Euphemisms have not received much attention in natural language processing, despite being an important element of polite and figurative language. Euphemisms prove to be a difficult topic, not only because they are subject to language change, but also because humans may not agree on what is a euphemism and what is not. Nonetheless, the first step to tackling the issue is to collect and analyze examples of euphemisms. We present a corpus of potentially euphemistic terms (PETs) along with example texts from the GloWbE corpus. Additionally, we present a subcorpus of texts where these PETs are not being used euphemistically, which may be useful for future applications. We also discuss the results of multiple analyses run on the corpus. Firstly, we find that sentiment analysis on the euphemistic texts supports that PETs generally decrease negative and offensive sentiment. Secondly, we observe cases of disagreement in an annotation task, where humans are asked to label PETs as euphemistic or not in a subset of our corpus text examples. We attribute the disagreement to a variety of potential reasons, including if the PET was a commonly accepted term (CAT).

This paper studies how the linguistic components of blogposts collected from Sina Weibo, a Chinese microblogging platform, might affect the blogposts' likelihood of being censored. Our results go along with King et al. (2013)'s Collective Action Potential (CAP) theory, which states that a blogpost's potential of causing riot or assembly in real life is the key determinant of it getting censored. Although there is not a definitive measure of this construct, the linguistic features that we identify as discriminatory go along with the CAP theory. We build a classifier that significantly outperforms non-expert humans in predicting whether a blogpost will be censored. The crowdsourcing results suggest that while humans tend to see censored blogposts as more controversial and more likely to trigger action in real life than the uncensored counterparts, they in general cannot make a better guess than our model when it comes to ‘reading the mind’ of the censors in deciding whether a blogpost should be censored. We do not claim that censorship is only determined by the linguistic features. There are many other factors contributing to censorship decisions. The focus of the present paper is on the linguistic form of blogposts. Our work suggests that it is possible to use linguistic properties of social media posts to automatically predict if they are going to be censored.

As big data heads towards big knowledge, data management and machine learning techniques work together to address several interesting problems. In this paper, we address a problem in natural language processing that involves learning by mining from large text databases. More specifically, we deal with the problem of preposition prediction, especially for ESL (English as a second language) learners. Prepositions are function words that typically show a relationship between a noun or a pronoun and other elements of a sentence. They play a key role in determining the meaning of a sentence. Accurate prediction of correct prepositions in a sentence is a challenging job since preposition usage is one of the most subtle aspects of the English grammar, making it difficult for non-native speakers. This paper proposes an approach for preposition prediction called WordPrep based on which we build a tool. WordPrep relies on mining based on the words themselves rather than on their lexical or syntactic connotations. This addresses the challenges of prepositions appearing in idiomatic phrases or in different semantic contexts, due to which the actual words are better than their grammatical positions. Our proposed solution entails a direct data-driven approach to predict the missing preposition in a sentence by learning from matching tokens consisting of ngrams with words before and after the preposition. Using various searches and pattern-matching methods against a large number of database records from big text corpora, this approach predicts the missing preposition(s). We describe our pilot approach, tool implementation and experiments in this paper. This work is particularly helpful for pedagogical applications.

Ribosome queuing is a fundamental phenomenon suggested to be related to topics such as genome evolution, synthetic biology, gene expression regulation, intracellular biophysics, and more. However, this phenomenon hasn't been quantified yet at a genomic level. Nevertheless, methodologies for studying translation (e.g. ribosome footprints) are usually calibrated to capture only single ribosome protected footprints (mRPFs) and thus limited in their ability to detect ribosome queuing. On the other hand, most of the models in the field assume and analyze a certain level of queuing. Here we present an experimental-computational approach for studying ribosome queuing based on sequencing of RNA footprints extracted from pairs of ribosomes (dRPFs) using a modified ribosome profiling protocol. We combine our approach with traditional ribosome profiling to generate a detailed profile of ribosome traffic. The data are analyzed using computational models of translation dynamics. The approach was implemented on the Saccharomyces cerevisiae transcriptome. Our data shows that ribosome queuing is more frequent than previously thought: the measured ratio of ribosomes within dRPFs to mRPFs is 0.2-0.35, suggesting that at least one to five translating ribosomes is in a traffic jam; these queued ribosomes cannot be captured by traditional methods. We found that specific regions are enriched with queued ribosomes, such as the 5'-end of ORFs, and regions upstream to mRPF peaks, among others. While queuing is related to higher density of ribosomes on the transcript (characteristic of highly translated genes), we report cases where traffic jams are relatively more severe in lowly expressed genes and possibly even selected for. In addition, our analysis demonstrates that higher adaptation of the coding region to the intracellular tRNA levels is associated with lower queuing levels. Our analysis also suggests that the Saccharomyces cerevisiae transcriptome undergoes selection for eliminating traffic jams. Thus, our proposed approach is an essential tool for high resolution analysis of ribosome traffic during mRNA translation and understanding its evolution.

Some expressions can be ambiguous between idiomatic and literal interpretations depending on the context they occur in, e.g., ‘sales hit the roof’ vs. ‘hit the roof of the car’. We present a novel method of classifying whether a given instance is literal or idiomatic, focusing on verb-noun constructions. We report state-of-the-art results on this task using an approach based on the hypothesis that the distributions of the contexts of the idiomatic phrases will be different from the contexts of the literal usages. We measure contexts by using projections of the words into vector space. For comparison, we implement Fazly et al. (2009)’s, Sporleder and Li (2009)’s, and Li and Sporleder (2010b)’s methods and apply them to our data. We provide experimental results validating the proposed techniques.

We describe an algorithm for automatic classification of idiomatic and literal expressions.Our starting point is that words in a given text segment, such as a paragraph, that are highranking representatives of a common topic of discussion are less likely to be a part of an idiomatic expression.Our additional hypothesis is that contexts in which idioms occur, typically, are more affective and therefore, we incorporate a simple analysis of the intensity of the emotions expressed by the contexts.We investigate the bag of words topic representation of one to three paragraphs containing an expression that should be classified as idiomatic or literal (a target phrase).We extract topics from paragraphs containing idioms and from paragraphs containing literals using an unsupervised clustering method, Latent Dirichlet Allocation (LDA) (Blei et al., 2003).Since idiomatic expressions exhibit the property of non-compositionality, we assume that they usually present different semantics than the words used in the local topic.We treat idioms as semantic outliers, and the identification of a semantic shift as outlier detection.Thus, this topic representation allows us to differentiate idioms from literals using local semantic contexts.Our results are encouraging.

This paper describes a system called SITAC based on our proposed approach to discover concepts (called SITACs) in text archives that are identical semantically but alter their names over time. Our approach integrates natural language processing, association rule mining and contextual similarity to discover SITACs in order to answer historical queries over text corpora.

Time-stamped documents such as newswire articles, blog posts and other web-pages are often archived online. When these archives cover long spans of time, the terminology within them could undergo significant changes. Hence, when users pose queries pertaining to historical information, over such documents, the queries need to be translated, taking into account these temporal changes, to provide accurate responses to users. For example, a query on Sri Lanka should automatically retrieve documents with its former name Ceylon. We call such concepts SITACs, i.e., Semantically Identical Temporally Altering Concepts. In order to discover SITACs, we propose an approach based on a novel framework constituting an integration of natural language processing, association rule mining, and contextual similarity as a learning technique. The proposed approach has been experimented with real data and has been found to yield good results with respect to efficiency and accuracy.

Text classification is a widely studied topic in the area of machine learning. A number of techniques have been developed to represent and classify text documents. Most of the techniques try to achieve good classification performance while taking a document only by its words (e.g. statistical analysis on word frequency and distribution patterns). One of the recent trends in text classification research is to incorporate more semantic interpretation in text classification, especially by using Wikipedia. This paper introduces a technique for incorporating the vast amount of human knowledge accumulated in Wikipedia into text representation and classification. The aim is to improve classification performance by transforming general terms into a set of related concepts grouped around semantic themes. In order to achieve this goal, this paper proposes a unique method for breaking the enormous amount of extracted Wikipedia knowledge (concepts) into smaller pieces (subsets of concepts). The subsets of concepts are separately used to represent the same set of documents in a number of different ways, from which an ensemble of classifiers is built. Experimental results show that an ensemble of classifiers individually trained on a different representation of the document set performs better with increased accuracy and stability than that of a classifier trained only on the original document set.

The main objective of Bolshakov and Gelbukh's new textbook in computational linguistics is to provide students in computer science with a foundation in the fundamentals of general linguistics necessary to develop applied software systems, and to enable students to make informed choices of proper models and data structures.Although Spanish-language applications are emphasized, the textbook contains examples from English, French, Portuguese, and Russian as well.This text is freely available in an electronic format.Supplementary materials, links to the relevant resources, and errata can be found on-line as well.The authors approach linguistic problems using two theoretical frameworks: meaning-text theory (MTT) (Mel'čuk 1974), and head-driven phrase structure grammar (HPSG) (Pollard and Sag 1994).These choices reflect primarily practical considerations.On the one hand, MTT can be used to describe any language but is particularly wellsuited for the description of free-word-order languages.On the other hand, HPSG is arguably the most advanced, widely used, and user-friendly formalism in naturallanguage description and processing.The structure of the book is as follows.Chapter I gives an introduction to different areas of linguistics and provides an overview of the state of the art in Spanish natural language processing (NLP).Chapter II is a brief survey of the history of linguistics from Ferdinand de Saussure to Leonard Bloomfield to Noam Chomsky, discussing context-free grammars, transformational grammars, and more up-to-date approaches to grammar.Covering concepts such as valence, constraints, and unification formalism, the authors introduce Fillmore's work (Fillmore 1968), generalized phrase structure grammar (Gazdar et al. 1985), and HPSG.The last several sections are devoted to MTT.MTT has influenced the design of new grammar formalisms such as Dependency Tree Grammars, and some aspects have been adopted into other theories (e.g., integrating lexical functions into Pustejovsky's Generative Lexicon; Wanner 1997).Chapter III provides a sketch of computational linguistics (CL) applications.The included discussion serves to illustrate that all of the relevant applications require sophisticated linguistic knowledge.The general lesson of this chapter can be summarized thus: "Most language processing tasks can be considered as special cases of the general task of language understanding, one of the ultimate goals of CL and AI."Chapter IV describes the MTT theory in more detail and makes interesting comparisons between MTT, HPSG, and Chomskian frameworks.Chapter V introduces the problem of language modeling in CL.The last section of the book includes various exercises, review questions, and multiple-choice problems.This book is a nice addition to the short list of CL textbooks.It is especially helpful for students in Spanish CL, given its numerous Spanish-related examples and