Matches in SemOpenAlex for { <https://semopenalex.org/work/W658232237> ?p ?o ?g. }
- W658232237 abstract "The most common approaches to analyzing algorithms are worst-case analysis, average-case analysis, and experimental analysis. In most cases, until recently, experimental analysis has not been given priority. The theoretical computer science community still continues to ignore experimental evaluation, although experimental work dominates algorithmic research in most other areas of computer science and related fields such as operations research. This thesis aims to show the importance of experimental approaches through the solution of three different problems, two in the field of computational geometry and one in the field of statistical machine translation. The GeoFilterKruskal algorithm solves the problem of computing geometric minimum spanning trees (GMST). A common tool in tackling clustering problems, GMSTs are an extension of the minimum spanning tree graph problem, applied to the complete graph of a point set. By using well separated pair decomposition, bi-chromatic closest pair computation, and partitioning and filtering techniques, GeoFilterKruskal greatly reduces the total computation required. It is also one of the only algorithms to compute GMSTs in a manner that lends itself to parallel computation; a major advantage over its competitors. The 1-Center problem is an important application in the field of Geographic Information Systems. Finding exact solution to the problem is computationally expensive. In this thesis, we develop an algorithm that uses computational geometric tools to find an approximate solution to the problem, instantaneously. Exhaustive experiments on real world data sets show that our solution incurs a small constant factor approximation. In the field of Statistical Machine Translation, the Minimum Error Rate Training (MERT) is a widely used algorithm for tuning error parameters. Most of the automated translation algorithms use the classical version of MERT, which fails to capture the underlying distribution of the translation lattice. We propose an algorithm which succeeds in overcoming this drawback, without compromising in the running time both in theory and in practice." @default.
- W658232237 created "2016-06-24" @default.
- W658232237 creator A5027433743 @default.
- W658232237 creator A5040154368 @default.
- W658232237 date "2012-01-01" @default.
- W658232237 modified "2023-09-26" @default.
- W658232237 title "Experimental approaches to computational geometric and statistical machine translation problems" @default.
- W658232237 cites W108314478 @default.
- W658232237 cites W108464071 @default.
- W658232237 cites W1555286493 @default.
- W658232237 cites W1595409123 @default.
- W658232237 cites W1600592911 @default.
- W658232237 cites W1714704734 @default.
- W658232237 cites W1916559533 @default.
- W658232237 cites W1965680834 @default.
- W658232237 cites W1970630246 @default.
- W658232237 cites W1972969203 @default.
- W658232237 cites W1979109797 @default.
- W658232237 cites W1981634604 @default.
- W658232237 cites W1988888548 @default.
- W658232237 cites W1992178764 @default.
- W658232237 cites W1993058867 @default.
- W658232237 cites W1995074666 @default.
- W658232237 cites W1996865658 @default.
- W658232237 cites W2000234319 @default.
- W658232237 cites W2001141328 @default.
- W658232237 cites W2008196645 @default.
- W658232237 cites W2009636660 @default.
- W658232237 cites W2011039300 @default.
- W658232237 cites W2022915405 @default.
- W658232237 cites W2024409132 @default.
- W658232237 cites W2024766881 @default.
- W658232237 cites W2027752345 @default.
- W658232237 cites W2029494882 @default.
- W658232237 cites W2032624334 @default.
- W658232237 cites W2043508111 @default.
- W658232237 cites W2053186076 @default.
- W658232237 cites W2053210693 @default.
- W658232237 cites W2056376561 @default.
- W658232237 cites W2056578260 @default.
- W658232237 cites W2066459185 @default.
- W658232237 cites W2067298033 @default.
- W658232237 cites W2068728923 @default.
- W658232237 cites W2070232376 @default.
- W658232237 cites W2071733808 @default.
- W658232237 cites W2074989152 @default.
- W658232237 cites W2075773495 @default.
- W658232237 cites W2078383150 @default.
- W658232237 cites W2083183567 @default.
- W658232237 cites W2083479198 @default.
- W658232237 cites W2086342204 @default.
- W658232237 cites W2087382696 @default.
- W658232237 cites W2089906307 @default.
- W658232237 cites W2093241092 @default.
- W658232237 cites W2094974501 @default.
- W658232237 cites W2097004206 @default.
- W658232237 cites W2097944783 @default.
- W658232237 cites W2099918640 @default.
- W658232237 cites W2101105183 @default.
- W658232237 cites W2103994377 @default.
- W658232237 cites W2104228215 @default.
- W658232237 cites W2105891181 @default.
- W658232237 cites W2106936040 @default.
- W658232237 cites W2110660003 @default.
- W658232237 cites W2111142112 @default.
- W658232237 cites W2114405797 @default.
- W658232237 cites W2118269922 @default.
- W658232237 cites W2121864884 @default.
- W658232237 cites W2124807415 @default.
- W658232237 cites W2129412583 @default.
- W658232237 cites W2131237536 @default.
- W658232237 cites W2131796944 @default.
- W658232237 cites W2136214327 @default.
- W658232237 cites W2137143056 @default.
- W658232237 cites W2140190241 @default.
- W658232237 cites W2143709779 @default.
- W658232237 cites W2145519405 @default.
- W658232237 cites W2146574666 @default.
- W658232237 cites W2149906774 @default.
- W658232237 cites W2150926065 @default.
- W658232237 cites W2154356512 @default.
- W658232237 cites W2154648702 @default.
- W658232237 cites W2159358338 @default.
- W658232237 cites W2160218441 @default.
- W658232237 cites W2161854574 @default.
- W658232237 cites W2165142526 @default.
- W658232237 cites W2165558283 @default.
- W658232237 cites W2167081989 @default.
- W658232237 cites W2167816765 @default.
- W658232237 cites W2169279899 @default.
- W658232237 cites W2169755259 @default.
- W658232237 cites W2169910125 @default.
- W658232237 cites W2171394996 @default.
- W658232237 cites W2295428206 @default.
- W658232237 cites W2427881153 @default.
- W658232237 cites W2437005631 @default.
- W658232237 cites W2606195968 @default.
- W658232237 cites W3145128584 @default.
- W658232237 cites W91928571 @default.
- W658232237 cites W1534270800 @default.