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!.  
Sequential Pattern Mining vesus "Sequential Pattern Mining"
Posted by: zhaozhao
Date: October 30, 2018 07:45PM

HiFournier-Viger.

i would like to ask if you ever hear the concept of "Lag Sequential Analysis".
you may find the inform at

http://methods.sagepub.com/base/download/ReferenceEntry/the-sage-encyclopedia-of-communication-research-methods/i7593.xml


somehow, I think it may have similar effort compared with "Sequential Pattern Mining". both can be used to predict the next behavior of action.

would you agree it or have different opinions?

Options: ReplyQuote
Re: Sequential Pattern Mining vesus "Sequential Pattern Mining"
Date: October 30, 2018 09:11PM

Hi,

There might be some similarities but I think there are some important differences between sequential pattern mining and lag sequence analysis.

Lag sequence analysis (LS):
- it is used to analyse time series (sequences of numbers) and find if there is some correlation between what happen at at time T with what happened at a time T + lag where lag is a parameter.

Sequential pattern mining (SPM):
- is used to analyse sequences of symbols and find subsequences of symbols that appear frequently in a set of sequences.

Differences:
- SPM is for a set of sequences of symbols. LS is for a time series (a sequence of number). T
- In time series, the time interval between consecutive values is assumed to always be the same. In sequences, this may not be true.
- SPM find patterns that appear in many sequences but there is no concept of lag. In LS there is a concept of lag.

Similarties
- You could always convert a time series to sequences of symbols to find sequential patterns in a time series. But there is still no concept of lag.
- You could always use both techniques to do some predictions.

I think that periodic pattern mining is closer to lag sequence analysis than sequential pattern mining.

Just my opinion ;-)



Edited 1 time(s). Last edit at 10/30/2018 09:12PM by webmasterphilfv.

Options: ReplyQuote


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