Research on Search

With Python 2.2 a new type of classes, named new-style classes , was introduced. New-style classes add some convenient functionality to classic classes. New-style classes are recognized by having...

http://researchonsearch.blogspot.com/2012/08/new-style-python-classes.html

Implementing a dynamic programming algorithm is made much easier in Python by using a caching/memorization decorator .

http://researchonsearch.blogspot.com/2012/08/python-caching-decorators.html

Python's handling of default parameter values becomes tricky when one uses a mutable object (e.g., list or dictionary) as a default value. > Default parameter values are evaluated when ...

http://researchonsearch.blogspot.com/2012/05/pythons-handling-of-default-parameter.html

Python actually has the splat operator (as in Perl or Ruby) that can unpack the arguments out of a list, tuple, or dictionary : func(*args), or func(**kwdargs).

http://researchonsearch.blogspot.com/2012/05/pythons-splat-operator.html

The Hungarian algorithm (aka Kuhn–Munkres algorithm or Munkres assignment algorithm) can solve the assignment problem in polynomial time O(n^3). It can be used to find the optimal mapping from...

http://researchonsearch.blogspot.com/2011/09/hungarian-algorithm-in-clustering.html

What is the most efficient method to check whether an item is in a given group or not? In Python, it seems that set (or frozenset) would be slightly faster than dict and much much faster than lis...

http://researchonsearch.blogspot.com/2011/08/fastest-membership-test-in-python.html

Intuitively, a submodular function over the subsets demonstrates "diminishing returns", which is related to the concept of marginal utility in economics. Its usefulness for machine learning is ...

http://researchonsearch.blogspot.com/2011/08/submodular-functions.html

Andrew Ng has proven in his ICML-2004 paper that sample complexity grows linearly in the number of irrelevant features when using L2 regularisation (in logistic regression, support vector machi...

http://researchonsearch.blogspot.com/2011/08/l1-regularisation-is-efficient-for.html

Juha Kärkkäinen, Peter Sanders , and Stefan Burkhardt: Linear Work Suffix Array Construction , Journal of the ACM (JACM), Volume 53 Issue 6, November 2006. As the authors have said, this algori...

http://researchonsearch.blogspot.com/2011/07/new-linear-time-algorithm-for-suffix.html

The British government's emphasis on the practical impact of research in REF reminds me of Feynman's following words. > Physics  is like sex: sure, it may give some practical > resu...

http://researchonsearch.blogspot.com/2011/07/research-impact-for-ref.html

Pablo Castells , Jun Wang , Ruben Lara , and Dell Zhang are organising an ACM RecSys-2011 workshop on Novelty and Diversity in Recommender Systems (DiveRS) . A special issue of ACM TIST in the...

http://researchonsearch.blogspot.com/2011/06/divers11.html

It is often desirable to measure the dissimilarity or distance between items using a proper metric . Jaccard coefficient can be converted to a metric by by subtracting the Jaccard coefficient f...

http://researchonsearch.blogspot.com/2011/06/couple-of-metrics.html

A very crude, but often good enough, method to achieve parallel processing (e.g., on multi-core computers) is to partition the large input data file into small chunks, run the program to process ...

http://researchonsearch.blogspot.com/2010/10/poor-mans-parallel-processing.html

The choice of the gain and discount function for the popular IR performance measure normalised Discounted Cumulative Gain (nDCG) has been discussed and empirically justified in a CIKM-2009 paper...

http://researchonsearch.blogspot.com/2010/09/ndcg.html

Here is a good tutorial with Matlab examples about Statistical Estimation for Large Numbers of Rare Events (LNRE) .

http://researchonsearch.blogspot.com/2010/08/lnre.html

The VLFeat open source computer vision library that implements popular feature extraction algorithms (such as SIFT , MSER , and quick shift ), clustering algorithms (such as integer k-means , ...

http://researchonsearch.blogspot.com/2010/06/vlfeat-computer-vision-toolbox.html

Locality Sensitive Hashing (LSH) of l-bits is achieved by carrying out l independent random cuts of the Euclidean space: if two data points are in the same side of all these cuts, they are very ...

http://researchonsearch.blogspot.com/2010/06/bloom-filters-and-locality-sensitive.html

It is said that Google's BigTable uses Bloom filters to reduce the disk lookups for non-existent rows or columns.

http://researchonsearch.blogspot.com/2010/05/bloom-filters-and-bigtable.html

It seems that there is currently no suffix tree implementation with Unicode support publicly available online. So I adapted Thomas Mailund's suffix tree implementation in C with a Python binding...

http://researchonsearch.blogspot.com/2010/05/suffix-tree-implementation-with-unicode.html

Given two strings, S of length m and T of length n, their longest common substrings can be found in O(m+n) time using a generalised suffix tree , or in O(mn) time through dynamic programming (e...

http://researchonsearch.blogspot.com/2010/05/longest-common-substring.html

Given the prior probability $p(mu) = mathcal{N}(x_0,sigma_0^2)$ and the likelihood $p(x_1|mu) = mathcal{N}(mu,sigma_1^2)$, the expectation of the posterior probability $p(mu|x_1)$ has a ver...

http://researchonsearch.blogspot.com/2010/03/bayesian-inference-for-gaussian.html

A succinct comparison of data analysis packages including R, Matlab, SciPy, Excel, SAS, SPSS and Stata, can be found here . I recently tried Stata, but found its language syntax ugly and awkward.

http://researchonsearch.blogspot.com/2010/02/comparing-data-analysis-packages.html

Walter Willinger et al. recently published a paper in which the scale-free network model of the preferential attachment type for Internet is said to be a myth, as it is based on fundamentally...

http://researchonsearch.blogspot.com/2009/11/myth-about-internet.html

A pretty striking finding in the WWW'08 paper from Leskovec etc. is that in nearly every network dataset they examined, there are tight but almost trivial communities at very small scales (up to...

http://researchonsearch.blogspot.com/2009/08/large-networks-are-not-modular.html

There are a number of methods in the family of Spectral Graph Partitioning, including the traditional min-cut and various balanced cut criteria (such as ratio-cut, average-cut, normalized-cut and...

http://researchonsearch.blogspot.com/2009/07/spectral-graph-partitioning.html