A Probabilistic Modeling Study of the Dynamics of Discourse Expression and the Construction of Discourse Power in an English News Corpus
19 mar 2025
INFORMAZIONI SU QUESTO ARTICOLO
Pubblicato online: 19 mar 2025
Ricevuto: 05 ott 2024
Accettato: 02 feb 2025
DOI: https://doi.org/10.2478/amns-2025-0368
Parole chiave
© 2025 Wanni Mo, published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Figure 1.
Figure 2.
Figure 3.
Figure 4.
Figure 5.
Figure 6.
The 10 features of the two groups of corpus differences
| Serial number | Feature name | World customs organization news | BROWNNews library | P value | Absolute difference |
|---|---|---|---|---|---|
| 1 | PHC | 5.7663 | 2.4081 | 0.001 | 3.28962 |
| 2 | NOMZ | 2.9353 | 0.4015 | 0 | 2.46538 |
| 3 | AWL | 2.8208 | 0.7048 | 0 | 2.0475 |
| 4 | DWNT | -0.766 | 0.4648 | 0 | 1.16432 |
| 5 | [SMP] | -0.7083 | 0.3648 | 0 | 1.00659 |
| 6 | [WZPRES] | 1.3081 | 0.3715 | 0.017 | 0.86811 |
| 7 | SYNE | -0.7901 | 0.0481 | 0 | 0.77174 |
| 8 | [PUBV] | -0.991 | -0.1885 | 0 | 0.73598 |
| 9 | RB | -2.7892 | -2.0485 | 0.001 | 0.67417 |
| 10 | TIME | -0.9315 | -0.1985 | 0 | 0.66644 |
The session of the context discourse is the result of building the experiment
| Folds | Performance measures | T-test | Goodness of fit test | |||||
|---|---|---|---|---|---|---|---|---|
| TP | sensitivity | TN | specificity | accuracy | Sig. | χ2 | Sig. | |
| fold 1 | 53 | 0.684 | 48 | 0.661 | 0.674 | 0.504 | 13.442 | 0.002 |
| fold 2 | 48 | 0.608 | 44 | 0.591 | 0.591 | 0.631 | 4.522 | 0.055 |
| fold 3 | 46 | 0.584 | 50 | 0.652 | 0.611 | 0.333 | 6.003 | 0.016 |
| fold 4 | 43 | 0.554 | 42 | 0.537 | 0.554 | 0.885 | 0.974 | 0.347 |
| Total | 190 | 0.608 | 184 | 0.610 | 0.608 | 0.588 | 24.941 | 0.000 |
The best number is the characteristics and their coefficients in F1
| Feature | The Explanation of Feature | Coefficient |
|---|---|---|
| WRDFRQmc | CELEX Log minimum frequency for content words, mean | 0.25745 |
| WRDHYPv | Hypernymy for verbs, mean | 0.172133 |
| WRDFRQc | CELEX word frequency for content words, mean | 0.1534 |
| DESWLltd | Word length, number of letters, standard deviation | -0.13745 |
| DESWLlt | Word length, number of letters, mean | 0.098632 |
| WRDHYPnv | Hypernymy for nouns and verbs, mean | -0.06452 |
| DESWLsyd | Word length, number of syllables, standard deviation | -0.05325 |
| WRDPOLc | Polysemy for content words, mean | 0.048352 |
| WRDHYPn | Hypernymy for nouns, mean | -0.04154 |
| DESWLsy | Word length, number of syllables, mean | -0.03544 |
| WRDFRQa | CELEX Log frequency for all words, mean | 0.027742 |
| LDTTRc | Lexical diversity, type-token ratio, content word lemmas | 0.009453 |
| LDMTLD | Lexical diversity, MTLD, all words | -0.0053 |
| LDTTRa | Lexical diversity, type-token ratio, all words | -0.00357 |
| WRDFAMc | Familiarity for content words, mean | -0.0027 |
| WRDMEAc | Meaningfulness, Colorado norms, content words, mean | 0.002464 |
| WRDIMGc | Imagability for content words, mean | -0.00225 |
Multidimensional analysis of the white corpus
| Brown sublibrary | Dimension 1 | Dimension 2 | Dimension 3 | Dimension 4 | Dimension 5 | Dimension 6 | The closest text type |
|---|---|---|---|---|---|---|---|
| News report | -17.96 | 0.5 | 4.61 | -1.74 | 0.89 | -1.27 | Academic article |
| editorial | -12.74 | -0.22 | 4.64 | 1.24 | 0.7 | -0.4 | Universal narrative |
| News review | -15.36 | -0.98 | 5.4 | -3.5 | 0.44 | -1.21 | Academic article |
| religion | -8.38 | 0.3 | 5.18 | 0.29 | 2.15 | 0.37 | Universal narrative |
| Skills, business and hobbies | -13.37 | -2.18 | 4.58 | -0.89 | 1.54 | -1.2 | Universal narrative |
| Social life | -14.7 | 0.4 | 4 | -0.96 | 1.51 | -0.77 | Universal narrative |
| Biographies and essays | -12.48 | 1.24 | 5.1 | -0.94 | 1.49 | -0.25 | Universal narrative |
| Government document | -17.68 | -2.64 | 8.39 | 0.54 | 2.67 | -0.13 | Academic article |
| Academic paper | -13.13 | -1.95 | 5.82 | -1.09 | 4.48 | -0.15 | Scientific article |
| General novel | -7.35 | 6.29 | 0.32 | -0.4 | -0.36 | -1.34 | Universal narrative |
| Detective story | -1.75 | 6.28 | -1.29 | 0.29 | -0.99 | -1.08 | The fantasy narrative |
| Science fiction | -3.49 | 5.46 | 1.33 | 0.2 | 0.82 | -0.89 | Universal narrative |
| Adventure and western fiction | -5.26 | 6.54 | -0.76 | -1.7 | -1.03 | -1.48 | Universal narrative |
| Love fiction | 0.11 | 6.43 | 0.55 | -0.16 | -1.1 | -1.22 | The fantasy narrative |
| Humor | -6.78 | 3.6 | 2.69 | -1.07 | 0.5 | -0.51 | Universal narrative |
| Library | -9.43 | 2.14 | 4.55 | -0.86 | 0.77 | -0.79 | Universal narrative |
Multidimensional analysis of the world customs organization news corpus
| The world customs organization news sub-library | Dimension 1 | Dimension 2 | Dimension 3 | Dimension 4 | Dimension 5 | Dimension 6 | The closest text type |
|---|---|---|---|---|---|---|---|
| Views | -20.4 | -3.54 | 10.4 | -0.45 | 3.06 | -0.76 | Academic article |
| Globe | -23.84 | -3.04 | 9.32 | -2.56 | 2.46 | -1.75 | Academic article |
| Book review | -22.19 | -1.71 | 8.25 | -2.39 | 3.72 | 0.09 | Academic article |
| Flocculus | -24.62 | -4.56 | 12.57 | -2.85 | 0.25 | -2.21 | Academic article |
| Close-up | -28.47 | -3.76 | 14.06 | -3.66 | 2.7 | -1.96 | Academic article |
| File | -22.46 | -4.14 | 11.49 | -1.24 | 3.5 | -1.78 | Academic article |
| Editor’s note | -17.02 | -3.91 | 10.23 | 3.12 | 0.74 | 0.94 | Academic article |
| Event | -23.37 | -4.23 | 11.91 | -2.69 | 1.7 | -1.43 | Academic article |
| Express | -23.05 | -4.43 | 11.13 | -2.77 | 1.32 | -1.56 | Academic article |
| Focal point | -19.85 | -3.33 | 10.2 | -0.21 | 3.6 | -1.22 | Academic article |
| Moderator | -19.91 | -4.81 | 12.7 | 1.02 | 0.74 | -1 | Academic article |
| Dialogue | -18.31 | -3.49 | 10.73 | 0.58 | 2.53 | -0.38 | Academic article |
| File | -22.85 | -3.95 | 14.25 | -0.73 | 2.11 | -0.94 | Academic article |
| Latest report | -22.26 | -1.8 | 10.27 | -2.55 | 1.2 | -0.79 | Academic article |
| Noncolumn name | -17.09 | -5.98 | 14.88 | -6.15 | -1.61 | -2.7 | Academic article |
| Member customs | -24.2 | -4.42 | 11.64 | -2.73 | 1.45 | -1.72 | Academic article |
| Panoramic view | -22.37 | -3.35 | 11.28 | -1.27 | 3.32 | -1.53 | Academic article |
| Publications | -26.31 | -6.65 | 12.73 | -6.92 | -0.61 | -2.62 | Academic article |
| Reader | -23 | -3.78 | 10.82 | -2.48 | 0.97 | -1.42 | Academic article |
| Special report | -21.72 | -3.7 | 12.23 | -1.2 | 2.61 | -1.2 | Academic article |
| Training log | -26.88 | -4.99 | 15.78 | -2.71 | -2.29 | -2.06 | Academic article |
| Focusing | -25.86 | -3.7 | 15.84 | -3.45 | 1.03 | -2.17 | Academic article |
| Library | -21.21 | -4.35 | 13.65 | -2.14 | 3.22 | -2.33 | Academic article |
The comparison of the context discourse and the non-contextual discourse
| Folds | Performance measures | Goodness of fit test | Effect comparison | Matched T-test | ||
|---|---|---|---|---|---|---|
| accuracy | χ2 | |||||
| with C. | without C. | with C. | with C. | Sig. | ||
| fold 1 | 0.653 | 0.596 | 13.454 | 4.325 | 1>2 | 0.108 |
| fold 2 | 0.612 | 0.671 | 5.255 | 17.543 | 1<2 | 0.067 |
| fold 3 | 0.607 | 0.604 | 7.000 | 5.224 | 1>2 | 0.692 |
| fold 4 | 0.564 | 0.624 | 0.972 | 0.970 | 1<2 | 0.174 |
| Total | 0.613 | 0.612 | 21.322 | 29.453 | 1<2 | 0.493 |