Beyond Similarity - Dimensions of Semantics and How to Detect them: Unterschied zwischen den Versionen

Aus SDQ-Institutsseminar
(Die Seite wurde neu angelegt: „{{Vortrag |vortragender=Felix Pieper |email=felix.pieper@thingsthinking.net |vortragstyp=Masterarbeit |betreuer=Sophie Corallo |termin=Institutsseminar/2022-01…“)
 
Keine Bearbeitungszusammenfassung
 
(Eine dazwischenliegende Version desselben Benutzers wird nicht angezeigt)
Zeile 4: Zeile 4:
|vortragstyp=Masterarbeit
|vortragstyp=Masterarbeit
|betreuer=Sophie Corallo
|betreuer=Sophie Corallo
|termin=Institutsseminar/2022-01-13
|termin=Institutsseminar/2023-01-13
|vortragsmodus=in Präsenz
|vortragsmodus=in Präsenz
|kurzfassung=Semantic similarity estimation is a widely used and well-researched area. Current state-of-the-art approaches estimate text similarity with large language models. However, semantic similarity estimation often ignores fine-grain differences between semantic similar sentences. This thesis proposes the concept of semantic dimensions to represent fine-grain differences between two sentences. A workshop with domain experts identified ten semantic dimensions. From the workshop insights, a model for semantic dimensions was created. Afterward, 60 participants decided via a survey which semantic dimensions are useful to users. Detectors for the five most useful semantic dimensions were implemented in an extendable framework. To evaluate the semantic dimensions detectors, a dataset of 200 sentence pairs was created. The detectors reached an average F1 score of 0.815.
}}
}}

Aktuelle Version vom 2. Januar 2023, 10:12 Uhr

Vortragende(r) Felix Pieper
Vortragstyp Masterarbeit
Betreuer(in) Sophie Corallo
Termin Fr 13. Januar 2023
Vortragsmodus in Präsenz
Kurzfassung Semantic similarity estimation is a widely used and well-researched area. Current state-of-the-art approaches estimate text similarity with large language models. However, semantic similarity estimation often ignores fine-grain differences between semantic similar sentences. This thesis proposes the concept of semantic dimensions to represent fine-grain differences between two sentences. A workshop with domain experts identified ten semantic dimensions. From the workshop insights, a model for semantic dimensions was created. Afterward, 60 participants decided via a survey which semantic dimensions are useful to users. Detectors for the five most useful semantic dimensions were implemented in an extendable framework. To evaluate the semantic dimensions detectors, a dataset of 200 sentence pairs was created. The detectors reached an average F1 score of 0.815.