Word Sense Discovery and Disambiguation

The work is based on the assumption that words with similar syntactic usage have similar meaning, which was proposed by Zellig S. Harris (1954,1968). We study his assumption from two aspects: Firstly, different meanings (word senses) of a word should manifest themselves in different usages (contexts), and secondly, similar usages (contexts) should lead to similar meanings (word senses). If we start with the different meanings of a word, we should be able to find distinct contexts for the meanings in text corpora. We separate the meanings by grouping and labeling contexts in an unsupervised or weakly supervised manner (Publication 1, 2 and 3). We are confronted with the question of how best to represent contexts in order to induce effective classifiers of contexts, because differences in context are the only means we have to separate word senses. If we start with words in similar contexts, we should be able to discover similarities in meaning. We can do this monolingually or multilingually. In the monolingual material, we find synonyms and other related words in an unsupervised way (Publication 4). In the multilingual material, we ?nd translations by supervised learning of transliterations (Publication 5). In both the monolingual and multilingual case, we first discover words with similar contexts, i.e., synonym or translation lists. In the monolingual case we also aim at finding structure in the lists by discovering groups of similar words, e.g., synonym sets. In this introduction to the publications of the thesis, we consider the larger background issues of how meaning arises, how it is quantized into word senses, and how it is modeled. We also consider how to define, collect and represent contexts. We discuss how to evaluate the trained context classi?ers and discovered word sense classifications, and ?nally we present the word sense discovery and disambiguation methods of the publications. This work supports Harris' hypothesis by implementing three new methods modeled on his hypothesis. The methods have practical consequences for creating thesauruses and translation dictionaries, e.g., for information retrieval and machine translation purposes. Keywords: Word senses, Context, Evaluation, Word sense disambiguation, Word sense discovery.

Readers' resposes versus reader-response theories : An empirical study

This study examines both theoretically an empirically how well the theories of Norman Holland, David Bleich, Wolfgang Iser and Stanley Fish can explain readers' interpretations of literary texts. The theoretical analysis concentrates on their views on language from the point of view of Wittgenstein's Philosophical Investigations. This analysis shows that many of the assumptions related to language in these theories are problematic. The empirical data show that readers often form very similar interpretations. Thus the study challenges the common assumption that literary interpretations tend to be idiosyncratic. The empirical data consists of freely worded written answers to questions on three short stories. The interpretations were made by 27 Finnish university students. Some of the questions addressed issues that were discussed in large parts of the texts, some referred to issues that were mentioned only in passing or implied. The short stories were "The Witch à la Mode" by D. H. Lawrence, "Rain in the Heart" by Peter Taylor and "The Hitchhiking Game" by Milan Kundera. According to Fish, readers create both the formal features of a text and their interpretation of it according to an interpretive strategy. People who agree form an interpretive community. However, a typical answer usually contains ideas repeated by several readers as well as observations not mentioned by anyone else. Therefore it is very difficult to determine which readers belong to the same interpretive community. Moreover, readers with opposing opinions often seem to pay attention to the same textual features and even acknowledge the possibility of an opposing interpretation; therefore they do not seem to create the formal features of the text in different ways. Iser suggests that an interpretation emerges from the interaction between the text and the reader when the reader determines the implications of the text and in this way fills the "gaps" in the text. Iser believes that the text guides the reader, but as he also believes that meaning is on a level beyond words, he cannot explain how the text directs the reader. The similarity in the interpretations and the fact that the agreement is strongest when related to issues that are discussed broadly in the text do, however, support his assumption that readers are guided by the text. In Bleich's view, all interpretations have personal motives and each person has an idiosyncratic language system. The situation where a person learns a word determines the most important meaning it has for that person. In order to uncover the personal etymologies of words, Bleich asks his readers to associate freely on the basis of a text and note down all the personal memories and feelings that the reading experience evokes. Bleich's theory of the idiosyncratic language system seems to rely on a misconceived notion of the role that ostensive definitions have in language use. The readers' responses show that spontaneous associations to personal life seem to colour the readers' interpretations, but such instances are rather rare. According to Holland, an interpretation reflects the reader's identity theme. Language use is regulated by shared rules, but everyone follows the rules in his or her own way. Words mean different things to different people. The problem with this view is that if there is any basis for language use, it seems to be the shared way of following linguistic rules. Wittgenstein suggests that our understanding of words is related to the shared ways of using words and our understanding of human behaviour. This view seems to give better grounds for understanding similarity and differences in literary interpretations than the theories of Holland, Bleich, Fish and Iser.

Identification and host-plant associations of Australian Sericothripinae (Thysanoptera, Thripidae)

The Sericothripinae is a largely tropical group of about 140 species that are often strikingly bicoloured and have complex surface sculpture, but for which the biology is poorly known. Although 15 genera have been described in this subfamily, only three of these are currently recognised, with five new generic synonymies indicated here. In Australia, Sericothrips Haliday is introduced, with one European species deployed as a weed biological control agent. Hydatothrips Karny comprises 43 species worldwide, with six species found in Australia, of which two are shared with Southeast Asia, and four are associated with the native vine genus, Parsonsia. Neohydatothrips John comprises 96 species worldwide, with nine species in Australia, of which one is shared with Southeast Asia and two are presumably introduced from the Americas. Illustrated keys are provided to the three genera and 16 species from Australia, including six new species [Hydatothrips aliceae; H. bhattii; H. williamsi; Neohydatothrips barrowi, N. bellissi, N. katherinae]. One new specific synonym is recognised [Hydatothrips haschemi Girault (= H. palawanensis Kudo)], also four new generic synonyms [Neohydatothrips John (= Faureana Bhatti; Onihothrips Bhatti; Sariathrips Bhatti; Papiliothrips Bhatti); Sericothrips Haliday (= Sussericothrips Han)].

Condition-specific associations of symptoms of depression and anxiety in adolescents and young adults with asthma and food allergy

Objective: This study examined associations of asthma and food allergy with symptoms of depression and anxiety at 14 and 21 years of age to determine whether condition-specific associations exist. Methods: Data come from 4972 adolescents in the Mater University Study of Pregnancy. Symptoms of depression and anxiety were assessed using the Youth Self-Report and Young Adult Self-Report. Results: Condition-specific associations between asthma and depression, OR=1.37 [1.12, 1.67] and between food allergy and anxiety, OR=1.26 [1.04, 1.76] were found during adolescence, but not in young adulthood. Whereas asthma was associated with resolved depression, OR=1.70 [1.13, 2.55], food allergy was associated with persistent anxiety, OR=1.26 [1.01, 1.59]. Conclusions: In adolescents, asthma is associated with an increased risk of clinically relevant symptoms of depression and food allergy with and increased risk of clinically relevant symptoms of anxiety. Future research is needed to clarify directionality and mechanisms explaining these relationships. Health professionals should be aware of the increased risk of mental health problems in adolescents with asthma or food allergy.

Language in Acquisition : Early Lexical Development and Associations between Lexicon and Grammar Findings from Full-Term and Very-Low-Birth-Weight Finnish Children

The aim was to analyse the growth and compositional development of the receptive and expressive lexicons between the ages 0,9 and 2;0 in the full-term (FT) and the very-low-birth-weight (VLBW) children who are acquiring Finnish. The associations between the expressive lexicon and grammar at 1;6 and 2;0 in the FT children were also studied. In addition, the language skills of the VLBW children at 2;0 were analysed, as well as the predictive value of early lexicon to the later language performance. Four groups took part in the studies: the longitudinal (N = 35) and cross-sectional (N = 146) samples of the FT children, and the longitudinal (N = 32) and cross-sectional (N = 66) samples of VLBW children. The data was gathered by applying of the structured parental rating method (the Finnish version of the Communicative Development Inventory), through analysis of the children´s spontaneous speech and by administering a a formal test (Reynell Developmental Language Scales). The FT children acquired their receptive lexicons earlier, at a faster rate and with larger individual variation than their expressive lexicons. The acquisition rate of the expressive lexicon increased from slow to faster in most children (91%). Highly parallel developmental paths for lexical semantic categories were detected in the receptive and expressive lexicons of the Finnish children when they were analysed in relation to the growth of the lexicon size, as described in the literature for children acquiring other languages. The emergence of grammar was closely associated with expressive lexical growth. The VLBW children acquired their receptive lexicons at a slower rate and had weaker language skills at 2;0 than the full-term children. The compositional development of both lexicons happened at a slower rate in the VLBW children when compared to the FT controls. However, when the compositional development was analysed in relation to the growth of lexicon size, this development occurred qualitatively in a nearly parallel manner in the VLBW children as in the FT children. Early receptive and expressive lexicon sizes were significantly associated with later language skills in both groups. The effect of the background variables (gender, length of the mother s basic education, birth weight) on the language development in the FT and the VLBW children differed. The results provide new information of early language acquisition by the Finnish FT and VLBW children. The results support the view that the early acquisition of the semantic lexical categories is related to lexicon growth. The current findings also propose that the early grammatical acquisition is closely related to the growth of expressive vocabulary size. The language development of the VLBW children should be followed in clinical work.

Genetics of heifer puberty in two tropical beef genotypes in northern Australia and associations with heifer- and steer-production traits

A total of 2115 heifers from two tropical genotypes (1007 Brahman and 1108 Tropical Composite) raised in four locations in northern Australia were ovarian-scanned every 4-6 weeks to determine the age at the first-observed corpus luteum (CL) and this was used to de. ne the age at puberty for each heifer. Other traits recorded at each time of ovarian scanning were liveweight, fat depths and body condition score. Reproductive tract size was measured close to the start of the first joining period. Results showed significant effects of location and birth month on the age at first CL and associated puberty traits. Genotypes did not differ significantly for the age or weight at first CL; however, Brahman were fatter at first CL and had a small reproductive tract size compared with that of Tropical Composite. Genetic analyses estimated the age at first CL to be moderately to highly heritable for Brahman (0.57) and Tropical Composite (0.52). The associated traits were also moderately heritable, except for reproductive tract size in Brahmans (0.03) and for Tropical Composite, the presence of an observed CL on the scanning day closest to the start of joining (0.07). Genetic correlations among puberty traits were mostly moderate to high and generally larger in magnitude for Brahman than for Tropical Composite. Genetic correlations between the age at CL and heifer- and steer-production traits showed important genotype differences. For Tropical Composite, the age at CL was negatively correlated with the heifer growth rate in their first postweaning wet season (-0.40) and carcass marbling score (-0.49), but was positively correlated with carcass P8 fat depth (0.43). For Brahman, the age at CL was moderately negatively genetically correlated with heifer measures of bodyweight, fatness, body condition score and IGF-I, in both their first postweaning wet and second dry seasons, but was positively correlated with the dry-season growth rate. For Brahman, genetic correlations between the age at CL and steer traits showed possible antagonisms with feedlot residual feed intake (-0.60) and meat colour (0.73). Selection can be used to change the heifer age at puberty in both genotypes, with few major antagonisms with steer- and heifer- production traits.

Integrating and evaluating neural word embeddings in information retrieval

Recent advances in neural language models have contributed new methods for learning distributed vector representations of words (also called word embeddings). Two such methods are the continuous bag-of-words model and the skipgram model. These methods have been shown to produce embeddings that capture higher order relationships between words that are highly effective in natural language processing tasks involving the use of word similarity and word analogy. Despite these promising results, there has been little analysis of the use of these word embeddings for retrieval. Motivated by these observations, in this paper, we set out to determine how these word embeddings can be used within a retrieval model and what the benefit might be. To this aim, we use neural word embeddings within the well known translation language model for information retrieval. This language model captures implicit semantic relations between the words in queries and those in relevant documents, thus producing more accurate estimations of document relevance. The word embeddings used to estimate neural language models produce translations that differ from previous translation language model approaches; differences that deliver improvements in retrieval effectiveness. The models are robust to choices made in building word embeddings and, even more so, our results show that embeddings do not even need to be produced from the same corpus being used for retrieval.

Associations with low rates of postpartum glucose screening after gestational diabetes among Indigenous and non-Indigenous Australian women

OBJECTIVES To explore factors associated with postpartum glucose screening among women with Gestational Diabetes Mellitus (GDM). METHODS A retrospective study using linked records from women with GDM who gave birth at Cairns Hospital in Far North Queensland, Australia, from 1 January 2004 to 31 December 2010. RESULTS The rates of postpartum Oral Glucose Tolerance Test (OGTT) screening, while having increased significantly among both Indigenous* and non-Indigenous women from 2004 to 2010 (HR 1.15 per year, 95%CI 1.08-1.22, p<0.0001), remain low, particularly among Indigenous women (10% versus 27%, respectively at six months postpartum). Indigenous women in Cairns had a longer time to postpartum OGTT than Indigenous women in remote areas (HR 0.58, 0.38-0.71, p=0.01). Non-Indigenous women had a longer time to postpartum OGTT if they: were born in Australia (HR 0.76, 0.59-1.00, 0.05); were aged <25 years (HR 0.45, 0.23-0.89, p=0.02); had parity >5 (HR 0.33, 0.12-0.90, p=0.03); smoked (HR 0.48, 0.31-0.76, p=0.001); and did not breastfeed (HR 0.09, 0.01-0.64, p=0.02). CONCLUSIONS Postpartum diabetes screening rates following GDM in Far North Queensland are low, particularly among Indigenous women, with lower rates seen in the regional centre; and among non-Indigenous women with indicators of low socioeconomic status. IMPLICATIONS Strategies are urgently needed to improve postpartum diabetes screening after GDM that reach women most at risk.

Associations among Emdogain®, human oral carcinoma cells and oral proteolytic enzymes

The Enamel matrix derivative Emdogain® (EMD) is a commercially available tissue extract preparation of porcine enamel origin. Studies have shown EMD to be clinically useful in promoting periodontal regeneration. EMD has been widely used in periodontal therapy for over ten years, but the mechanism of its action and the exact composition are not completely clear. EMD is predominantly amelogenin (>90%). However, unlike amelogenin, EMD has a number of growth factor-like effects and it has been shown to enhance the proliferation, migration and other cellular functions of periodontal ligament fibroblasts and osteoblasts. In contrast, the effects of EMD on epithelial cell lines and in particular on oral malignant cells have not been adequately studied. In addition, EMD has effects on the production of cytokines by several oral cell lines and the product is in constant interaction with different oral enzymes. Regardless of the various unknown properties of EMD, it is said to be clinically safe in regenerative procedures, also in medically compromised patients. The aim of the study was to examine whether gingival crevicular fluid (GCF), which contains several different proteolysis enzymes, could degrade EMD and alter its biological functions. In addition, the objective was to study the effects of EMD on carcinogenesis-related factors, in particular MMPs, using in vitro and in vivo models. This study also aimed to contribute to the understanding of the composition of EMD. GCF was capable of degrading EMD, depending on the periodontal status, with markedly more degradation in all states of periodontal disease compared to healthy controls. EMD was observed to stimulate the migration of periodontal ligament fibroblasts (PLF), whereas EMD together with GCF could not stimulate this proliferation. In addition, recombinant amelogenin, the main component of EMD, decreased the migration of PLFs. A comparison of changes induced by EMD and TGF-β1 in the gene profiles of carcinoma cells showed TGF-β1 to regulate a greater number of genes than EMD. However, both of the study reagents enhanced the expression of MMP-10 and MMP-9. Furthermore, EMD was found to induce several factors closely related to carcinogenesis on gene, protein, cell and in vivo levels. EMD enhanced the production of MMP-2, MMP-9 and MMP-10 proteins by cultured carcinoma cells. In addition, EMD stimulated the migration and in vitro wound closure of carcinoma cells. EMD was also capable of promoting metastasis formation in mice. In conclusion, the diseased GCF, containing various proteases, causes degradation of EMD and decreased proliferation of PLFs. Thus, this in vitro study suggests that the regenerative effect of EMD may decrease due to proteases present in periodontal tissues during the inflammation and healing of the tissues in vivo. Furthermore, EMD was observed to enhance several carcinoma-related factors and in particular the production of MMPs by benign and malignant cell lines. These findings suggest that the clinical safety of EMD with regard to dysplastic mucosal lesions should be further investigated.

Loci affecting gamma-glutamyl transferase in adults and adolescents show age x SNP interaction and cardiometabolic disease associations

Serum gamma-glutamyl transferase (GGT) activity is a marker of liver disease which is also prospectively associated with the risk of all-cause mortality, cardiovascular disease, type 2 diabetes and cancers. We have discovered novel loci affecting GGT in a genome-wide association study (rs1497406 in an intergenic region of chromosome 1, P = 3.9 x 10(-8); rs944002 in C14orf73 on chromosome 14, P = 4.7 x 10(-13); rs340005 in RORA on chromosome 15, P = 2.4 x 10(-8)), and a highly significant heterogeneity between adult and adolescent results at the GGT1 locus on chromosome 22 (maximum P(HET) = 5.6 x 10(-12) at rs6519520). Pathway analysis of significant and suggestive single-nucleotide polymorphism associations showed significant overlap between genes affecting GGT and those affecting common metabolic and inflammatory diseases, and identified the hepatic nuclear factor (HNF) family as controllers of a network of genes affecting GGT. Our results reinforce the disease associations of GGT and demonstrate that control by the GGT1 locus varies with age.

Associations among working conditions and behavioral risk factors : The Helsinki Health Study with International Comparisons

Adverse health behaviors as well as obesity are key risk factors for chronic diseases. Working conditions also contribute to health outcomes. It is possible that the effects of psychosocially strenuous working conditions and other work-related factors on health are, to some extent, explained by adverse behaviors. Previous studies about the associations between several working conditions and behavioral outcomes are, however, inconclusive. Moreover, the results are derived mostly from male populations, one national setting only, and with limited information about working conditions and behavioral risk factors. Thus, with an interest in employee health, this study was set to focus on behavioral risk factors among middle-aged employees. More specifically, the main aim was to shed light on the associations of various working conditions with health behaviors, weight gain, obesity, and symptoms of angina pectoris. In addition to national focus, international comparisons were included to test the associations across countries thereby aiming to produce a more comprehensive picture. Furthermore, a special emphasis was on gaining new evidence in these areas among women. The data derived from the Helsinki Health Study, and from collaborative partners at the Whitehall II Study, University College London, UK, and the Toyama University, Japan. In Helsinki, the postal questionnaires were mailed in 2000-2002 to employees of the City of Helsinki, aged 40 60 years (n=8960). The questionnaire data covered e.g., socio-economic indicators and working conditions such as Karasek s job demands and job control, work fatigue, working overtime, work-home interface, and social support. The outcome measures consisted of smoking, drinking, physical activity, food habits, weight gain, obesity, and symptoms of angina pectoris. The international cohorts included comparable data. Logistic regression analysis was used. The models were adjusted for potential confounders such as age, education, occupational class, and marital status subject to specific aims. The results showed that working conditions were mostly unassociated with health behaviors, albeit some associations were found. Low job strain was associated with healthy food habits and non-smoking among women in Helsinki. Work fatigue, in turn, was related to drinking among men and physical inactivity among women. Work fatigue and working overtime were associated with weight gain in Helsinki among both women and men. Finally, work fatigue, low job control, working overtime, and physically strenuous work were associated with symptoms of angina pectoris among women in Helsinki. Cross-country comparisons confirmed mostly non-existent associations. High job strain was associated with physical inactivity and smoking, and passive work with physical inactivity and less drinking. Working overtime, in turn, related to non-smoking and obesity. All these associations were, however, inconsistent between cohorts and genders. In conclusion, the associations of the studied working conditions with the behavioral risk factors lacked general patters, and were, overall, weak considering the prevalence of psychosocially strenuous work and overtime hours. Thus, based on this study, the health effects of working conditions are likely to be mediated by adverse behaviors only to a minor extent. The associations of work fatigue and working overtime with weight gain and symptoms of angina pectoris are, however, of potential importance to the subsequent health and work ability of employees.

Multiple socioeconomic circumstances and health : associations and explanations among Finnish and British public sector employees

Socioeconomic health inequalities have been widely documented, with a lower social position being associated with poorer physical and general health and higher mortality. For mental health the results have been more varied. However, the mechanisms by which the various dimensions of socioeconomic circumstances are associated with different domains of health are not yet fully understood. This is related to a lack of studies tackling the interrelations and pathways between multiple dimensions of socioeconomic circumstances and domains of health. In particular, evidence from comparative studies of populations from different national contexts that consider the complexity of the causes of socioeconomic health inequalities is needed. The aim of this study was to examine the associations of multiple socioeconomic circumstances with physical and mental health, more specifically physical functioning and common mental disorders. This was done in a comparative setting of two cohorts of white-collar public sector employees, one from Finland and one from Britain. The study also sought to find explanations for the observed associations between economic difficulties and health by analysing the contribution of health behaviours, living arrangements and work-family conflicts. The survey data were derived from the Finnish Helsinki Health Study baseline surveys in 2000-2002 among the City of Helsinki employees aged 40-60 years, and from the fifth phase of the London-based Whitehall II study (1997-9) which is a prospective study of civil servants aged 35-55 years at the time of recruitment. The data collection in the two countries was harmonised to safeguard maximal comparability. Physical functioning was measured with the Short Form (SF-36) physical component summary and common mental disorders with the General Health Questionnaire (GHQ-12). Socioeconomic circumstances were parental education, childhood economic difficulties, own education, occupational class, household income, housing tenure, and current economic difficulties. Further explanatory factors were health behaviours, living arrangements and work-family conflicts. The main statistical method used was logistic regression analysis. Analyses were conducted separately for the two sexes and two cohorts. Childhood and current economic difficulties were associated with poorer physical functioning and common mental disorders generally in both cohorts and sexes. Conventional dimensions of socioeconomic circumstances i.e. education, occupational class and income were associated with physical functioning and mediated each other’s effects, but in different ways in the two cohorts: education was more important in Helsinki and occupational class in London. The associations of economic difficulties with health were partly explained by work-family conflicts and other socioeconomic circumstances in both cohorts and sexes. In conclusion, this study on two country-specific cohorts confirms that different dimensions of socioeconomic circumstances are related but not interchangeable. They are also somewhat differently associated with physical and mental domains of health. In addition to conventionally measured dimensions of past and present socioeconomic circumstances, economic difficulties should be taken into account in studies and attempts to reduce health inequalities. Further explanatory factors, particularly conflicts between work and family, should also be considered when aiming to reduce inequalities and maintain the health of employees.