README for /home/uav/correlation_matrix Created by Melanie Martin on 4-09-02 This directory contains the correleation matricies by team, by mission created from the corpus: /home/uav/lsa/uavcorpus-fold.doc The program to run the matricies and the index file Command line: python run_martix.py index This program reads the index file, which contains the beginning and ending document numbers for each mission of each team, and calls syn on the range on documents. syn -S -d [range] -S -d [range] > cormatrixXXYY where [range] indicates the beginning_doc_#-ending_doc_# -------------------------------------------------------------------------- Checked by running syn directly and comparing the files: gortyn[114]% syn -S -d 3230-3557 -S -d 3230-3557 > test0107 gortyn[115]% diff cormatrix0107 test0107 gortyn[116]% syn -S -d 8488-8834 -S -d 8488-8834 > test0402 gortyn[117]% diff cormatrix0402 test0402 gortyn[118]% -------------------------------------------------------------------------- Details of index file: start end msgs team <-- line ommitted 1 2257 2257 0000 <-- line ommitted 2258 2569 312 0103 . . 22294 22565 272 1106 22566 22802 237 1107 Here "team" is in the following format: XXYY where XX is the team number and YY is the mission number ============================================================================= 05-07-02 Made directory /home/uav/correlation_matrix/square_matrices that contains the correlation matricies in square, tab-separated format These are the results of running python process_format_matrix.py index -v which calls: format_matrix.py both programs are in this directory Some spot checking done to make sure that the cormatrix and the sqcormatrix files correspond and to make sure that there are the right number of lines: gortyn[144]% wc -l sqcormatrix0607.txt 350 sqcormatrix0607.txt gortyn[145]% wc -l sqcormatrix1101.txt 212 sqcormatrix1101.txt gortyn[146]% wc -l sqcormatrix1107.txt 237 sqcormatrix1107.txt