Title :
A new optimal algorithm for the solution of a generalised assignment problem-application in automatic text alignment
Author :
Cranias, Lamhros
Author_Institution :
Inst. for Language & Speech Process., Athens, Greece
Abstract :
A new optimal algorithm is presented, which is a modification of the Viterbi algorithm (VA), that produces the optimum (in a maximum likelihood sense) set of connections between the entries of two not necessarily aligned vectors. Each connection may treat single entries of the two vectors or groups of (not necessarily consecutive) entries, and therefore the two vectors need not be of the same length. The proposed algorithm is described, its complexity is evaluated and potential applications are proposed from the area of natural language processing (NLP)
Keywords :
computational complexity; maximum likelihood estimation; probability; search problems; Viterbi algorithm; automatic text alignment; generalised assignment problem; maximum likelihood; natural language processing; optimal algorithm; Dynamic programming; Joining processes; Natural language processing; Natural languages; Speech processing; Viterbi algorithm;
Conference_Titel :
Systems, Man and Cybernetics, 1995. Intelligent Systems for the 21st Century., IEEE International Conference on
Conference_Location :
Vancouver, BC
Print_ISBN :
0-7803-2559-1
DOI :
10.1109/ICSMC.1995.538214