data mining

[vmishra85]

i have to find the common simillar subsequence among the large
no.s(>21,000) of input sequences, and not neccesarily present in all
the input sequences.subsequence need not be the exact one.it can vary
upto 20%.

can anybody help to design the algorithm for the above problem.
Take it as a challenge.

 

[Vladimir Oka]

i have to find the common simillar subsequence among the large
no.s(>21,000) of input sequences, and not neccesarily present in all
the input sequences.subsequence need not be the exact one.it can vary
upto 20%.

can anybody help to design the algorithm for the above problem.
Take it as a challenge.

This is a question more suited to comp.programming; here you're only
likely to get advice once you try to implement it in C, and experience
C language problems.

PS
I feel something biotechy about it (looking for a gene in different DNA
strands). Maybe there lies an alternative path to the answer. Are there
any Usenet groups dedicated to it?

 

[pemo]

This is a question more suited to comp.programming; here you're only
likely to get advice once you try to implement it in C, and experience
C language problems.

PS
I feel something biotechy about it (looking for a gene in different
DNA strands). Maybe there lies an alternative path to the answer. Are
there any Usenet groups dedicated to it?

My misses is a Population Geneticist, and yes, it sounds like the sort of
thing she does with DNA sequences in C++ ... when casually asked about this,
she replied 'needleman-wunsch' - I didn't seek clarification - it usually
does my head in if I do.

 

[osmium]

i have to find the common simillar subsequence among the large
no.s(>21,000) of input sequences, and not neccesarily present in all
the input sequences.subsequence need not be the exact one.it can vary
upto 20%.

can anybody help to design the algorithm for the above problem.
Take it as a challenge.

There is a newsgroup devoted to the specialty of genetic algorithms, it is
comp.ai.genetic. But I think your chances of getting any detailed help,
even there, are well under 1%.

 

[Juuso Hukkanen]

i have to find the common simillar subsequence among the large
no.s(>21,000) of input sequences, and not neccesarily present in all
the input sequences.subsequence need not be the exact one.it can vary
upto 20%.

Many C data-mining code examples are shown in
http://www.cosc.canterbury.ac.nz/tad.takaoka/manuscript3.pdf


Juuso Hukkanen
(to reply by e-mail set addresses month and year to correct)

 

[Your Uncle]

Many C data-mining code examples are shown in
http://www.cosc.canterbury.ac.nz/tad.takaoka/manuscript3.pdf

I'd love to hear anything you have to say about this topic. I believe it
has forensic dimensions. bfx