Finally, the shallow linguistic kernel (SLK) uses only shallow parsing information such as word occurrences and part-of-speech tags (Giuliano et al., 2006).
(ii) KERNEL relies on a supervised classifier [shallow linguistic kernel (Giuliano et al., 2006), identical to French et al. (2012)] that requires only shallow parsing information such as word occurrences and part-of-speech tags.
(People who have a faulty copy of the gene — an extremely rare occurrence — are capable of speech, but what they say is, to strangers, mostly incomprehensible).
Each prototype word is assigned a prior likelihood to be observed thus capturing the uneven frequency of occurrence of different words in natural speech.
Single words served as tools for different speech acts (depending on the context) in a fully balanced design, which also controlled the probability of occurrence of the different speech act types across trials, thus closely matching speech act comprehension in natural setting.
Given the set of words in CELEX that match the current speech sequence, and the frequency of occurrence of these words, we can compute the conditional probability of each of the possible segments that could follow the current segment.
Frequencies of occurrence of language-specific Part-Of-Speech (POS) patterns involving emotion words provide further information with respect to the properties of culturally bound emotion concepts and their linguistic expression.
Given a set of relevant words, either explicitly as a list, or by capturing all occurrences of a given part-of-speech, we compute a document-word matrix.
The distractor candidates are collected from several sources, including the synonyms of the co-occurrence words with the same part-of-speech as the target word in the reading passage, and the sibling and hyponym words of the target word in a lexical taxonomy.
Since lack of punctuation and occurrence of disfluencies in spontaneous speech transcripts are factors that disturb their processing by natural language processing systems, parsers or information extraction systems, automatic analysis of pauses can help to make spontaneous speech transcripts more readable for both human and NLP systems.
