Search the T-BDLD Corpus

Free corpus Question

Free corpus query, the word written is a corpus reading style, which shows the gravitational shape or cluster at the center and separates the context around it.

Left context The wanted word is the previous five words, right context The wanted word shows the next six words.

Merkez, the actual surface format caught in the sentence; even if the search expands through the Lemma, the sentence appears in the central sentence.

The search scans the first type written; if necessary, the ClickHouse surface-lemma match will also test lemma and multiword expression candidates.

Last observed, sorgunun tamamıyla eşleşen en yeni cümlenin T-BDLD'ye işlendiği zamanı Türkiye saatine göre gösterir.

Co-use

Co-occurrenceshows another word that is often seen near the wanted word.

Co-use items are kept on ClickHouse on this page 2-4 word ngram It is removed from its plates and is ranked by the number of utilizations.

Density The bar shows relative weight based on the highest number of use in the list.

Serbestlik derecesi, shows whether the wanted word is balanced by a large number of disposable elements or a few patterns.

Area Spreading

Domain dispersion, shows in what areas the wanted word leaves a sentence mark in the T-BDLD field/structure analysis.

Most common area By the number of sentences, dominant area It's calculated by the field score.

Spreading scoreIt sums up with Shannon's diversity whether the word piles up in one area or is it balanced in many areas.

Statistical Scores

NPMImeasures how many words appear by random expectation between negative 1 and +1; higher value means more specific relationship.

logDice, the common use gives power less affected by frequency differences; the higher value points to more established co-use.

t-score, shows how digitally strong it is to be seen together; frequent and stable co-uses are high.

Serbestlik, Shannon considers the share of the most common item with diversity; high value free use tends to be low value patterning.

Okuma Notu

The high number of common utilizations suggests high NPMI is specific, so the number of general words is higher, and the number may be lower.

The most powerful interpretation is that the number of utilizations, the NPMI, logice, and t-score are read together.

Valency

Valency describes which complements a lexical unit selects and the case markers, postpositions, or clauses through which it organises them.

Single words, phrases, and compounds are examined with the same method. Words inside a multiword unit are not counted as complements; only external relations are measured.

Core, strong, optional, and rare labels are corpus tendencies computed from occurrence share, source diversity, proximity, direction, the Wilson lower confidence bound, and frame stability.

Results do not express absolute grammatical obligation; they are re-evaluated through the live ClickHouse signature whenever a new sentence is written to the corpus.