Maximum Entropy (maxent) models are an attractive formalism for statistical models of many types and have been used for a number of purposes, including language modeling (Rosenfeld 1994), part of speech tagging (Ratnaparkhi 1996), prepositional phrase attachment (Ratnaparkhi 1998), sentence breaking (Reynar and Ratnaparkhi 1997) and parsing (Ratnaparkhi 1997). Maxent models allow the combinatio...

This paper describes work in progress that is identifying shortcomings of existing Prepositional Phrase (PP) attachment algorithms and producing a new resource derived from the Penn TreeBank (PTB) corpus. The aim is to use this new resource (PTB Prime) to improve the accuracy of PP attachment algorithms and use this in an existing text processing system (LaSIE-II).

We present a new approach to the classification problem for instances with discrete features, based on Bayesian statistics and explicit assumptions about the smoothness of the data. Combined with an optimizing method based on tailored genetic algorithm, our classifier yields performance which rivals that of the best existing methods for several problems like prepositional phrase attachment disa...

emphasis on sustainable agriculture in iran like the other developing countries, is responsedto the adverse environmental and economic impacts of conventional agriculture. thisproblem, particularly about “greenhouse production” is considerable. the perceptions ofgreenhouse owners about the environmental, economical and social aspects of sustainableagriculture was discussed in this article. the ...

This paper describes how to automatically classify the functional relations from the Factotum knowledge base via a statistical machine learning algorithm. This incorporates a method for inferring prepositional relation indicators from corpus data. It also uses lexical collocations (i.e., word associations) and class-based collocations based on the WordNet hypernym relations (i.e., is-subset-of)...

An approach to the machine translation of locative prepositional phrases (PP) is presented. The technique has been implemented for use in an experimental transfer-based, multi-lingual machine translation system. Previous approaches to this problem are described and they are compared to the solution presented.

This paper proposes an LMT analysis for valence alternations in German of the form NP V NP [P NP ] NP V NP [P NP ], where the indices denote referential identity. These alternations involve direct internal arguments (i.e., objects) and indirect prepositional complements, and characterize among others the behaviour of verbal predicates which participate in the so-called Locative Alternation phen...

Prepositional phrase (PP) attachment is one of the major sources for errors in traditional statistical parsers. The reason for that lies in the type of information necessary for resolving structural ambiguities. For parsing, it is assumed that distributional information of parts-of-speech and phrases is sufficient for disambiguation. For PP attachment, in contrast, lexical information is needed.