Namhlanje sifunda iphepha lika Reiter no-Dale lijonga ii-systems ze-NLG. Eliphepha, nangona amagama efana, lahlukile kweli lithi “Building Natural Language Generation Systems”. Kukho igama elinye elihlukileyo. Abababhali bazijonga ii-NLG systems ngengabantu abafuna ukufundisa ukuba zakhiwa njani. Bazakuthetha ngokuba zisetyenziswa nini ezinye ii-techniques ze-NLG. Banikezela ngeengcebiso malunga ne-requirements analyses kunye nezinye izinto ezenziwayo kwi-NLG. Bazakugxininisa kwi-techniques ezisele ziphucukile, le nto bayenziswa yinto yokuba ezi-techniques zingasetyenziswa ekwakheni ii-systems ezisebenzayo. Bathi bazakusinonisa ukuba ngawaphi amaphepha esingawafunda xa sidilishana nezinto ezibunzinyarha.

I-natural language generation (NLG) ifumaneka kwizifundo ze-artificial intelligence (AI) kunye neze-computational linguistics. Umsebenzi wesisifundo kukwakha imibhalo yesiXhosa okanye ezinye ilwimi zabantu, ithatha ulwazi olubhalwe ngohlobo olulungele i-computer. Ii-systems ze-NLG zisetyenziswa ekukhupheni iimpepha, iingaciso kunye nezinye intlobo zemibhalo.

Eyona ndlela ibalaseleyo yokusebenzisa ii-systems ze-NLG kukunika abantu ulwazi ngendlela ezakubangela ukubelula ukufunda. Ngaphakathi, indlela ii-computer ezigcina ngalo ulwazi ibangela kubelula kuzo ukudilishana nolwazi, ezi zizinto ezifana nee-airline schedule databases, accounting spreadsheets kunye nezinye. Kumaxesha amaninzi ulwazi olunje lulungele iincutshe zodwa. Umsebenzi wee-NLG system kukuvulela wonke umntu kolulwazi. I-technology ye-NLG iyasetyenziswa kwezinye izinto. Isetyenziswa ekwakheni ii-authoring aids. Ezi zi-systems ezincedisa ekwakheni impepha ezibhalwa rhoqo.

Lo mbhalo uhoye ukwakhiwa kwe-systems ezizokunceda abantu ebomini. Ngenxa yalonto, kubalulekile ukuba sichaze ukuba akunyanzelekanga usebenzise i-techniques ze-NLG kumaxesha onke. Akhona amaxesha apho i-graphs ziingcono kunamagama.

Umbuzo ohamba phambili ngoku ndiyaqonda ukuba uthi ingaba sizakha njani Zizokuqala ngokuqwalasela ii-tasks ze-NLG. Singajonga umsebenzi we-NLG system njengokufuna i-function ethatha imibhalo ethile ikhuphe enye. Kodwa, njenge nto yonke kubalulekile ukuba siyohlule le nto, sijonge indawo ezincinci. Abantu abangqilelanani bonke ngokuba zeziphi ii-subtask ezizona zilungele ii-NLG systems. Zikhona kodwa ezisithoba ii-task ezixhaphakileyo. Kubalulekile kwakhona ukugxininisa ukuba ayithi le nto yonke i-NLG system idinga ii-modules ezisithoba. Ii-systems ezininzi zisebenzisa i-architecture apho i-module enye yenza izinto ezininzi. Ezi-task ziquka:

content determination, discourse planning, sentence aggregation, lexicalization, referring expression generation, kunye ne-linguistic realisation.

Ngoku sizokucacisa ukuba zintoni okanye zenza ntoni ezi-task. I-content determination kulapho sibonayo okanye sizixelela khona ukuba loluphi ulwazi oluzokuba kumbhalo osizakuwukhupha. I-discourse planning kulapho sibonayo ukuba loluphi uhlobo umbhalo esizakuwukhupha uzoma ngawo. I-sentence aggregation yona yi-process apho sithatha imiyalezo siyidibanise kuba sisakha izivakalisi. I-lexicalization yona ithetha ngokukhethwa kwamagama azakucacisa ii-relations. Referring expression generation kulapho kukhethwa amagama alungele ukubonisa i-entities zalento kuthethwa ngayo. Njengamntu omameleyo, kumele ukuba uyazibuza ngoku ukuba kutheni sisohlula i-lexicalization kunye ne-referring expression generation. Kutheni zingadityaniswa mhlawumbi senze into ekuthiwa yi-lexicalisation ne-referring expressionation? Isizathu yinto yokuba i-referring expression generation kufuneka iqwalasele zeziphi i-entities ekusele zisetyenzisiwe kwaye kusetyenziswe amaphi amagama. Le nto yenziwa ngoba yokuba kubalekwa ukusetyenziswa kwegama elinye ezintweni izininzi. I-linguistic realisation kulapho sisebenzisa imithetho ye-grammar sikhuphe imibhalo yelwimi eluthile, kwaye izivakalisi ezo zithethe into eyiyo.