Glossary

Search results

Automatic Caption Evaluation(also: ACE, ACE Framework, ACE Metric)

A caption-quality evaluation framework introduced by Sushant Kafle and Matt Huenerfauth (2017-2018) that scores automatically generated captions based on their usability for Deaf and Hard-of-Hearing readers, rather than simply counting transcription errors. For each mismatch…

Caption Quality(also: Subtitle Quality)

The overall fitness of a set of captions or subtitles for their intended accessibility purpose. Quality is multi-dimensional: it includes text accuracy (whether spoken words are correctly transcribed, commonly measured by Word Error Rate or the NER model), synchronicity with the…

Discriminative Ability(also: Discriminative ability of a metric, Discriminability)

In accessibility research methodology, the property of an evaluation metric to reveal statistically significant differences between stimuli that are known to differ along the dimension being measured. For example, a comprehension-question metric has discriminative ability for…

Failure Rate(also: FR (accessibility metric))

An accessibility metric introduced by Sullivan and Matson (2000) that, for a given page and a given checkpoint, divides the number of checkpoint violations found by the maximum number of violations that could have occurred on that page. Failure Rate produces a normalised value…

Flesch-Kincaid Grade Level(also: Flesch-Kincaid, FKGL, Flesch-Kincaid readability)

A readability formula that estimates the U.S. school grade level required to comfortably read a given English text, based on average sentence length and average syllables per word. Developed for the U.S. Navy in 1975 by Rudolf Flesch and J. Peter Kincaid, the formula is widely…

Fluency(also: Text fluency, Grammatical fluency)

In natural language processing and text simplification, fluency is the degree to which a piece of text is grammatically correct and reads naturally in the target language. It is one of three standard evaluation dimensions for automatic text simplification alongside complexity…

Grammaticality(also: Grammatical correctness, Grammatical acceptability)

The degree to which a sentence conforms to the grammatical rules of a language. In accessibility and NLP research, grammaticality is typically assessed via a 5-point Likert subjective judgement (e.g., "This sentence is grammatically correct") and is used as a component of…

Listenability(also: Auditory Readability, Speech-Output Quality)

A web-accessibility usability metric that measures how appropriate a page's rendered text is when read aloud by a screen reader or voice browser — complementary to, and distinct from, raw WCAG conformance. Listenability penalises meaningless or placeholder ALT text (such as…

Literacy Bias(also: Literacy bias of a metric)

In accessibility research methodology, a literacy bias describes the phenomenon where an evaluation metric systematically produces different scores for participants with different reading-literacy levels, independent of the characteristic being measured. For example,…

NER Model(also: Number, Edition, Recognition Model, NER Accuracy Model)

A caption-quality evaluation model developed by Pablo Romero-Fresco and Juan Martínez Pérez for measuring the accuracy of live subtitling and respeaking. Unlike Word Error Rate, which penalises all errors equally, the NER model weights each error by how much it affects the…

Quality of Perception(also: QoP)

An evaluation framework from the multimedia-accessibility research literature for measuring how well a user can understand and use a media presentation, combining objective comprehension metrics (e.g., fact-recall or multiple-choice quiz accuracy) with subjective judgements…

SARI(also: System output Against References and against the Input sentence)

An automatic evaluation metric for text simplification systems that compares a system’s output against both the original input sentence and a set of human-written simplification references, rewarding the system for adding appropriate words, keeping important words, and deleting…

Score Prediction(also: Predicted grade, Comprehension self-prediction)

A subjective-response evaluation item in which a research participant, immediately after reading a passage and before learning their actual comprehension-question score, estimates the percentage of questions they will have answered correctly. Score prediction is used in…

Sense of Control(also: Perceived control, Locus of control (task-level))

A psychological construct describing a user's subjective feeling of agency over what a system does, distinct from objective control measures such as available options or task-completion rates. In accessibility research on AI-assisted tools, sense of control has emerged as a…

Text Complexity(also: Linguistic complexity, Text difficulty)

The degree to which a piece of writing demands advanced reading skills to comprehend, driven by factors such as vocabulary frequency, syntactic structure, sentence length, passage organisation, and background-knowledge assumptions. In Automatic Text Simplification and…

15 results.