325 KB0 KB194820252025Mečkovskaja, Nina Borisovnaštevilka:3letnik:73str. 435–449ISSN:0350-6894DOI:10.57589/srl.v73i3.4171COBISSID_HOST:264140803URN:URN:NBN:SI:doc-48R8X0U0ruSlavistično društvo SlovenijeSlavistična revijafrekvenčni slovarjifrequency dictionarieshigh-frequency words as semantic multiplierskomponentna pomenska analiza 100.000-besednega slovarjasemantic component analysis of a 100,000-word dictionarysinteza razlagalnih in frekvenčnih slovarjev v slovarjih Macmillan (2007)synthesis of explanatory and frequency dictionaries in the Macmillan dictionary (2007)visokofrekvenčne besede kot semantični multiplikatorjiKorpusnaja lingvistika, častotnaja i tolkovaja leksikografija| vektory vzaimodejstvija| Corpus linguistics, frequency and explanatory dictionaries| interaction vectors|With the overall growth in the number of corpora, their volumes and diversity, there is a specialization of corpora depending on their targeted content. Electronic corpora of the first generation (with a volume of approximately 100-million-word tokens), called or perceived as “national” or “state”, retain a relative balance of subcorpora and a broad social science and humanities audience. As the volumes of later corpora increase, there is a specialization of their purpose along two vectors: 1) content-oriented monitor (replenished) megacorpora of newspaper and magazine texts; the target groups of corpus content of this class include sociologists and political scientists, economists, demographers, journalists, etc. 2) thematically unlimited (non-selective) corpora accumulating digitalized texts (printed and electronic) used in computer science as raw material for “natural language processing” (machine pre-training 436 Slavistična revija, letnik 73/2025, št. 3, julij–september of neural networks) and creation of statistical algorithms for self-linking words into adequate verbal responses of artificial intelligence. Two most significant innovative developments in corpus lexicography are named: 1) synthesis of explanatory and frequency dictionaries in the Macmillan dictionaries (2007) and later Collins, Longman; 2) component semantic analysis of a 100,000-word dictionary using the 2,500 most frequent lexemes in Macmillan (2007) as semantic componentsPri obščem roste količestva korpusov, ih ob"emov i raznoobrazija proishodit specializacija korpusov v zavisimosti ot sostava ih targetirovannogo kontenta. Elektronnye korpusy pervogo pokolenija (ob"emom primerno 100 mln slovoupotreblenij), nazyvaemye ili osoznavaemye kak “nacional'nye” ili “gosudarstvennye”, cohranjajut otnositel'nuju sbalansirovannost' podkorpusov i širokuju social'no-gumanitarnuju adresaciju. Po mere uveličenija ob"emov bolee pozdnih korpusov proishodit ih specializacija po dvum vektoram: 1) soderžatel'no orientirovannye monitornye (popolnjaemye) megakorpusy gazetno-žurnal'nyh tekstov; v celevye gruppy korpusnogo kontenta dannogo klassa vhodjat sociologi i politologi, ekonomisty, demografy, žurnalisty i dr.; 2) tematičeski bezgraničnye (neizbiratel'nye) korpusy, akkumulirujuščie ocifrovannye teksty (pečatnye i elektronnye), ispol'zuemye v informatike kak syr'e dlja “obrabotki estestvnnogo jazyka” (natural language processing): mašinnogo predobučenija nejronnyh setej i sozdanija statističeskih algoritmov samosvjazyvaemosti slov v adekvatnye tekstovye reakcii iskusstvennogo intellekta. Nazvany dve naibolee značitel'nye novatorskie razrabotki v korpusnoî leksikografii: 1) sintez tolkovogo i častotnogo slovareî v slovarjah Macmillan (2007), pozže Collins, Longman; 2) komponentnyj semantičeskij analiz 100-tysjačnogo slovnika s ispol'zovaniem v kačestve semantičeskih komponentov 2.500 samyh častyh leksem v Macmillan 2007. Vozmožnosti korpusov v skorom vremeni privedut k krupnym dostiženijam v diahroničeskoj lingvistikeTEXTznanstveno časopisjejournalsSlovenian National E-content AggregatorNational and University Library of SloveniaSlavistično društvo Slovenije