We present a new framework for classifying common nouns that extends namedentity classification. We used a fixed set of 26 semantic labels, which we called supersenses. These are the labels used by lexicographers developing WordNet. This framework has a number of practical advantages. We show how information contained in the dictionary can be used as additional training data that improves accur...

Previous research has demonstrated the utility of clustering in inducing semantic verb classes from undisambiguated corpus data. We describe a new approach which involves clustering subcategorization frame (SCF) distributions using the Information Bottleneck and nearest neighbour methods. In contrast to previous work, we particularly focus on clustering polysemic verbs. A novel evaluation schem...

―Tree SRL system‖ is a Semantic Role Labelling supervised system based on a tree-distance algorithm and a simple k-NN implementation. The novelty of the system lies in comparing the sentences as tree structures with multiple relations instead of extracting vectors of features for each relation and classifying them. The system was tested with the English CoNLL-2009 shared task data set where 79%...

We describe an approach to classifying images found on the WWW for the purpose of information extraction (IE). Among features used for classification are image sizes, colour histograms, and the similarity of the classified image’s content to images in a training collection. Our content similarity metric is based on the latent semantic index. Results are presented on a collection of 1624 image o...

This paper describes an algorithm whereby an initial, naïve user query to a search engine can be subsequently refined to improve both its recall and precision. This is achieved by manually classifying the documents retrieved by the original query into relevant and irrelevant categories, and then finding additional Boolean terms which successfully discriminate between these categories. Latent se...

There is a constant need to extend and tune medical vocabularies to account for new words and new word usages. Robust natural language processing (NLP) tools can be applied to medical texts corpora such as patient narratives and help collect and analyze unknown words1,2. The aim of the present work is to assess the potential for classifying unknown words based on the semantic categories of “nei...

Semantic segmentation is the task of clustering together parts an image that belong to same object class. webpages important for inferring contextual information from webpage. This study examines and compares deep learning methods classifying based on imagery obscured by semantic segmentation. Fully convolutional neural network architectures (UNet FCN-8) with defined hyperparameters loss functi...

This paper presents a generative model based on the language modeling approach for sentiment analysis. By characterizing the semantic orientation of documents as “favorable” (positive) or “unfavorable” (negative), this method captures the subtle information needed in text retrieval. In order to conduct this research, a language model based method is proposed to keep the dependent link between a...

In this paper, we describe a multi-tier Natural Language (NL) clustering approach to text classification for classifying students’ essays for tutoring applications. The main task of the classifier is to map the students’ essay statements into target concepts, namely physics principles and misconceptions. A simple `Bag-Of-Words (BOW)’ classifier using a naïve-Bayes algorithm was unsatisfactory f...

Discriminating between similar languages or language varieties aims to detect lexical and semantic variations in order to classify these varieties of languages. In this work we describe the system built by the Pattern Recognition and Human Language Technology (PRHLT) research center Universitat Politècnica de València and Autoritas Consulting for the Discriminating between similar languages (DS...