Adriana Vlad, Adrian Mitrea * Estimating Conditional
Probabilities and Digram Statistical Structure in Printed Romanian
Table 4.1. Hierarchy
of digrams in text without blank
Rank
| _ 0
| _ 1
| _ 2
| _ 3
| _ 4
| _ 5
| _ 6
| _ 7
| _ 8
| _ 9
|
0 _
| - | RE
| EA
| AR
| TE
| IN
| DE
| RI
| AT
| ER
|
1 _
| UL
| AL
| ES
| EC
| ÎN
| LE
| IC
| OR
| ST
| EL
|
2 _
| UN
| NT
| CE
| ªI
| RA
| EN
| IA
| IT
| NE
| AC
|
3 _
| TA
| CA
| IE
| TR
| CU
| SE
| TI
| AS
| LA
| PR
|
4 _
| ÞI
| LU
| PE
| DI
| N
| - | - | - | -
| - |
Table 4.2. Hierarchy
of digrams in text with blank
Rank
| _ 0
| _ 1
| _ 2
| _ 3
| _ 4
| _ 5
| _ 6
| _ 7
| _ 8
| _ 9
|
0 _
| - | E-
| I-
| A-
| -C
| -D
| Ã-
| -A
| RE
| -S
|
1 _
| -P
| AR
| TE
| IN
| DE
| N-
| RI
| EA
| -Î
| U-
|
2 _
| ST
| ÎN
| UL
| AT
| ER
| L-
| OR
| CE
| AL
| LE
|
3 _
| -ª
| T-
| RA
| NT
| UN
| -L
| -M
| CU
| ªI
| CA
|
4 _
| NE
| PR
| TA
| TR
| EL
| EN
| -N
| ES
| PE
| R-
|
5 _
| SE
| ÞI
| DI
| ÂN
| - | - | - | -
| - | - |
5. Final remarks
The method we proposed allowed
the p(j/i) conditional probabilities and consequently the
p(i,j) digram probabilities calculation, with an error
control (by means of the confidence intervals). The experimental
results obtained by different methods showed a good concordance,
as expected.
Though the digrams for which the
experimental results could be expressed with sufficient accuracy
(i.e. (S/N) > 20 both for p(i) and p(j/i)
estimation) were about 1/18 of the total possible digram number,
they covered about 54% of the total digram occurrences in the
texts with blanks, or 43% in the texts without blanks.
APPENDIX: List of processed publications (printed Romanian)
- George Cãlinescu, Bietul Ioanide, Edit. Minerva,
Bucureºti, 1995, ISBN 973-21-0432-5:
vol. 1: ISBN 973-21-0431-7, pp. 1-214;
vol. 2: ISBN 973-21-0433-3, pp. 5-256;
vol. 3: ISBN 973-21-0434-1, pp. 5-238.
- Neculai Moghior, Ion Dãnilã,
Leonida Moise, FERDINAND I - Cuvânt pentru întregirea
neamului românesc, pp. 15-198, Edit. Metropol, Bucureºti,
1994, ISBN 973-562-050-2.
- Lucian Blaga, Hronicul
ºi cântecul vârstelor (whole text), Edit.
Minerva, Bucureºti, 1995, ISBN 973-21-0448-1.
- Gala Galaction, Opere 2
(whole text), Edit. Minerva, Bucureºti, 1995, ISBN 973-21-0450-3.
- Gib I. Mihãescu, Rusoaica,
pp. 9-256, Edit. Minerva, Bucureºti, 1995, ISBN 973-21-0443-0.
- Radu Anton Roman, Precum
fumul (whole text), Edit. Cartea Româneascã,
Bucureºti, 1996, ISBN 973-23-0274-7.
- Mircea Stoian, Ultima cursã
de la Struma la Mefküre (whole text), Edit. Hasefer,
Bucureºti, 1995, ISBN 973-96819-6-4.
- Nicolae Volonciu, Tratat
de procedurã penalã, vol. I, Parte generalã,
pp. 7-492 (with the new Romanian orthography), Edit. Paideia,
Bucureºti, 1993, ISBN 973-9131-01-8.
- Nicolae Volonciu, Tratat
de procedurã penalã, vol. II, Parte specialã,
pp. 7-495, Edit. Paideia, Bucureºti, 1994, ISBN 973-9131-24-7.
- Corneliu Zaharia, Îndreptar
de anatomie practicã ºi chirurgicalã a membrelor,
pp. 5-465 (without figure captions and the English and French
prefaces), Edit. Paideia, Bucureºti, 1994, ISBN 973-9131-07-7.
53