Hi Robin, Isabel and Shannon:

Thank you for your responses--I've been learning more about Mahout by
reading the code and the wiki, and running the examples.

While I'm still learning about Mahout, I'd like to elaborate further on my
ideas:

1) Hidden Markov Models: Applications are, as Shannon mentioned, in Spliced
Alignment for Gene Prediction where Viterbi's Algorithm can be applied. The
other application, which I'm investigating for the Bioinformatics course
project could be to predict the side chain structure of proteins. This has
been traditionally done using the rotamer libraries and I think HMMs could
be of use there. I've also noticed that the current HMM does not use
MapReduce, so one possible project would be to provide that functionality,
although it is not trivial...

2) Reinforcement Learning using Markov Decision Process for Recommender
Systems: Solve Optimal Control (Planning) and/or Q Learning. For planning
problems where transition probabilities and reward function is known in
advance, the task remains to solve for optimal policy using the Bellman
Ford/Policy Iteration/Matrix Inversion methods. For cases where this
information is not available a priori, one can implement the Q Learning
algorithm.

I'm also curious where Haloop (http://code.google.com/p/haloop/) stands in
relation to Mahout because a lot of Machine Learning algorithms are
iterative.

Any feedback on my approaches are welcome, especially any information on
fitting HMM/MDPs with MapReduce.

Thank you.
Dhruv.



On Tue, Feb 15, 2011 at 10:43 AM, Shannon Quinn <[email protected]> wrote:

> Hi Dhruv,
>
> > It would be interesting to see how the current implementation can be
> > integrated into your Bioinformatics course work, what extensions you
> > need for this particular domain.
>
> I assume your HMM would be for the purpose of protein/nucleic acid sequence
> analysis and alignment. HMMs/MDPs are fairly domain-independent, but this
> would definitely be an interesting pursuit; I'm a grad student in comp bio,
> and I know a lot of folks who would love to use this.
>
> On the topic of GSoC, I meant to mention: I'm happy to volunteer my
> services as a potential mentor for the summer.
>
> Shannon
>
> Apologies for the brevity, this was sent from my iPhone

Reply via email to