[SRILM User List] Interpreting ngram.lm file

ADEYANJU A. sosimi asosimi at unilag.edu.ng
Thu May 15 10:16:03 PDT 2014


Dear all,

Have just generated a LM file but I need your tips in interpreting the results. See some extract of .lm file below. Anticipating a quick response. Kindest regards

\data\
ngram 1=16536
ngram 2=7998
ngram 3=3200

\1-grams:
-0.8297427	</s>
-99	<s>	-0.1990616
-2.382785	a	-0.3606565
-4.994339	aabo
-4.994339	aadọ́ta
-3.596113	aago	-0.1141468
-4.994339	aawọ̀
-4.994339	aayé
-4.994339	aayọ̀
-4.994339	aàrẹ
-4.994339	aàrùn
-4.994339	aàrẹ
-4.994339	aáràbìnrin
-4.83583	aba
-4.994339	abajáde
-4.477304	abala
-4.630644	abara	-0.6027466
-4.994339	abaraméjì
-4.994339	abà
-4.994339	abàmì
\2-grams:
-2.143121	<s> a	-0.0767099
-3.407348	<s> aago	0.003341295
-3.833085	<s> abẹnugan	0.008342609
-4.054651	<s> adé
-3.622831	<s> adébọ̀wálé
-3.833085	<s> adégbìtẹ́
-4.054651	<s> adégétọ̀
-4.054651	<s> adégòkè
-3.407348	<s> adékọ̀yà
-4.054651	<s> adéníyì
-3.774448	<s> adífálà
-3.269046	<s> agbára
-2.968016	<s> agbègbè
-4.054651	<s> agbo
-3.407348	<s> agbára
-3.774448	<s> agbègbè
-3.310438	<s> agbẹ́gilére
-4.054651	<s> agẹmọ
-3.134347	<s> agogo	-0.0162892
-3.833085	<s> ajo
-3.356196	<s> ọ̀pọ̀lọpọ̀
-3.407348	<s> ọ̀rẹ́	-0.1759373
-2.306117	<s> ọ̀rọ̀	0.03070859
-1.854306	a </s>
-2.501609	a bá
-1.104184	a bá	-0.05966994
-2.501609	a dé
-2.501609	a fagbára	-0.07674935
-1.502123	a fi	-0.05625019
-2.501609	a fipá
-1.803154	a fẹ́
-2.069788	a gbọ́	-0.05816576
-2.501609	a jà
-1.643453	a kì	-0.3194264
-2.221406	a kìí
-1.502123	a kò	-0.1097101
-2.280043	a le
-1.957232	a lè
-1.611268	a lè
-1.957232	a lóhùn	0.1669339
-1.854306	a lọ
-1.414973	a máa	-0.05020244
-2.501609	a mọ̀
-2.501609	a mú
-2.501609	a mọ
-2.280043	a mọ̀
-1.206611	ajá ò	-0.2693426
-0.3606157	ajé </s>
-1.335277	ajé èsìn
-1.335277	ajé ìmú	0.1278539
-0.3462729	akápò ìjọba
-0.8581563	akitiyan </s>
-0.3462729	akoni tàbí
-0.4431829	akọ̀wé àgbà
-0.8037053	akẹ́kọ̀ọ́ </s>
-1.589311	akẹ́kọ̀ọ́ ní
-1.589311	akẹ́kọ̀ọ́ ní
-1.589311	akẹ́kọ̀ọ́ sọ
-1.367745	akẹ́kọ̀ọ́ yorùbá	-0.1815158
-1.367745	akẹ́kọ̀ọ́ yóò
-1.589311	akẹ́kọ̀ọ́ àti
-1.367745	akẹ́kọ̀ọ́ èdè	-0.1630108
-0.2213341	aladani kékèké
-0.3462729	alaga ẹgbẹ́
-0.7530956	alága ìjọba
-0.9746618	alága ẹgbẹ́
-0.1574904	alágbára </s>
-0.2213341	alápapọ̀ jáde
-0.4768893	alẹ́ olúwa
-0.353785	alohùn </s>
-0.9746618	alàgbà </s>
-0.9746618	alàgbà bíṣọ́ọ̀bù
-0.9746618	alàgbà líjàdú	0.1278539
-0.8233941	aládàáni kékèké
-0.4785899	aládùúrà </s>
-0.1247067	alágbàá </s>
-0.2216167	alágbádá iná
-0.2213341	alágà ìjọba
-0.4431829	alákọ̀ọ́bẹ̀rẹ̀ </s>
\3-grams:
-1.763766	<s> a fi
-2.02365	<s> a gbọ́
-1.167317	<s> a kì
-1.614281	<s> a kìí
-1.483251	<s> a kò
-1.763766	<s> a lè
-1.483251	<s> a lè
-2.02365	<s> a máa
-2.02365	<s> a mọ̀
-1.614281	<s> a sì
-1.614281	<s> a ti
-1.614281	<s> a à
-1.088136	<s> a ò
-0.9368684	<s> a ó
-2.02365	<s> a ń
-2.02365	<s> a ṣe
-0.5383909	bá a bá
-1.436117	bá a jà
-1.026747	bá a sọ̀rọ̀
-1.436117	bá a wí
-0.4658403	bí a bá
-1.026747	bí a kò
-1.026747	bí a ṣe
-1.026747	bí a ti
-1.436117	bí a ó
-0.3861521	dá a lóhùn
-1.114962	ni a fi
-1.374847	ni a fẹ́
-1.374847	ni a ti
-1.198756	pé a máa
-0.9388713	pé a ò
-1.198756	pé a ń
-1.271307	tí a ti
-0.6378412	tá a bá
-0.7056006	tí a bá
-1.266508	tí a fi
-0.7594231	àwòrán adúláwọ̀ </s>
-1.032424	<s> agogo márùn
-0.4995385	ọ̀rọ̀ ajé </s>
-0.9355144	<s> ajá ò
-0.1610316	orò ajé </s>
-0.7725398	ọrọ̀ ajé </s>
-0.4583931	<s> akápò ìjọba
-0.4583931	ère akoni tàbí
-1.178552	<s> akẹ́kọ̀ọ́ yóò
-0.7773587	àwọn akẹ́kọ̀ọ́ </s>
-1.361483	àwọn akẹ́kọ̀ọ́ sọ
-1.361483	àwọn akẹ́kọ̀ọ́ yorùbá
-1.361483	àwọn akẹ́kọ̀ọ́ èdè
-0.8563331	<s> alága ìjọba
-0.4995385	ewì alohùn </s>
-0.3334544	lábẹ̀ alàgbà bíṣọ́ọ̀bù
-0.6378412	eégún aláré </s>
-0.2264513	kábíyèsí aláṣẹ èkejì
-0.2251149	àbàtà aláwẹ́ mẹ́rin
-0.3334544	òyìnbó aláwọ̀ funfun
-0.2954186	kábíyèsí aláyélúwà ọba
-0.2190236	okóòwò aláàdani kéréje
-0.3334544	okoòwò aláàdáni kéréje
-0.3334544	gbá aláṣọ </s>
-0.3645163	gan an </s>



More information about the SRILM-User mailing list