The Experts below are selected from a list of 52470 Experts worldwide ranked by ideXlab platform
Mark Lee - One of the best experts on this subject based on the ideXlab platform.
-
acknowledging Discourse Function for sentiment analysis
International Conference on Computational Linguistics, 2014Co-Authors: Phillip Smith, Mark LeeAbstract:In this paper, we observe the effects that Discourse Function attribute to the task of training learned classifiers for sentiment analysis. Experimental results from our study show that training on a corpus of primarily persuasive documents can have a negative effect on the performance of supervised sentiment classification. In addition we demonstrate that through use of the Multinomial Naive Bayes classifier we can minimise the detrimental effects of Discourse Function during sentiment analysis.
-
WASSA@ACL - Cross-Discourse Development of Supervised Sentiment Analysis in the Clinical Domain
2012Co-Authors: Phillip Smith, Mark LeeAbstract:Current approaches to sentiment analysis assume that the sole Discourse Function of sentiment-bearing texts is expressivity. However, the persuasive Discourse Function also utilises expressive language. In this work, we present the results of training supervised classifiers on a new corpus of clinical texts that contain documents with an expressive Discourse Function, and we test the learned models on a subset of the same corpus containing persuasive texts. The results of this indicate that despite the difference in Discourse Function, the learned models perform favourably.
Phillip Smith - One of the best experts on this subject based on the ideXlab platform.
-
acknowledging Discourse Function for sentiment analysis
International Conference on Computational Linguistics, 2014Co-Authors: Phillip Smith, Mark LeeAbstract:In this paper, we observe the effects that Discourse Function attribute to the task of training learned classifiers for sentiment analysis. Experimental results from our study show that training on a corpus of primarily persuasive documents can have a negative effect on the performance of supervised sentiment classification. In addition we demonstrate that through use of the Multinomial Naive Bayes classifier we can minimise the detrimental effects of Discourse Function during sentiment analysis.
-
WASSA@ACL - Cross-Discourse Development of Supervised Sentiment Analysis in the Clinical Domain
2012Co-Authors: Phillip Smith, Mark LeeAbstract:Current approaches to sentiment analysis assume that the sole Discourse Function of sentiment-bearing texts is expressivity. However, the persuasive Discourse Function also utilises expressive language. In this work, we present the results of training supervised classifiers on a new corpus of clinical texts that contain documents with an expressive Discourse Function, and we test the learned models on a subset of the same corpus containing persuasive texts. The results of this indicate that despite the difference in Discourse Function, the learned models perform favourably.
Judy Delin - One of the best experts on this subject based on the ideXlab platform.
-
the Function and interpretation of reverse wh clefts in spoken Discourse
Language and Speech, 1996Co-Authors: Jon Oberlander, Judy DelinAbstract:We examine the characteristics of the headless reverse wh-cleft, with a view to embedding its description into a model of Discourse Function. On the basis of 302 examples taken from the London-Lund Corpus of spoken English and on Collins' (1991) analysis of the same data, we discuss the distinctive features of the construction, as well as those it shares with other kinds of cleft. In common with other clefts, the reverse wh-cleft is presuppositional, and has the property of conveying part of its content aspectually as a STATE. In addition, it has commonalities with the IP it-cleft in its information structure, and the ability to allocate novel CASE ROLES to existing Discourse referents. Its distinctive features include the summative or ending Function previously observed; we further suggest that the construction is more likely to play a metalinguistic or transaction management role than it is to move a Discourse forward. In addition, its referential behavior differs from that of the it-cleft in a manner likely to affect cognitive processing of its content. We present a formal model, combining the general features of cleft constructions with the distinctive features of reverse wh-clefts in spoken Discourse
Jon Oberlander - One of the best experts on this subject based on the ideXlab platform.
-
the Function and interpretation of reverse wh clefts in spoken Discourse
Language and Speech, 1996Co-Authors: Jon Oberlander, Judy DelinAbstract:We examine the characteristics of the headless reverse wh-cleft, with a view to embedding its description into a model of Discourse Function. On the basis of 302 examples taken from the London-Lund Corpus of spoken English and on Collins' (1991) analysis of the same data, we discuss the distinctive features of the construction, as well as those it shares with other kinds of cleft. In common with other clefts, the reverse wh-cleft is presuppositional, and has the property of conveying part of its content aspectually as a STATE. In addition, it has commonalities with the IP it-cleft in its information structure, and the ability to allocate novel CASE ROLES to existing Discourse referents. Its distinctive features include the summative or ending Function previously observed; we further suggest that the construction is more likely to play a metalinguistic or transaction management role than it is to move a Discourse forward. In addition, its referential behavior differs from that of the it-cleft in a manner likely to affect cognitive processing of its content. We present a formal model, combining the general features of cleft constructions with the distinctive features of reverse wh-clefts in spoken Discourse
Mira Ariel - One of the best experts on this subject based on the ideXlab platform.
-
accessibility marking Discourse Functions Discourse profiles and processing cues
Discourse Processes, 2004Co-Authors: Mira ArielAbstract:When accounting for the usage of some linguistic form, one can refer to its Discourse profile, all concomitant features frequently co-occurring with that form in Discourse, or abstract a more general claim about its Discourse Function, referring only to the necessary and sufficient conditions for the proper occurrence of the form. This article discusses Discourse profile and Discourse Function approaches to referring expressions in linguistic and psycholinguistic research. Focusing on English prenominal possessive noun phrases (NPs, e.g., my career), I argue that although their (prototypical) Discourse profile is amenable to clearly defined automatic processing or production cues, and to explaining potential grammaticizations, only their Discourse Functions can account for all the forms actually attested in Discourse. I conclude that both types of accounts are linguistically relevant, and I propose that psycholinguists devise experiments to test the three hypotheses outlined (Discourse profiles, Discourse...
-
the possessive np construction Discourse Function and Discourse profile
Annual Meeting of the Berkeley Linguistics Society, 2002Co-Authors: Mira ArielAbstract:Pragmatists and Discourse grammarians have benefited tremendously from examining naturally occurring data. An examination of all and only the necessary discoursal conditions obtaining when a certain linguistic expression occurs has paved the way for pragmatists to describe the appropriateness conditions placed on linguistic forms. These were identified as the Discourse Functions of those linguistic expressions. For example, based on an empirical study of English itand wh-clefts in Discourse, Prince ( 1978) defined distinct Discourse Functions for it-clefts and wh-clefts. She argued that the Discourse Function of the embedded clause in it-clefts is to mark "known (factual) information" and that of wh-clefts is to mark information "Given (currently available) to the addressee". These Discourse Functions should hold true in (virtually) every case the clefts are used.