Natural Language Processing Complexity and Parallelism
نویسندگان
چکیده
This paper reviews the processes involved in Natural Language Processing (NLP). It then demonstrates the various kinds of choices that need be taken during the execution of the word morphology, the syntactic text analysis, or text generation components. It compares the time complexity of traditional serial algorithms and examines the possible expected gain in some corresponding parallel counterparts.
منابع مشابه
Effective Exploitation of Parallelism in NLP
The increasing demand for accuracy and robustness for today’s natural language processing systems brings on an increasing demand for computing power. Experiments on our unification-based parser indicated that only fine-grained parallelism could yield acceptable speedup. We present a parallelization technique to exploit this fine-grained parallelism. Although the approach was designed for the pa...
متن کاملProcessing Underspecified Semantic Representations in the Constraint Language for Lambda Structures
The constraint language for lambda structures (CLLS) is an expressive language of tree descriptions which combines dominance constraints with powerful parallelism and binding constraints. CLLS was introduced as a uniform framework for defining underspecified semantics representations of natural language sentences, covering scope, ellipsis, and anaphora. This article presents saturation-based al...
متن کاملParallelizing the CKY and Earley Parsing Algorithms
Context-free parsing algorithms are one of the oldest and most well-understood aspects of natural language processing. Efforts to reduce the time complexity of these algorithms have produced two particularly popular algorithms: the Cocke-Kasami-Younger (CKY) bottomup parsing algorithm [5, 9], and the Earley top-down parsing algorithm [2, 3]. However, despite these efforts, parsing remains a tim...
متن کاملNatural Language Multiprocessing: A Case Study
This paper presents two case studies of parallelization of large Natural Language Processing (NLP) applications using a parallel logic programming system (called "ACE") that automatically exploits implicit parallelism. The first system considered is Artwork, a system for semantic disambiguation, speech act resolution, and temporal reference resolution. The second system is ULTRA, a multilingual...
متن کاملParallel Logic Programming with Extensions
A programming language is a tool and a vehicle for applications. The need to promote the evolution of Prolog towards a more expressive new logic programming language. A large medium-term national research project. Extensions to Logic Programming: new forms of computational reasoning, with explicit negation, constraint programming, parallelism and distribution support. The resulting language: PR...
متن کامل