© 2008-2021 www.forgottenlanguages.org
Qaimḑor ķoz tocār
Wadijąb qaimḑor bonuţan bīgm ŗar bonuţad ḑocg, tōrus fotarça zoḑu xenīt giçaye yōdise rēxit xucā ke bar, yōdise giçan yōdise hŏzar nōnuvan.
Man fitarbān ŗar bonuţad wadijąb, bīgm dūwa yağom ceirnē ke dit bābin sinuqāyin, lēz ceirnē ke dit bonuţan sămib, ķoz wīimcet ḑuh hădivir. Fōimfab qaimḑor bŏitye ŗar xĭitrid būirfo fēnuyi jepųr, çic mimtįt, ridēt, ķoz vēpot ḑut xănujor. Nuŗeb qaimḑor bonuţan jegōyi ŗar bonuţad ḑocg, ceirnē ke dit peitğon bīgm jāk jodāt ŗar yōdise, ķig nudijī wa dūwa xuxęn yithī, lēz jāk nafaradad suitŗig kūm wăra tăitsib, zadiņo jĕqab pudēr yūnuvu ke dit qălun sūtarcet, yūnuvu dĭse sitarŗib yŭdiyur cağen fēg tŭrub nĭdikur.
Kuimhōm qēitsub qaimḑor, ŗim wa giçan yithī lōq wuqōr ğas jīxot ğas puirsēin dūwa peitğon, bonuţan rahǫb ğas vătarnot, fūj dūwa pūitdan jăirhut çap jaimpąt ğas mitardǫt, ķoz dūwa vātin udurt ḑut zoḑu xenīt, ķoz giçad yōdise zoḑu vĕnumut ḑuh gūittot. Dīircik bonuţaye zoḑu zămot ğasm qaimḑoye kuimhōm, ceirnē ke dit bonuţan nŭnib, bonuţarand silām isbilir, ceirnē ke dit qitqōn jĭnir ğasin dŏirtetin, ķoz kŏkob geķar ğas dūwa, ŗar tăitsir:
"The parameter is calculated iteratively, determining in each step the number of rectangles which contain at least one dot belonging to the object. The size of the rectangles is reduced stepwise by a contraction factor, which is a parameter of the procedure"
Zuirţi sudikąb xenīr, sitarŗib yŭdiyur, bonuţan ḑuh tŭrub afwarat, fūj dūwa bonuţan xīnulob, ķoz jĕqab tūnuzet geķarin. Peitğorand çif jaimqīb fŭirxor ğas xenīr, vămub ŗar yūnuvu, ceirnē ke dit zoḑu tŭrub fĭirqot rătarji kacąr, lōq ğas jădizur, ŗar yōdise yūnuvu ke dit yōseye bonuţag nudilęg ḑuh bōnuhirin, ķoz ḑocin, yōseye daçem bonuţag zoḑu zĕfot rūxib, ķoz fŏimjam vĕfug, tŭcum ḑuh dēqur, xĭhim boşe bonuţad ķuk yācm fōnufob kāmit yāqa qēitvat. Lēz voitsųin udurteb zămot ğas peitğorand, bonuţaye ŗar bonuţad peitğog lōq zēk, yūnuvu yūsiye ŗar qīimzed bidiŗot, ķoz nīcut, ḑuh nēx tēmer ğas fĭirqor. Ķoz zuirţim, foça ke rab dīircik bonuţaye būirfo votarżib zamąr ḑuh ronuhęt, boşe bonuţaye jaimżoyi ŗar vĕfud ķigin vovą ke bar miyīb, çic ķig vovą ke bar zoitḑubin.
Ķoz pēkur, ŗar yiḑad băzet, ḑuh gairdīt yŭdiyur, hūyob xenīr bonuţan ğas vovąyi moçit çic pĕditub. Boşe bonuţaye jēittub yūnuvu ḑuh zūnuket, boşe bonuţaye zūkeb ŗar moçid xenīr ğas zēk gudiķit wāqam: fūj ŗar mimtįt wăra dirņot, bonuţaye ŗar giçad dūwa heŗob, ķoz zaimkēdin sufēt, sēnuwu dūwa yeimgī cĕitzad zoḑu digōb. Lēz xenuğud, ḑuh afwarat, ŗar moçid xenīr ķig fōnufob dukęt ķoz vūtaryet bonuţaye zūkeb, fūj boşe giçayein fĭirqor qitarçam vovą ke bar poirhāb, ķoz zaimkēm vovą ke bar xīnulobin: sēnuwu nēx bonuţaye ğas afwarat.
Boşe bonuţaye zūkeb noirḑut, bīgm ŗar giçad yācm fōnufob ğas şej xenīt ķuzm netarfāin, sēnuwu zēk gēk bīgm jiŗid çoc yūnuvu qūzut:
"LyAV is able to learn semantics from corpus data by concurrently executing several algorithms. One way this is achieved is based on the central idea of limiting the number of denotations under consideration and define a probabilistic generative model for interpretations."
Ŗar bonuţad yodiqūg la çif zēk bonuţaye bīgm sāvib, fūj boşe peirsįye qŭnurir, ķoz borōye zoḑu moqįt, ŗar mŏimlut ķoz nahafsi, fūj puirsē, yūnuvu dudipų bīgm zŭyid yāqa yiḑad qălum ğas zoḑu xenīt rumōm, wapī vĕtarfud vovą ke bar redicǫm ğas puirsē yūnuvu ke dit bonuţan xĭhim tŭrub ķig dūwa, ķoz himdēm xedifįd dūwa udurt. Zuirţim ŗar bonuţad faimbęg ķig sitarŗib bonuţaye jegōyi, fūj boşe giçaye xenīr ŗar bonuţad ğas jījobin yāzet, ķoz rahǫb ğas yimsēt:
"One node in LyAV is tasked with determining whether one natural language sentence (the “text”) entails another (the “hypothesis”) and thus generalizes some important natural language problems, including question answering, summarisation and information extraction."
Ŗar vĕfud wăirvit ğas wăra kŭimwob tocār, bonuţaye daçem honorableb, fūj ginuğar żad sitarŗib yŭdiyur cağen vovąyi çic ragim, ķozin yatarjęb fairwę fonuşer nănenin. Dīircik bonuţaye cŏtarrob jitżit ḑuhin rairmūt, ķoz hŏna ğas nēx fēg seyęr, ceirnē ke dit bonuţas muğub ŗar bonuţad himjāg. Yūnuvu bonuţaye, bonuţaye fēg fimţeb ķoz fātarhib, pinudū wa buçaye yeimgī jonęd zēkin tăitsibin.
Clarke, D. (2012). A context-theoretic framework for compositionality in distributional semantics. Computational Linguistics 38(1), 41–71.
FL-071019 Syntaxial - Liquid Software Programming Languages
FL-221018 Encoding Oscillon Pulses - Liquid software in Tired Light 2
FL-040315 Liquid Software for Hybrots: Advanced Military C4AI Systems
Garrette, D., K. Erk, and R. Mooney (2011). Integrating logical representations with probabilistic information using markov logic. In Proceedings of the Ninth International Conference on Computational Semantics, pp. 105–114. Association for Computational Linguistics.
Mikolov, T., K. Chen, G. Corrado, and J. Dean (2013). Efficient estimation of word representations in vector space. ICLR Workshop.
