A new thresholding strategy for a text categorization problem is proposed. It is based on Zadeh’s calculus of linguistically quantified propositions. The strategy may be also interpreted in terms of fuzzy integral.

Sentiment lexicons and other linguistic knowledge proved to be beneficial in polarity classification. This paper introduces a linguistically informed Convolutional Neural Network (lingCNN), which incorporates this valuable kind of information into the model. We present two intuitive and simple methods: The first one integrates word-level features, the second sentence-level features. By combinin...

An increasing number of enterprises are beginning to include semantic web ontologies into their Information Extraction (IE) and Text Analytics (TA) applications. This can be challenging for a TA group wishing to avail of semantic web ontologies due to the manual effort of retargeting and tailoring language resources within the TA system to a new domain to meet customer needs. A lightweight lexi...

We try to improve the classifier-based deterministic dependency parsing in two ways: by introducing a better search method based on a non-deterministic nbest algorithm and by devising a series of linguistically richer models. It is experimentally shown on a ConLL 2007 shared task that this results in a system with higher performance while still keeping it simple enough for an efficient implemen...

Sentiment understanding has been a long-term goal of AI in the past decades. This paper deals with sentence-level sentiment classification. Though a variety of neural network models have been proposed very recently, however, previous models either depend on expensive phrase-level annotation, whose performance drops substantially when trained with only sentence-level annotation; or do not fully ...

Systems now exist which are able to compile uni cation grammars into language models that can be included in a speech recognizer, but it is so far unclear whether non-trivial linguistically principled grammars can be used for this purpose. We describe a series of experiments which investigate the question empirically, by incrementally constructing a grammar and discovering what problems emerge ...

This paper deals with sentence-level sentiment classification. Though a variety of neural network models have been proposed recently, however, previous models either depend on expensive phrase-level annotation, most of which has remarkably degraded performance when trained with only sentence-level annotation; or do not fully employ linguistic resources (e.g., sentiment lexicons, negation words,...

In recent years, there has been a lot of research on wide-coverage statistical natural language processing with linguistically expressive grammars such as Combinatory Categorial Grammars (CCG), Head-driven Phrase-Structure Grammars (HPSG), Lexical-Functional Grammars (LFG) and Tree-Adjoining Grammars (TAG). But although many young researchers in natural language processing are very well trained...

Usually, the development of an information system (or some part of it) starts with an requirements elicitation, collection and analysis phase that results in a set of natural language requirements specifications. These then serve as a source for the phase of conceptual design where a semantic model of the given universe of discourse is established using modeling and representation concepts like...

In this paper, we propose a probabilistic phrase alignment model based on dependency trees. This model is linguistically-motivated, using syntactic information during alignment process. The main advantage of this model is that the linguistic difference between source and target languages is successfully absorbed. It is composed of twomodels: Model1 is using content word translation probability ...