We consider translating natural language sentences into a formal language using direct translation models built automatically from training data. Direct translation models have three components: an arbitrary prior conditional probability distribution, features that capture correlations between automatically determined key phrases or sets of words in both languages, and weights associated with t...

This paper describes how any theory which assigns propositions to conditional sentences can be lifted to the setting of inquisitive semantics, where antecedents and consequents may be associated with multiple propositions. We show that the lifted account improves on the original account in two ways: first, it leads to a better analysis of disjunctive antecedents, which are treated as introducin...

Third class conditional sentences, a very frequent type of conditional sentence, are identified and characterized by their use of the subjunctive mood in the protasis. The subjunctive indicates potentiality, contingency, or simple futurity. It is the condition which points to a future eventuality. The common notion that it indicates a degree of probability is examined by inductive study of all ...

This paper describes experiments in classifying sentences of medical abstracts into a number of semantic classes given by section headings in structured abstracts. Using conditional random fields, we obtain F -scores ranging from 0.72 to 0.97. By using a small set of sentences that appear under the PARTICPANTS heading, we demonstrate that it is possible to recognize sentences that describe popu...

This book offers a new and in-depth analysis of English conditional sentences. In a wide-ranging discussion, Dancygier classifies conditional constructions according to time-reference and modality. She shows how the basic meaning parameters of conditionality, such as causation and logical sequence, correlate to formal parameters of the linguistic constructions which are used to express them. Da...

Existing models based on artificial neural networks (ANNs) for sentence classification often do not incorporate the context in which sentences appear, and classify sentences individually. However, traditional sentence classification approaches have been shown to greatly benefit from jointly classifying subsequent sentences, such as with conditional random fields. In this work, we present an ANN...

The question whether material implication is a proper translation of “if ... then ...” in conditional sentences has been known and discussed as “material implication paradox”. The source of this paradox comes from the interpretation of material implication that any falsehood, or contradiction, entails every proposition. For example, consider truth conditions of the conditional sentence (1) and ...

Are people willing and able to disambiguate content for the Semantic Web? We asked subjects to use two methods (paraphrasal and truth-conditional selection) to disambiguate sentences from the Web. Native speakers did better with the paraphrasal method, and non-native speakers with the truth-conditional method. Unpaid volunteers performed better than paid subjects. Subjects’ average disambiguati...

The paper discusses a prominent one of Kratzer’s (1986, 1991, 2012) arguments against material implication analyses the denotation (indicative) conditional sentences. This is argument based on sentence Most time, if man buys horse, he pays cash for it. It shown that makes prediction does conform to speakers’ intuitions, contrary claim. also argues Lewis’s (1975) attack sentences examples where ...

Many languages allow for ”fake” uses of their past tense marker: the marker can occur in certain contexts without conveying temporal pastness. Instead it appears to bear a modal meaning. Iatridou [2000] has coined this phenomenon Fake Tense. Fake Tense is particularly common to conditional construction. This paper proposes to explain Fake Tense in English conditional sentences as a certain kind...