An important component of nativelike language production is linked to the knowledge of near-synonyms and their collocational patterns. The purpose of this corpus-based study was three-fold: (a) an examination of the frequency distribution across registers of the three near-synonyms change, alter, and modify; (b) their collocational patterns; and (c) their interchangeability. A lexical study on their distribution across registers in the BNC was conducted, examining the collocational strength between near-synonyms and their collocations in Sketch Engine. Subsequently, attempts were made to determine the potential interchangeability of the near-synonyms. The comparative analysis indicates that the near-synonyms differ from each other with reference to formality, as alter and modify tend to be used more in written and formal language than change. Further to this, and to some extent, they reveal different collocational patterns. More specifically, they often did not share collocates, although the verbs were near-synonyms. Change was more polysemous than alter and modify, which have more restricted collocational patterns. Furthermore, the interchangeability of change and alter were higher than for modify. Nevertheless, the formality and meaning were affected as change is less formal than alter and modify. The findings from the analyses are of value to second language learners and can support further knowledge in the field of linguistics, specific to near-synonyms and collocations.

This paper investigates the dominant metaphors in corpora constructed of British online news publications. It uses a combination of qualitative and quantitative analysis, consisting of corpus linguistics, discourse analysis, and agenda setting theory. The corpus, broken down into several sub-corpora, was investigated using collocate and KWIK tools in order to shed light on possible conceptual metaphors in the disease domain. The results showed that there were three major conceptual metaphor mappings; DISEASE IS A POSSESSION, PEOPLE ARE CONTAINERS and CONTROL IS UP. These metaphors in turn emphasised the individual’s culpability in the spreading of the disease and that the government’s preferred response was containing the disease. The most salient attributes in the disease discourse related to the severity of the disease or the government’s defensive response to the virus. In sum, the results provide discursive framings of COVID-19 disease in the British news and show how the conventional metaphor mappings function might affect the public opinion.

Two studies are presented, linking word-frequency information within the Global Corpus of Web-based English to real world facts. The first study concerns how patterns of the use of place names reflect geospatial and geopolitical relationships of English-speaking nations. The second study concerns how the emotional connotation of words before place names reflects general well-being in that place. Taken together, these studies demonstrate that the surface structure of language, as embodied in word frequencies, is a useful source of information about the real world. / Thesis / Master of Science (MSc) / This thesis involves two studies, using the Global Corpus of Web-based English. The first study shows how you can reconstruct a rudimentary map of English speaking countries of the world purely on the basis of how often different words happen in texts. The second study shows that when we discuss countries of the world online, how happy and exciting the adjectives are before place names relate to how long people live in that country.

An important component of nativelike language production is linked to the knowledge of near-synonyms and their collocational patterns. The purpose of this corpus-based study was three-fold: (a) an examination of the frequency distribution across registers of the three near-synonyms change, alter, and modify; (b) their collocational patterns; and (c) their interchangeability. A lexical study on their distribution across registers in the BNC was conducted, examining the collocational strength between near-synonyms and their collocations in Sketch Engine. Subsequently, attempts were made to determine the potential interchangeability of the near-synonyms. The comparative analysis indicates that the near-synonyms differ from each other with reference to formality, as alter and modify tend to be used more in written and formal language than change. Further to this, and to some extent, they reveal different collocational patterns. More specifically, they often did not share collocates, although the verbs were near-synonyms. Change was more polysemous than alter and modify, which have more restricted collocational patterns. Furthermore, the interchangeability of change and alter were higher than for modify. Nevertheless, the formality and meaning were affected as change is less formal than alter and modify. The findings from the analyses are of value to second language learners and can support further knowledge in the field of linguistics, specific to near-synonyms and collocations.

This study investigates the discursive ways in which the Syrian war and the Syrian refugees are being portrayed in the American news media. It was conducted by combining the method Corpus Linguistics with the concepts of Critical Discourse Analysis. The data for this research consists of a virtual sub-corpus created from the NOW (News on the Web) corpus, hosted by English-Corpora. The virtual corpus contains all the data from the NOW corpus with the word SYRIA* in the title and the words SYRIA* and REFUGEE* weref urther investigated by the use of concordance, collocate and KWIC tools. The results for SYRIA* showed that the Syrian war was not the prime focus for media coverage and were therefore underrepresented. The collocations around REFUGEE* demonstrated that the conceptual metaphor REFUGEES ARE WATER, and by extension, REFUGEES ARE A DISASTER, were mapped onto the Syrian refugees. Consequently, the nuisance and inconvenience of the refugees was emphasised and represented them negatively.

The objective of this thesis is to analyze the variation between two competing structures--on the one hand, a transitive verb that takes a finite nominal clause as its complement, and on the other hand, a transitive verb that takes an infinitive as its complement. This thesis seeks to address three questions: (1) Which semantic categories are more likely to use coreferential nominal clauses as complements? (2) How do coreferential finite nominal clauses compare with coreferential infinitive complements, in terms of frequency of usage? And (3) do the corpora show any variation among different countries? The corpora CREA and Web/Dialects were used to determine the frequency of usage between these two structures with four different semantic categories of verbs used in the main clause: assertive, dubitative, evaluative, and volitive. The U.S., Spain, Argentina, and Mexico were also used to compare the results by country. The findings show that when the main verb is assertive the use of a subordinate clause is favored, while main clauses with dubitative showed mixed results between the corpora, although Web/Dialects showed that dubitatives favor an infinitive complement. The evaluative verbs lamentar and odiar did not produce any coreferential results with direct object nominal clauses. Volitive verbs never accepted the use of a coreferential finite. The Web/Dialects results indicate that Spain and the U.S were more likely than Argentina and Mexico to use the finite construction after a main clause with a dubitative phrase, while still favoring the infinitive complement.

Many teachers, test designers, textbook writers, and instructional designers turn to books written by usage experts to determine what is correct standard written American English. Unfortunately, though, experts often disagree about what is correct and what is incorrect, and this disagreement can create problems with validity when people create and assess instruction about usage. One way to discover the rules of standard English usage is to describe what writers actually do in printed, edited English. Researchers can access large collections of standard English through digital text archives, which can be searched electronically. The text archives for this study were taken from EBSCO and ProQuest digital libraries and divided into three different registers: (a) newspapers, (b) magazines, and (c) scholarly journals. This study examines 30 representative items of controversial usage; such as "a lot" or "alot," "between you and I" or "between you and me," "had proved" or "had proven"; to determine the actual occurrence in these three registers of standard written American English. The results list the percentage of use in each register, as well as the total averaged percentage of use in all three registers. Items showing 90% to 100% usage in the total averaged percentages are considered standard English, but items showing 90% to 95% usage are borderline cases that should be monitored for future use. If a variant form is used more than 10% of the time, then it should be considered a possible alternative usage in dictionaries, in text books, and in tests. This study shows the results of using corpus linguistics to answer questions about usage in standard American English.

Dictionaries and, to a lesser extent, usage guides provide writers, editors, and users of American English information on how to use the language appropriately. Dictionaries, in particular, hold authority over correct usage of words. However, historically, usage guides and dictionaries were created using the knowledge of a small group of people. Lexicographers like Noah Webster set out to prescribe a proper way of using American English. To make these judgments, they often relied on a combination of study and idiosyncratic intuitions. A similar process took place in creating usage guides. Though these manuals profess to explain how the language is used by American English speakers-or rather by the selected group of speakers deemed "standard" by usage guide editors and lexicographers-ultimately the manuals can only express the perspectives of the editors and lexicographers on this language. Historically, the views of these editors and lexicographers were the best tools available to assess language, but now computer-based corpora allow for studying larger swaths of language usage. This study examines how much dictionaries and usage guides agree with real-world usage found in corpus data. Using the Corpus of Historical American English, a set of dictionaries and usage guides published throughout the last two hundred years were analyzed to see how much agreement they had with corpus data in noting the addition of denominal verbs (i.e., verbs formed by the conversion of nouns as in 'They taped together the box.') in American English usage. It was found that the majority of the time dictionaries noted new denominal verbs before corpus data reflected accepted usage of these verbs. However, about a quarter of the time dictionaries noted new denominal verbs concurrently with the corpus data. These results suggest that dictionaries-and the subjective opinions of the lexicographers that created them-are more aligned with real-world usage than would be expected. Because of sparse listings, results for usage guide agreement was inconclusive.

Vocabulary in a second language is an indispensable building block of all comprehension (Folse, 2006; Nation, 2006). Teachers in content area classes such as science, math, and social studies frequently teach content specific vocabulary, but are not aware of the obstacles that can occur when students do not know the basic words. Word lists such as the General Service List (GSL) were created to assist students and teachers (West, 1953). The GSL does not adequately take into account the high level of polysemy of many common English words, nor has it been updated by genre to reflect specific content domains encountered by secondary science students in today's high stakes classes such as chemistry. This study examines how many words of the first 1000 words of the GSL occurred in the secondary chemistry textbooks sampled, how often the first 1000 words of the GSL were polysemous, and specifically which multiple meanings occurred. A discussion of results includes word tables that list multiple meanings present, example phrases that illustrate the context surrounding the target words, suggestions for a GSL that is genre specific to secondary chemistry textbooks and that is ranked by meaning as well as type, and implications for both vocabulary materials and classroom instruction for ELLs in secondary chemistry classes. Findings are essential to second language (L2) researchers, materials developers, publishers, and teachers.

Communication Accommodation Theory (CAT) states that individuals adapt to each other’s communicative behaviors. This adaptation is called “convergence.” In this work we explore the convergence of writing styles of users of the online music distribution plat- form SoundCloud.com. In order to evaluate our system we created a corpus of over 38,000 comments retrieved from SoundCloud in April 2014. The corpus represents comments from 8 distinct musical genres: Classical, Electronic, Hip Hop, Jazz, Country, Metal, Folk, and World. Our corpus contains: short comments, frequent misspellings, little sentence struc- ture, hashtags, emoticons, and URLs. We adapt techniques used by researchers analyzing other short web-text corpora in order to deal with these problems. We use a supervised machine learning approach to classify the genre of comments in our corpus. We examine the effects of different feature sets and supervised machine learning algorithms on classification accuracy. In total we ran 180 experiments in which we varied: number of genres, feature set composition, and machine learning algorithm. In experiments with all 8 genres we achieve up to 40% accuracy using either a Naive Bayes classifier or C4.5 based classifier with a feature set consisting of 1262 token unigrams and bigrams. This represents a 3 time improvement over chance levels.