FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W1185756938> ?p ?o ?g. }

• W1185756938 abstract "A large number of problems in natural language processing (NLP) involve outputs with complex structure. Conceptually in such problems, the task is to assign values to multiple variables which represent the outputs of several interdependent components. A natural approach to this task is to formulate it as a two-stage process. In the first stage, the variables are assigned initial values using machine learning based programs. In the second, an inference procedure uses the outcomes of the first stage classifiers along with domain specific constraints in order to infer a globally consistent final prediction. This dissertation introduces a framework, inference with classifiers, to study such problems. The framework is applied to two important and fundamental NLP problems that involve complex structured outputs, shallow parsing and semantic role labeling. In shallow parsing, the goal is to identify syntactic phrases in sentences, which has been found useful in a variety of large-scale NLP applications. Semantic role labeling is the task of identifying predicate-argument structure in sentences, a crucial step toward a deeper understanding of natural language. In both tasks, we develop state-of-the-art systems which have been used in practice. In this framework, we have shown the significance of incorporating constraints into the inference stage as a way to correct and improve the decisions of the stand alone classifiers. Although it is clear that incorporating constraints into inference necessarily improves global coherency, there is no guarantee of the improvement in the performance measured in terms of the accuracy of the local predictions---the metric that is of interest for most applications. We develop a better theoretic understanding of this issue. Under a reasonable assumption, we prove a sufficient condition to guarantee that using constraints cannot degrade the performance with respect to Hamming loss. In addition, we provide an experimental study suggesting that constraints can improve performance even when the sufficient conditions are not fully satisfied." @default.
• W1185756938 created "2016-06-24" @default.
• W1185756938 creator A5023802054 @default.
• W1185756938 creator A5058998686 @default.
• W1185756938 date "2005-01-01" @default.
• W1185756938 modified "2023-09-23" @default.
• W1185756938 title "Inference with classifiers: a study of structured output problems in natural language processing" @default.
• W1185756938 cites W12836875 @default.
• W1185756938 cites W1482728761 @default.
• W1185756938 cites W1501005381 @default.
• W1185756938 cites W1510793943 @default.
• W1185756938 cites W1525460779 @default.
• W1185756938 cites W1530251607 @default.
• W1185756938 cites W1531859941 @default.
• W1185756938 cites W1535936886 @default.
• W1185756938 cites W1542537254 @default.
• W1185756938 cites W1552508833 @default.
• W1185756938 cites W1554663460 @default.
• W1185756938 cites W1560621288 @default.
• W1185756938 cites W1566346388 @default.
• W1185756938 cites W1597379537 @default.
• W1185756938 cites W1602846073 @default.
• W1185756938 cites W1623072288 @default.
• W1185756938 cites W1629870923 @default.
• W1185756938 cites W1632114991 @default.
• W1185756938 cites W1651266332 @default.
• W1185756938 cites W1676820704 @default.
• W1185756938 cites W1734803745 @default.
• W1185756938 cites W1773803948 @default.
• W1185756938 cites W1774901127 @default.
• W1185756938 cites W1813610639 @default.
• W1185756938 cites W1934019294 @default.
• W1185756938 cites W1963547452 @default.
• W1185756938 cites W1970958875 @default.
• W1185756938 cites W1979711143 @default.
• W1185756938 cites W1980501707 @default.
• W1185756938 cites W1985367963 @default.
• W1185756938 cites W1991133427 @default.
• W1185756938 cites W1991566372 @default.
• W1185756938 cites W1996430422 @default.
• W1185756938 cites W2008652694 @default.
• W1185756938 cites W2008830554 @default.
• W1185756938 cites W2009955311 @default.
• W1185756938 cites W2034916234 @default.
• W1185756938 cites W2036265926 @default.
• W1185756938 cites W204260652 @default.
• W1185756938 cites W2049633694 @default.
• W1185756938 cites W2059933135 @default.
• W1185756938 cites W2061130579 @default.
• W1185756938 cites W2069634543 @default.
• W1185756938 cites W2086699924 @default.
• W1185756938 cites W2092654472 @default.
• W1185756938 cites W2097125878 @default.
• W1185756938 cites W2098921539 @default.
• W1185756938 cites W2099942883 @default.
• W1185756938 cites W2105644991 @default.
• W1185756938 cites W2116632765 @default.
• W1185756938 cites W2117400858 @default.
• W1185756938 cites W2118925939 @default.
• W1185756938 cites W2118996379 @default.
• W1185756938 cites W2119821739 @default.
• W1185756938 cites W2125838338 @default.
• W1185756938 cites W2126851059 @default.
• W1185756938 cites W2129113961 @default.
• W1185756938 cites W2134681090 @default.
• W1185756938 cites W2135843243 @default.
• W1185756938 cites W2137644731 @default.
• W1185756938 cites W2138689857 @default.
• W1185756938 cites W2140785063 @default.
• W1185756938 cites W2141732516 @default.
• W1185756938 cites W2144578941 @default.
• W1185756938 cites W2147880316 @default.
• W1185756938 cites W2148638861 @default.
• W1185756938 cites W2155693943 @default.
• W1185756938 cites W2156373089 @default.
• W1185756938 cites W2156515921 @default.
• W1185756938 cites W2157800108 @default.
• W1185756938 cites W2159080219 @default.
• W1185756938 cites W2160225842 @default.
• W1185756938 cites W2160482687 @default.
• W1185756938 cites W2161290181 @default.
• W1185756938 cites W2296427152 @default.
• W1185756938 cites W2337098149 @default.
• W1185756938 cites W2911823959 @default.
• W1185756938 cites W2915429162 @default.
• W1185756938 cites W2915816387 @default.
• W1185756938 cites W2153936946 @default.
• W1185756938 hasPublicationYear "2005" @default.
• W1185756938 type Work @default.
• W1185756938 sameAs 1185756938 @default.
• W1185756938 citedByCount "0" @default.
• W1185756938 crossrefType "journal-article" @default.
• W1185756938 hasAuthorship W1185756938A5023802054 @default.
• W1185756938 hasAuthorship W1185756938A5058998686 @default.
• W1185756938 hasConcept C119857082 @default.
• W1185756938 hasConcept C154945302 @default.
• W1185756938 hasConcept C162324750 @default.
• W1185756938 hasConcept C186644900 @default.
• W1185756938 hasConcept C187736073 @default.
• W1185756938 hasConcept C195324797 @default.

• next