Nick

1,052 reputation

1615

bio	website
	location	CA
	age
visits	member for	2 years, 6 months
	seen	May 18 at 0:54
stats	profile views	77

I'm a phd student in computer science.

	1,052 reputation	bio	website		visits	member for	2 years, 6 months
1615 badges		location	CA		seen	May 18 at 0:54

summary answers questions tags badges favorites bounties reputation activity

140 Actions

suggestions reviews revisions comments badges posts accepts all

Dec 27	revised	Most interesting statistical paradoxes improved formatting
Dec 13	comment	Matrix factorization vs random walk with restart for recommender systems How are you planning to integrate friendship links into matrix factorization? Also, it'd be useful to know exactly what kind of data you have available for your task.
Dec 11	revised	How to compile data from annual reports? fixed grammar
Dec 11	suggested	suggested edit on How to compile data from annual reports?
Nov 13	comment	Simple customer response modelling problem Look into collaborative filtering and related techniques for recommendation systems. (for starters have a look at: downloads.hindawi.com/journals/aai/2009/421425.pdf)
Nov 8	awarded	Yearling
Oct 30	comment	Most interesting statistical paradoxes This is a nice example, but I think this is Simpson's Paradox: en.wikipedia.org/wiki/Simpson%27s_paradox
Sep 14	comment	Trouble minimizing perplexity in LDA How many topics have you tried so far (on what size corpus)? Maybe you just haven't yet hit the right number of topics? Also, for inferring the number of topics from data you may want to look into the Hierarchical Dirichlet Process (HDP) with code on David Blei's site: cs.princeton.edu/~blei/topicmodeling.html
Sep 8	comment	How to determine if two sequences are significantly different? It will depend on your specific problem and assumptions, but these links may provide some insight: stats.stackexchange.com/questions/1047/…, and mathworks.com/matlabcentral/newsreader/view_thread/270729
Aug 30	comment	Looking for datasets to practice Text Mining Depends on what kind of text mining you want to do (e.g. topic modelling, parts of speech tagging, ...), but you can download all or part of Wikipedia (dumps.wikimedia.org/enwiki)
Aug 24	awarded	Nice Question
Jul 31	answered	Alternatives to MAP estimator
Jul 17	comment	How to go about selecting an algorithm for approximate Bayesian inference Very interesting points - thanks for the links. I'm wondering more if the structure of the problem could influence one's decision (e.g. method X may be a natural fit for inferring network structure based on node attributes)
Jul 13	asked	How to go about selecting an algorithm for approximate Bayesian inference
Jun 26	comment	Estimation of parameters as a mode of posterior distribution Just to reinforce @Marco's point, the specific estimator one chooses should be driven by the purpose of the model. This is usually defined either implicitly or explicitly by a loss function en.wikipedia.org/wiki/Loss_function.
Jun 22	comment	How to do clustering on monthly collected data? What is the purpose of clustering in this case (i.e. what are you trying to find from this data)? Are you looking for sites with similar attributes over the entire 2yr period? What if some attributes match very well and others do not, would these patterns be of interest as well?
Jun 22	comment	Summarization of correlated but noisy measurement data Do you have a single measurement per device for each experiment, or are measurements taken over time? Do you have a model of the types errors you see for each device?
May 31	comment	Main challenges in data-mining Data mining is an extremely broad field, you might get more informative responses if you specify particular areas that you'd like to learn more about.
May 31	comment	Proposal for transition matrix for Metropolis-Hastings phylogenetic inference If they are independent, you could draw the substitution rates from Beta distributions.
May 30	comment	Proposal for transition matrix for Metropolis-Hastings phylogenetic inference You say that the substitution probabilities are not required to sum to 1, are they independent? is the sum required to be <= 1?