tailieunhanh - Báo cáo khoa học: "Disentangling Chat with Local Coherence Models"

internet h tD we ontinue to do etter on onE str ined disent nglement t skD though so f rD we re un le to pply these improvements to the full t skF e suspe t th tD with etter lowElevel nnot tion tools for the h t dom in nd good w y of integr tE ing prior inform tionD our improvements on SWBD ould tr nsfer fully to s g h tF. | Disentangling Chat with Local Coherence Models Micha Elsner School of Informatics University of Edinburgh Eugene Chamiak Department of Computer Science Brown University Providence RI 02912 ec@ Abstract We evaluate several popular models of local discourse coherence for domain and task generality by applying them to chat disentanglement. Using experiments on synthetic multiparty conversations we show that most models transfer well from text to dialogue. Coherence models improve results overall when good parses and topic models are available and on a constrained task for real chat data. 1 Introduction One property of a well-written document is coherence the way each sentence fits into its context- sentences should be interpretable in light of what has come before and in turn make it possible to interpret what comes after. Models of coherence have primarily been used for text-based generation tasks ordering units of text for multidocument summarization or inserting new text into an existing article. In general the corpora used consist of informative writing and the tasks used for evaluation consider different ways of reordering the same set of textual units. But the theoretical concept of coherence goes beyond both this domain and this task setting- and so should coherence models. This paper evaluates a variety of local coherence models on the task of chat disentanglement or threading separating a transcript of a multiparty interaction into independent conversations1. Such simultaneous conversations occur in internet chat a public implementation is available via https melsner browncoherence. 1179 rooms and on shared voice channels such as push-to-talk radio. In these situations a single correctly disentangled conversational thread will be coherent since the speakers involved understand the normal rules of discourse but the transcript as a whole will not be. Thus a good model of coherence should be able to disentangle .