SPoT: A Trainable Sentence Planner

SPoT: A Trainable Sentence Planner Marilyn A. Walker AT&T Labs – Research Florham Park, NJ, USA walker@research.att.com Owen Rambow AT&T Labs – Research Florham Park, NJ, USA rambow@research.att.com Monica Rogati Carnegie Mellon University Pittsburgh, PA, USA mrogati+@cs.cmu.edu Abstract Sentence planning is a set of inter-related but distinct tasks, one of which is sentence scoping, i.e. the choice of syntactic structure for elementary speech acts and the decision of how to combine them into one or more sentences. In this paper, we present SPoT, a sentence planner, and a new methodology for automatically train- ing SPoT on the basis of feedback provided by human judges. We reconceptualize the task into two distinct phases. First, a very simple, randomized sentence-plan- generator (SPG) generates a potentially large list of pos- sible sentence plans for a given text-plan input. Second, the sentence-plan-ranker (SPR) ranks the list of output sentence plans, and then selects the top-ranked plan. The SPR uses ranking rules automatically learned from train- ing data. We show that the trained SPR learns to select a sentence plan whose rating on average is only 5% worse than the top human-ranked sentence plan. 1 Introduction Sentence planning is a set of inter-related but distinct tasks, one of which is sentence scoping, i.e. the choice of syntactic structure for elementary speech acts and the decision of how to combine them into sentences.1 For example, consider the required capabilities of a sentence planner for a mixed-initiative spoken dialog system for travel planning: (D1) System1: Welcome.... What airport would you like to fly out of? User2: I need to go to Dallas. System3: Flying to Dallas. What departure airport was that? User4: from Newark on September the 1st. System5: What time would you like to travel on Septem- ber the 1st to Dallas from Newark? Utterance System1 requests information about the caller’s departure airport, but in User2, the caller takes the initiative to provide information about her destina- tion. In System3, the system’s goal is to implicitly con- firm the destination (because of the possibility of error 1We would like to thank Michael Collins and Rob Schapire for their help, comments, and encouragement, and Noemie Elhadad and three anonymous reviewers for very useful feedback. This work was partially funded by DARPA under contract MDA972-99-3-0003. in the speech recognition component), and request in- formation (for the second time) of the caller’s departure airport. In User4, the caller provides this information but also provides the month and day of travel. Given the system’s dialog strategy, the communicative goals for its next turn are to implicitly confirm all the information that the user has provided so far, i.e. the departure and desti- nation cities and the month and day information, as well as to request information about the time of travel. The system’s representation of its communicative goals for utterance System5 is in Figure 1. The job of the sentence planner is to decide among the large number of potential realizations of these communicative goals. Some exam- ple alternative realizations …

Content truncated to 500 words