Pîn-lu̍t-pió

Kè-sǹg gí-giân-ha̍k kóng ê pîn-lu̍t-pió (frequency list) tō sī kā jī (word, word type) chiàu pîn-lu̍t (frequency) koân-kē lâi pâi, ah pîn-lu̍t tō khoàⁿ hit ê jī tī gí-liāu-khò͘ (corpus) chhut-hiān kúi pái lâi sǹg.

Eng-gí ê lē

the	3789654
he	2098762
...
king	57897
boy	56975
...
outragious	76
...
stringyfy	5
...
transducionalify	1

Pe̍h-oē-jī ê lē

Ē-kha sī 2005 nî 7 goe̍h POJ gí-liāu-khò͘ thóng-kè--chhut-lâi ê pîn-lu̍t. Pîn-lu̍t siōng koân ê 31 ê jī lóng sī toaⁿ-im-chat.

sūn-sū	POJ	pîn-lu̍t	pí-lē	ha̍p-chóng
1	ê	198203	8.13%	8.13%
2	i	59922	2.46%	10.59%
...
9	bô	28770	1.18%	20.96%
...
20	kaù	13968	0.57%	30.26%
...
32	in-ūi	10634	0.44%	36.36%
...
42	ēng	8010	0.33%	40.17%
...
94	toà	2934	0.12%	50.03%
...
228	sèng	1219	0.05%	60.03%
...
559	kan-taⁿ	466	0.019%	70.01%
...
1000	tīn	255	0.0105%	76.18%
...
1463	piáu-bêng	162	0.0066%	80.00%
...
4864	lat-bí-ah	35	0.0014%	90.00%
...
10000	saⁿ-chiap	11	0.0005%	94.15%
...
20000	haū-iú-hoē	4	0.0002%	96.83%
...

Lō͘-ēng

Kè-sǹg gí-giân-ha̍k chha-put-to ta̍k hong-bīn ê gián-kiù lóng tio̍h-ài ū tiān-chú jī-tián (electronic dictionary), ah beh chò tiān-chú jī-tián tiāⁿ-tio̍h ài seng ū pîn-lu̍t-pió.

Beh pian kaù-châi, mā tio̍h lī-iōng pîn-lu̍t-pió kā ta̍k ê kai-toāⁿ beh o̍h ê jī kap ē-tit ēng ê jī lia̍t--chhut-lâi, ha̍k-si̍p haū-kó chiah ē khah hó.

Beh pian jī-tián, 10000 jī, 20000 jī, 40000 jī, ... toā-sè pún ê jī-tián beh siu sím-mih jī, pîn-lu̍t-pió mā sī siōng hó ê chham-khó. Án-ne pian--chhut-lâi ê jī-tián chiah ē-tit hù-ha̍p bô-siâng sú-iōng-chiá ê su-iàu.

Lí-lūn

Hó chhan-chhiūⁿ ta̍k chióng chū-jiân gí-giân ê pîn-lu̍t-pió lóng hù-ha̍p Zipf tēng-lu̍t (Zipf's law).

Goā-pō͘ liân-kiat

• Chôan-lô Tâi-gú-bûn sû-pîn thóng-kè chu-liāu Archived 2007-12-22 at the Wayback Machine.