The Data Mining Forum                             open-source data mining software data science journal data mining conferences high utility mining workshop
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!.  
Ph.D topics in data mining or big data
Posted by: David
Date: September 07, 2018 04:18AM

Hi,


Looking for a topic in data mining or big data?

Please help if you have good topics for Ph.D. study.


Kind regards,
David

Options: ReplyQuote
Re: Ph.D topics in data mining or big data
Date: September 09, 2018 04:02PM

Hi,

There is a lot of possible topics. You can choose to work on something more fundamental like algorithm design or something more applied such as how to best solve a given applied problem.

A good way of choosing a research problem is to look at some recent papers and find something that you are interested in. Personally, I am quite interested in pattern mining problems and algorithm design so this is usually the kind of topics that I choose in my research.

But it can take some time to find a good topic.

Best regards,

Options: ReplyQuote
Re: Ph.D topics in data mining or big data
Posted by: David
Date: October 16, 2018 12:33AM

Thanks,
I am Also interested in pattern mining problems and algorithm design.

It is so appreciated if you can suggest me some Ph.D. topics in this field.

Best

Options: ReplyQuote
Re: Ph.D topics in data mining or big data
Date: October 17, 2018 09:45PM

In pattern mining, there is a lot of possible topics:
- new applications of algorithms
- designing faster algorithms for an existing pattern mining problem (you can choose a problem and find a way of improving the performance)
- designing a new problem for pattern mining (you need to think about some limitations of the current pattern mining problems and think about something new that you can add to these problems, and then design an algorithm and show that your new problem is useful. A simple way of creating a new problem is to combine two problems together. For example, if there is a problem of sequential rule mining and another problem of mining patterns in a stream, you could combine to do sequential rule mining in a stream. It would be a new problem. But you need to make sure the problem is not too easy and is useful.

Also, a good start is to check my SPMF data mining software. There is a lot of code for existing algorithms. If you start from an existing algorithm where you already have some code and datasets, you can save a lot of time. Also you can learn from the code.

Hope this is helpful
Best,

Philippe

Options: ReplyQuote


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