Random commentary about Machine Learning, BigData, Spark, Deep Learning, C++, STL, Boost, Perl, Python, Algorithms, Problem Solving and Web Search
Tuesday, March 16, 2010
Document processing with score and cost function
Suppose you have a stream of document that you are evaluating. Each document has a score function s(i) >= 0, and a cost function c(i) >= 0. if you analyze a document i you have to pay the cost c(i) and you get access to the associated score (i). The goal is to maximize the score, but minimize the cost. What strategy would you adopt?
Subscribe to:
Post Comments (Atom)
Use greedy algorithm by maximizing score(i)/cost(i)
ReplyDelete