Adriana Vlad, Adrian Mitrea * Estimating Conditional Probabilities and Digram Statistical Structure in Printed Romanian



Table 2.2. Letter probabilities estimation in text with blanks
Letter
i 		Probability
estimate 		Signal to
noise ratio 		Relative
error 		Confidence limits *) Relative
frequency
upper lower
- 0.1607 120.3 0.0163 0.1634 0.1581 0.1620
E 0.1028 93.1 0.0211 0.1050 0.1007 0.1023
I 0.0890 86.0 0.0228 0.0911 0.0870 0.0885
A 0.0823 82.3 0.0238 0.0842 0.0803 0.0822
R 0.0624 70.9 0.0276 0.0641 0.0607 0.0623
N 0.0546 66.1 0.0297 0.0562 0.0530 0.0541
T 0.0533 65.2 0.0301 0.0549 0.0517 0.0527
U 0.0505 63.4 0.0309 0.0521 0.0490 0.0496
C 0.0450 59.7 0.0328 0.0465 0.0436 0.0438
L 0.0387 55.2 0.0355 0.0401 0.0374 0.0403
S 0.0371 54.0 0.0363 0.0385 0.0358 0.0351
O 0.0347 52.1 0.0376 0.0360 0.0334 0.0355
Ã0.0283 46.9 0.0418 0.0295 0.0271 0.0284
D 0.0276 46.4 0.0423 0.0288 0.0265 0.0279
P 0.0264 45.3 0.0433 0.0275 0.0253 0.0265
M 0.0240 43.1 0.0455 0.0251 0.0229 0.0239
Î0.0103 28.1 0.0697 0.0111 0.0096 0.0101
F 0.0101 27.8 0.0706 0.0108 0.0094 0.0101
ª0.0101 27.8 0.0705 0.0109 0.0094 0.0107
Þ0.0094 26.7 0.0733 0.0101 0.0087 0.0096
V 0.0089 26.0 0.0753 0.0096 0.0082 0.0094
G 0.0078 24.4 0.0804 0.0085 0.0072 0.0079
B 0.0073 23.6 0.0830 0.0080 0.0067 0.0077
Z 0.0060 21.4 0.0916 0.0066 0.0055 0.0062
Â0.0053 20.1 0.0975 0.0059 0.0048 0.0054
H 0.0033 (15.9) - - - 0.0034
J 0.0017 (11.5) - - - 0.0018
X 0.0016 (10.9) - - - 0.0018
K 0.0003 - - - - 0.0003
W 0.0002 - - - - 0.0002
Y 0.0002 - - - - 0.0003
Q 0.0000 - - - - 0.0000

*) Calculated on the whole X text (on correlated data).

50

Previous Index Next