"> ">
Matrix
Compression Techniques
These are sparse
parser table compression techniques. I've modified it to be used
in
matrices. These are the pseudocodes ! not the programs!
Graph coloring scheme
for i=0 to m-1
for j =0 to n-1
if T[i][j] == insignificant
sigmap[i][j] = 0
else
sigmap[i][j] = 1
color = 1
for i=0 to m-1
for i2=0 to m-1
for j=0 to n-1
if T[i][j] == T[i2][j] && T[i][j] != insignificant && T[i2][j] != insignificant
f = 1
else
f = 0
rowmap[i2] = color
if f == 1
if rowmap[i] == color
rowmap[i2] = color-1
for j=0 to n-1
newT[rowmap[i2]][j] = T[i][j]
else
rowmap[i2] = color
for j=0 to n-1
newT[rowmap[i2]][j] = T[i][j]
else
if f == 0
color++
rowmap[i2] = color
for j=0 to n-1
newT[rowmap[i2]][j] = T[i][j]
for j=0 to n-1
for j2=0 to n-1
for i=0 to m-1
if T[i][j] == T[i][j2] && T[i][j] != insignificant && T[i][j2] != insignificant
f = 1
else
f = 0
columnmap[j2] = color
if f == 1
if columnmap[j] == color
columnmap[j2] = color-1
for i=0 to m-1
value[columnmap[i]][j2] = T[i][j]
else
columnmap[j2] = color
 hirschberg/