ACL RD-TEC 1.0 Summarization of C02-1162
Paper Title:
IDENTIFYING CONCEPTS ACROSS LANGUAGES: A FIRST STEP TOWARDS A CORPUS-BASED APPROACH TO AUTOMATIC ONTOLOGY ALIGNMENT
IDENTIFYING CONCEPTS ACROSS LANGUAGES: A FIRST STEP TOWARDS A CORPUS-BASED APPROACH TO AUTOMATIC ONTOLOGY ALIGNMENT
Authors: Grace Ngai and Marine Carpuat and Pascale Fung
Primarily assigned technology terms:
- algorithm
- classification
- corpus-based approach
- cross-lingual identification
- data acquisition
- dictionary induction
- direct translation
- forward match
- grouping
- hownet construction
- identification
- illustration
- induction
- information retrieval
- lexicon induction
- machine translation
- measuring
- multilingual information retrieval
- segmentation
- semantic similarity calculation
- similarity calculation
- tagger
- word segmentation
Other assigned terms:
- american english
- approach
- bilingual corpora
- bilingual corpus
- case
- chinese corpora
- chinese corpus
- chinese words
- chinese-english lexicon
- co-occurrence
- co-occurrence statistics
- collocation
- comparable corpora
- compounds
- concepts
- context vector
- context vectors
- corpora
- correlation
- dictionary
- distribution
- english translations
- english verbs
- fact
- hownet
- idiomatic expressions
- knowledge
- large corpus
- lexical semantics
- lexicon
- machine-readable dictionary
- mandarin chinese
- mandarin chinese hownet
- manual translation
- mapping
- mappings
- measure
- method
- multilingual information
- news corpus
- nouns
- ontologies
- ontology
- parallel corpora
- part-of-speech
- seed
- semantic
- semantic similarity
- sense granularity
- senses of a word
- sentences
- similarity scores
- sparse data
- sparse data problem
- statistics
- structural information
- style
- synonym
- synonyms
- synset
- synsets
- technical terms
- term
- term frequency
- terms
- test set
- theory
- translation research
- translations
- user
- verb
- word
- word co-occurrence
- word distribution
- word senses
- word similarity
- word usage
- word-net
- wordnet
- wordnet synset
- wordnet synsets
- words
