This paper summarizes work over the past ve years on the automated planning and generation of multisentence texts using discourse structure relations, placing it in context of ongoing eeorts by Computational Linguists and Linguists to understand the structure of discourse. Based on a series of studies by the author and others, the paper describes how the orientation of generation toward communi...

We experiment with different ways of training LSTM networks to predict RST discourse trees. The main challenge for RST discourse parsing is the limited amounts of training data. We combat this by regularizing our models using task supervision from related tasks as well as alternative views on discourse structures. We show that a simple LSTM sequential discourse parser takes advantage of this mu...

This paper presents a novel latent variable recurrent neural network architecture for jointly modeling sequences of words and (possibly latent) discourse relations between adjacent sentences. A recurrent neural network generates individual words, thus reaping the benefits of discriminatively-trained vector representations. The discourse relations are represented with a latent variable, which ca...

Scheduling a sequence of tasks — in the acceptation of finding the execution times — is not a trivial problem when the optimization criterion is irregular as for instance in earliness-tardiness problems. This paper presents an efficient Dynamic Programming algorithm to solve the problem with general cost functions depending on the end time of the tasks, idle time costs and variable durations al...

This paper applies machine learning techniques to acquiring aspects of the meaning of discourse markers. Three subtasks of acquiring the meaning of a discourse marker are considered: learning its polarity, veridicality, and type (i.e. causal, temporal or additive). Accuracy of over 90% is achieved for all three tasks, well above the baselines.

The present paper describes procedures to visualise diachronic language changes in academic discourse to support analysis. These changes are reflected in the distribution of different lexico-grammatical features according to register. Findings about register differences are relevant for both linguistic applications (e.g., discourse analysis and translation studies) and NLP tasks (notably automa...

This paper proposes a research methodology for attacking the problem of providing fluent and natural discourse about space and spatially situated tasks between naïve users and robots. We suggest flexible and adaptive ontology mediation, parameterized according to empirically determined discourse and contextual factors, as a suitable architecture with clear applications for the treatment of natu...

We examined the relations between verbal communication disturbances and several hypothesized etiological factors in 47 schizophrenia spectrum individuals. Both alogia and disturbed discourse coherence were associated with poor planning abilities. Alogia and discourse coherence were differentially associated with performance on tasks measuring fluency, working memory, word finding abilities, and...

the present research aimed at investigating the interaction of the classroom talk to have a better understanding of the teachers’ practice teaching which can pave the way for reflection on teaching. the study benefitted from conversation analytic perspective to qualitatively and interpretively examine the classroom discourse of the teacher-learner interaction. the classroom talk was analyzed th...

veryfew attempts have been made in the past to develop instruments to measure pragmatic knowledge of second language (l2) learners. the absence of such instruments in the literature of english language teaching (elt) underscores the need for the researchers to develop new tests that are specifically designed to assess this crucial but less explored aspect of language learners’ (lls) knowledge. ...