Phambi kokuba sithethe into ngeliphepha, kuyafuneka ukuba siliqale ngokucacisa izinto ezimbalwa. Ukuba umfundi akayazi ukuba yintoni i-morphological analyser; sizokuqala siyicacise. I-morphological analyser yi-software esetyenziswayo ekufumaneni, ekuqwalaseleni kunye nasekucaciseni amalungu olwimi; amalungu lawo zizinto ezifana neziqu zamagama, izigaba zentetho kunye nezinye izinto.
Eliphepha lifuna ukuqwalasela ukuba ingaba yintoni efanayo kunye nengafaniyo phakathi kwesiXhosa nesiZulu. Isizathu salento yinto yokuba kukho i-morphological analyser yesiZulu esele yakhiwe - umbuzo obalulekileyo uthi; ingaba engatshintshwa kuyo ukwenzela ikwazi ukusebenza nesiXhosa. Iballekile lento ngoba kuzakuthatha ixesha ukwakha i-morphological analyser yesiXhosa kabutsha. Ababhali baqwalasela izinto ezimbini elwimini:
house & to house, igama eli lithi house lifundwa ngeendlela ezingafaniyo.Ababhali bafuna ukutshintsha izinto ezincinci kule-morphological analyser ukwenzela ukuba ikwazi ukusebenza nesiXhosa. Emva koko bazakuhlola ukusebenza kwale analyser. Umbuzo ekumele ukuba unawo yinto yokuba ababhali bayithatha phi i-corpus?
Besele sitshilo ukuba eliphepha lijonga isiZulu kunye nesiXhosa. Ezilwimi zimbini ziyingxenye yelwimi zesiNguni. Ingaba kutheni ababhali bekhethe isiXhosa? Kutheni bengakhethanga isiSwati okanye isiNdebele. Kwizifundo ze-natural language processing iilwimi zaBantu azifundwa kakhulu - zinqabile. Iyinyani le nto kwisiZulu kunye nesiXhosa. Ulwimi lwaBantu olune-morphological analyser epheleleyo sisiSwahili. I-morphological analyser eyenzwe ngababhali yesiZulu ithiywe ngokuba yi-ZulMorph. Indlela abazokuyenza isebenze nesiXhosa ilelihlbo:
The extensions concern the word root lexicon, followed by the grammatical morpheme lexicons and finally by the appropriate morphophonological rules
Akukacaci ukuba ze-extentions ezithini ezo. Eliphepha lithetha kakhulu nge “open” kunye ne “closed” class. Lithatha ixesha ukude licacise ukuba zintoni ezo. Andiqondi ukuba yinto enyamezelekayo le. Ababhali bathi, ngaphezulu kwaleyo, ukuba i-morphological analyser le bayakhayo izakuba ne “guesser variant”. Abayicacisi ukuba yintoni leyo.
Eyona nto sizakuyiqwalasela apha kweliphepha ngumsebenzi wababhali kwi-morphotatics. Ii-morphophonological alternations sizakhe sizibeke ecaleni. Isizathu yintoyokuba indlela amagama abizwa ngayo ayibalulekanga kuthi. Bathi ababhali zikhona izinto esiXhoseni ekufuneka zilungiselelwe ngoba azikho esiZulwini. Umzekelo wezizinto yi-temporal form. Ababhali bathetha ngathi zikhona ezinye - kodwa abazichazi. I-temporal form kukubonakalisa ukuba isenzo senzeke nini isenzo so. Kongezwa i-morphome esiXhoseni ukubonakalisa i-temporal form.