In this paper, we consider the problem of sentiment classification of English Twitter messages using machine learning techniques. We systematically evaluate the use of different feature types on the performance of two text classification methods: Naive Bayes (NB) and Support Vector Machines (SVM). Our goal is threefold: (1) to investigate whether or not partof-speech (POS) features are useful f...

This study investigated the integration of word- and phrase-level prominences in speech produced by twenty-five school-aged children (6;2 to 7;3) and twenty-five adults. Participants produced disyllabic number words in a straight count condition and in two phrasal conditions, namely, a stress clash and non-clash phrasal context. Duration and amplitude measures of syllable rhymes were used to as...

Pancheva (2009) argues that phrasal comparatives in Polish exhibit a subject-island effect. She proposes an account of the island effect as a combination of several factors: than has a small clause complement in phrasal comparatives; wh-movement turns the than-clause into a degree predicate; wh-movement of the vP subject is prohibited by an anti-locality constraint; sub-extraction of the vP sub...

Text reuse is a common way to transfer historical texts. It refers to the repetition of text in a new context and ranges from nearverbatim (literal) and para-phrasal reuse to completely non-literal reuse (e.g., allusions or translations). To improve the detection of reuse in historical texts, we need to better understand its characteristics. In this work, we investigate the relationship between...

-

Mandarin Chinese has been claimed to have phrasal stress which falls on a nonhead constituent: on the modifier in a modifier-noun phrase, and on the object in a verb-object phrase (MODNh and VhOBJ, respectively; the subscript h stands for head, and the stressed constituent is underlined). This NONHEAD STRESS RULE is motivated by greater information load carried by the nonhead than its syntactic...