Papers by Esther Seyffarth
Corpus-based Identification of Verbs Participating in Verb Alternations Using Classification and Manual Annotation (2020.coling-main)
Copied to clipboard
| Challenge: | Verb alternations allow verbs to appear in a set of syntactically different constructions whose associated semantic frames are systematically related. |
| Approach: | They use ENCOW and VerbNet data to train classifiers to predict the instrument subject alternation and the causative-inchoative alternation . they use count-based and vector-based features as well as perplexity-based language model features to reflect each alternation’s felicity by simulating it. |
| Outcome: | The proposed approach reduces the required annotation effort by only presenting annotators with the highest-scoring candidates from the previous classification. |
Verb Alternations and Their Impact on Frame Induction (N18-4)
Copied to clipboard
| Challenge: | Frame induction is the automatic creation of frame-semantic resources similar to FrameNet or PropBank, which map lexical units of a language to frame representations of each lexical unit’s semantics. |
| Approach: | They propose to use frames to map lexical units to frame representations of each lexical unit's semantics. |
| Outcome: | The proposed framework compares the semantics of alternating verbs and their similarities and differences. |
AET: Web-based Adjective Exploration Tool for German (L18-1)
Copied to clipboard
| Challenge: | AET enables research on the modificational behavior of German adjectives and adverbs . currently available online corpus query tools for German do not lend themselves specifically to research on adjectives - e.g., syntactic relationships or morphological properties. |
| Approach: | They propose a web-based corpus query tool that can be used to query German corpus . they extracted modifiers and modifiees from a print media corpus and stored them in a database . |
| Outcome: | The proposed tool can be transferred to other languages and modification phenomena. |
The Maaloula Aramaic Speech Corpus (MASC): From Printed Material to a Lemmatized and Time-Aligned Corpus (2022.lrec-1)
Copied to clipboard
| Challenge: | The corpus contains 64,845 words, including lemmas, tokens, types, lemmas, sentences, narratives, and speakers. |
| Approach: | They present the first electronic speech corpus of Maaloula Aramaic . it is a Western Neo-Aramaic variety spoken in three Syrian villages . |
| Outcome: | The corpus contains transcriptions, lemmatized transcriptions and audio files . it is available in four formats: transcriptions with audio and phonetic transcriptions . |