Guide-yakabatana mibvunzo · Pakati Nhanho

Relation Extraction kubva Text Quiz

Edza kuti kuburitsa kwehukama kunoshandura sei zvinyorwa kuita chokwadi chakarongeka, uye kuti mapaipi ayo akajairwa uye nzira dzekutarisa dzinogona kutadza sei.

Nzira dzekutungamira dzinoenderanaRelation Extraction Kubva Muzvinyorwa
Mubvunzo 1 ye 8

Ndeipi iyo yakajairika kubuda kweiyo relation extraction system?

Tarisa mubvunzo wese8 mibvunzo

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

  1. Ndeipi iyo yakajairika kubuda kweiyo relation extraction system?
    • Yakarongeka katatu inobatanidza masangano maviri nehukama
    • Muti wezvivakamutauro wemutsara
    • Manzwiro akaenzana kubva kune negative kusvika kune positive
    • Kuturikirwa kwechirevo kune mumwe mutauro
  2. In a common pipeline for relation extraction, what may identify entity mentions before pair classification?
    • Chinyorwa-ku-kutaura-manyorerwo
    • Yakapihwa zita rekuziva
    • Image segmentation
    • Audio source separation
  3. Why can distant supervision produce noisy relation-extraction labels?
    • Knowledge bases contain no structured facts
    • It requires every sentence to be hand-labeled
    • A sentence may mention two linked entities without stating their knowledge-base relation
    • It removes entity mentions before training
  4. Mazhinji sei transformer-based RE modhi anoratidza kuti ndeipi mubatanidzwa pairi yekuisa?
    • Nokushandura chirevo chekutanga
    • Nekudzima chimwe chemasangano
    • Nekushandura zvinyorwa kuita audio
    • Nekuisa ma tokeni akakosha kutenderedza masangano maviri
  5. Chii chinosiyanisa 'yakavhurika' hukama hwekubvisa kubva kune yakajairwa RE?
    • Inobvisa mitsara yehukama zvakananga kubva kune zvinyorwa pasina kufanotsanangurwa seti yemhando dzehukama
    • Inoshanda chete panhamba
    • Inoda data rakanyorwa kupfuura chero imwe nzira
    • Inogona chete kuwana hukama humwe pagwaro
  6. Why include a no-relation label when classifying candidate entity pairs?
    • To translate the sentence into another language
    • To represent pairs for which the text does not assert a target relation
    • To replace named entity recognition
    • To guarantee that every pair has a relation
  7. What does document-level relation extraction allow a system to use?
    • Only the spelling of a single entity
    • Audio features from a spoken recording
    • Evidence that may be spread across multiple sentences
    • A fixed relation without reading text
  8. How does open relation extraction differ from a closed relation schema?
    • It cannot use text as input
    • It always requires a knowledge base to label every sentence
    • It only identifies entity boundaries
    • It can express relation phrases without choosing only from a predefined label inventory