuals by using cross-mutation operator

> should be valid and diverse as many as possible so

> that the convergence of the proposed algorithm

> could be speeded up."

To get some accurate answer, it would be better to ask the corresponding author. Otherwise, I would have to read the paper to explain it to you. But currently, I dont have time to do that.

But generally, for genetic algorithm, you have some solutions called "individuals. And then the algorithm will combine the existing solutions (individuals) to generate some new solutions. The process of combining solutions is called mutation in genetic algorithms.

"diverse solutions" means that we don't want solution that are too similar. Because that would mean that the genetic algorithm would be only exploring one part of the search space. If the solution are diverse (not similar to each other), it means that you are exploring a larger part of the search space and have more chance to find the best solution.

As for the "convergence", a genetic algorithm will be applied iteratively. The algorithm neeeds to perform many iterations to find an optimal solution. After several iterations, the algorithm may converge toward some best solution. So In that sentence, the idea is to look for solutions that are diverse to quickly find a near-optimal solution. I think that it is the main idea.

>

> And why meta-itemset and transaction-itemset are

> combined together? I have already seen that

> meta-itemset and vertical-transaction are

> combined, but the combination of meta-itemset and

> transaction-itemset is incomprehensible

This, I don't know. I would have to read the paper. For that you may ask the corresponding author.

Edited 4 time(s). Last edit at 12/01/2018 04:37AM by webmasterphilfv.

-------------------------------------------------------

> Thank you for your attention, Yes, I'll definitely

> be looking for spmf.

> But just one question of "get the initial

> population P with the proposed problem-specific

> initialize strategy" :

>

> What is the concept of the following sentence?

>

> "the child individuals generated from parent

> individuals by using cross-mutation operator

> should be valid and diverse as many as possible so

> that the convergence of the proposed algorithm

> could be speeded up."

To get some accurate answer, it would be better to ask the corresponding author. Otherwise, I would have to read the paper to explain it to you. But currently, I dont have time to do that.

But generally, for genetic algorithm, you have some solutions called "individuals. And then the algorithm will combine the existing solutions (individuals) to generate some new solutions. The process of combining solutions is called mutation in genetic algorithms.

"diverse solutions" means that we don't want solution that are too similar. Because that would mean that the genetic algorithm would be only exploring one part of the search space. If the solution are diverse (not similar to each other), it means that you are exploring a larger part of the search space and have more chance to find the best solution.

As for the "convergence", a genetic algorithm will be applied iteratively. The algorithm neeeds to perform many iterations to find an optimal solution. After several iterations, the algorithm may converge toward some best solution. So In that sentence, the idea is to look for solutions that are diverse to quickly find a near-optimal solution. I think that it is the main idea.

>

> And why meta-itemset and transaction-itemset are

> combined together? I have already seen that

> meta-itemset and vertical-transaction are

> combined, but the combination of meta-itemset and

> transaction-itemset is incomprehensible

This, I don't know. I would have to read the paper. For that you may ask the corresponding author.

Edited 4 time(s). Last edit at 12/01/2018 04:37AM by webmasterphilfv.

Posted by:
**
tianikowa
**

Date: December 01, 2018 08:06AM

Dear Professor, Thank you very much