Phrase

About: Phrase is a research topic. Over the lifetime, 12580 publications have been published within this topic receiving 317823 citations. The topic is also known as: syntagma & phrases.

Computer system for generating architectural specifications and project control instructions

Abstract: A computer system is disclosed for preparing a complete set of architectural specifications for constructing a building, including, if desired, sets of instructions that will be used by personnel during preparation of construction drawings and during construction. A specifier uses brief symbolic characters representative of individual phrases from a large information repository called a Master Phrase Catalog, to mark decisions and phrase selections in code on an order form designed to facilitate the specifier's task. The information in the Master Phrase Catalog includes construction materials, techniques, reference information, testing and industry standards which are correlated with instructional information and work assignments for various groups of people. The specifier uses only a limited number of commands to completely specify a project. The code is fed into a computer which has all the information in the Master Phrase Catalog stored on permanent record file, called the specification data file, with an index file correlating the specifier's code with the phrases that will eventually make up the specification. The computer is programmed to edit the input code for errors, prepare a listing of the input specifier's code and prepare an architectural specification from the specification data file together with a table of contents and listing of errors, if any. If called for by the specifier, sets of instructions, correlated to the particular phrases selected from the Master Phrase Catalog and stored in permanent memory will be printed out as separate booklets.

Jointly optimizing word representations for lexical and sentential tasks with the C-PHRASE model

Abstract: We introduce C-PHRASE, a distributional semantic model that learns word representations by optimizing context prediction for phrases at all levels in a syntactic tree, from single words to full sentences. C-PHRASE outperforms the state-of-theart C-BOW model on a variety of lexical tasks. Moreover, since C-PHRASE word vectors are induced through a compositional learning objective (modeling the contexts of words combined into phrases), when they are summed, they produce sentence representations that rival those generated by ad-hoc compositional models.

Syntactically Guided Neural Machine Translation

Abstract: We investigate the use of hierarchical phrase-based SMT lattices in end-to-end neural machine translation (NMT). Weight pushing transforms the Hiero scores for complete translation hypotheses, with the full translation grammar score and full n-gram language model score, into posteriors compatible with NMT predictive probabilities. With a slightly modified NMT beam-search decoder we find gains over both Hiero and NMT decoding alone, with practical advantages in extending NMT to very large input and output vocabularies.

The acquisition of German relative clauses: a case study.

Abstract: This paper investigates the development of relative clauses in the speech of one German-speaking child aged 2;0 to 5;0. The earliest relative clauses we found in the data occur in topicalization constructions that are only a little different from simple sentences: they contain a single proposition, express the actor prior to other participants, assert new information and often occur with main-clause word order. In the course of the development, more complex relative constructions emerge, in which the relative clause is embedded in a fully-fledged main clause. We argue that German relative clauses develop in an incremental fashion from simple non-embedded sentences that gradually evolve into complex sentence constructions.

Method and apparatus for recognizing handwritten chinese characters

Abstract: A method for recognizing handwritten Chinese characters based on stroke recognition comprises steps of: recognizing handwritten strokes, updating stroke code sequences; retrieving in dictionaries/lexicons at least one corresponding character/phrase entry so as to obtain at least one candidate Chinese character/phrase; dynamically displaying the at least one candidate Chinese character/phrase; jumping to the step of recognizing strokes if it is judged that a next stroke is being written; inputting a displayed Chinese character/phrase into computers as the result of recognition if this character/phase is selected by the user.