tailieunhanh - Báo cáo khoa học: "An Optimization Tool for MaltParser"

Data-driven systems for natural language processing have the advantage that they can easily be ported to any language or domain for which appropriate training data can be found. However, many data-driven systems require careful tuning in order to achieve optimal performance, which may require specialized knowledge of the system. We present MaltOptimizer, a tool developed to facilitate optimization of parsers developed using MaltParser, a data-driven dependency parser generator. MaltOptimizer performs an analysis of the training data and guides the user through a three-phase optimization process, but it can also be used to perform completely automatic optimization. Experiments show that. | MaltOptimizer An Optimization Tool for MaltParser Miguel Ballesteros Complutense University of Madrid Spain miballes@ Joakim Nivre Uppsala University Sweden Abstract Data-driven systems for natural language processing have the advantage that they can easily be ported to any language or domain for which appropriate training data can be found. However many data-driven systems require careful tuning in order to achieve optimal performance which may require specialized knowledge of the system. We present MaltOptimizer a tool developed to facilitate optimization of parsers developed using MaltParser a data-driven dependency parser generator. MaltOptimizer performs an analysis of the training data and guides the user through a three-phase optimization process but it can also be used to perform completely automatic optimization. Experiments show that MaltOptimizer can improve parsing accuracy by up to 9 percent absolute labeled attachment score compared to default settings. During the demo session we will run MaltOptimizer on different data sets user-supplied if possible and show how the user can interact with the system and track the improvement in parsing accuracy. 1 Introduction In building NLP applications for new languages and domains we often want to reuse components for tasks like part-of-speech tagging syntactic parsing word sense disambiguation and semantic role labeling. From this perspective components that rely on machine learning have an advantage since they can be quickly adapted to new settings provided that we can find suitable training data. However such components may require careful feature selection and parameter tuning in order to give optimal performance a task that can be difficult for application developers without specialized knowledge of each component. A typical example is MaltParser Nivre et al. 2006 a widely used transition-based dependency parser with state-of-the-art performance for many languages as .