This paper proposes a fusion model to reinforce fuzzy association rules, which contains two main procedures: (1) employing the cumulative probability distribution approach (CPDA) to partition the universe of discourse and build membership functions; and (2) using the AprioriTid mining algorithm to extract fuzzy association rules. The proposed model is more objective and reasonable in determinin...

Lexical bundles are empirically derived formulaic units of language which are register-specifi c and perform a variety of discourse functions. Because these units of language contribute to the linguistic make up of specifi c registers, they can be important indicators for determining the success of language users within these discourse communities. Therefore, language learners need to assimilat...

This paper reports a series of three experimental studies (described in detail in Ong 2013) with three related goals/results. The Vrst goal is to empirically evaluate two competing – syntax vs. discourse driven – accounts of strict vs. sloppy readings of reWexives in VP ellipsis, building on the experimental investigation in Kim and Runner (2009). The results strongly suggest that discourse-bas...

Automatic metrics for the evaluation of machine translation (MT) compute scores that characterize globally certain aspects of MT quality such as adequacy and fluency. This paper introduces a reference-based metric that is focused on a particular class of function words, namely discourse connectives, of particular importance for text structuring, and rather challenging for MT. To measure the acc...

This paper investigates the structure of Mandarin spoken dialogues by analysing the distribution of words and turns used in dialogues. The results of an empirical quantitative study show that independent of speakers, there exists a kind of basic vocabulary for daily Mandarin conversations. It is proposed that this is the minimal set of a lexicon for the use of spoken Mandarin. Moreover, a numbe...

“Uptalk” or the use of rising and high pitch at the end of statements is common in interactive discourse in Australian English. The distribution and discourse functions of complex and compound rising tunes were examined in a corpus of Australian English map task dialogues. Each utterance was analysed in terms of Dialogue acts (classified using DAMSL) and intonational tune. It was found that mos...

This study seeks to provide evidence of the dynamics associated with the configurations of discourse-voice regulatory strategies in patient-therapist interactions in relevant episodes within psychotherapeutic sessions. Its central assumption is that discourses manifest themselves differently in terms of their prosodic characteristics according to their regulatory functions in a system of intera...

Critical discourse analysis has focussed extensively on argumentation in anti-immigration discourse where a specific suite of argumentation strategies has been identified as constitutive of the discourse. The successful perlocutionary effects of these arguments are analysed as products of pragmatic processes based on ‘common-sense’ reasoning schemes known as topoi. In this paper, I offer an alt...

Syntactic knowledge is important for discourse relation recognition. Yet only heuristically selected flat paths and 2-level production rules have been used to incorporate such information so far. In this paper we propose using tree kernel based approach to automatically mine the syntactic information from the parse trees for discourse analysis, applying kernel function to the tree structures di...