ACL RD-TEC 1.0 Summarization of C04-1136
Paper Title:
SIGNIFICANCE TESTS FOR THE EVALUATION OF RANKING METHODS
SIGNIFICANCE TESTS FOR THE EVALUATION OF RANKING METHODS
Primarily assigned technology terms:
- approximation
- binomial test
- classification
- collocation extraction
- comparative evaluation
- cross-validation
- cutoff
- database
- estimator
- evaluation procedure
- identification
- information retrieval
- learning
- machine learning
- machine learning and classification
- maximum-likelihood
- natural-language processing
- parser
- parsers
- partial parser
- pre-processing
- processing
- random selection
- ranking
- ranking method
- scoring
- scoring function
- search
- validation
Other assigned terms:
- annotation
- annotators
- approach
- association measure
- binomial distribution
- case
- classification tasks
- collocation
- conditional distribution
- conditional probabilities
- contingency table
- corpora
- correlation
- distribution
- empirical results
- euclidean space
- evaluation experiment
- evaluation strategy
- fact
- feature
- feature space
- feature vector
- feature vectors
- gold standard
- heuristics
- human annotators
- hypothesis
- implementation
- inferences
- inter-annotator agreement
- interpretation
- lexical association
- log-likelihood
- log-likelihood ratio
- manual annotation
- measure
- measures
- method
- methodology
- multilingual corpus
- n-best list
- natural-language
- noise
- normal distribution
- null hypothesis
- pp-verb
- precision
- priori
- probabilities
- probability
- procedure
- standard deviation
- statistic
- statistical model
- statistical significance
- statistics
- terms
- text
- text corpora
- text type
- toolkit
- words
