Made available in DSpace on 2016-04-28T18:22:33Z (GMT). No. of bitstreams: 1 Evelin Tesser.pdf: 787983 bytes, checksum: 1af38de98de5d393f5e4c3c8cc749be2 (MD5) Previous issue date: 2012-08-10 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / A double desire motivated me to develop this work: to study in depth the reflections about dialogue described in my master`s thesis and to talk about a case which memories return by each reading about transference. Regarding dialogue, this thesis rejects any rapprochement between communication or intersubjectivity and dialogue, assumed here as a guide in an insoluble dissymmetry (non-coincidence) among speakers. The dialogue allows the relationship missing (not complete) among different speakers: the other is always the other - there will never be "communion", "union" or "communication" among the speakers. The dialogue does not hide the differences; these differences anchor the dialogue, while it supports them. Thus, speaking about clinical dialogue is to take advantage of uncertainty and unpredictability. It is said that no one knows what the patient will say until he says. This opposes the idea that to have dialogue, they need to get the "common knowledge". In the clinical dialogue with patients who present symptomatic speech disorders, which comes as a stronger experience is a fracture in the illusion of control about speech - symptomatic speech disorders, as Lier-DeVitto (2011) reported, these disorders brought "the truth of non-coincidence" to light . Indeterminacy and unpredictability of speech indicate, therefore, a specific notion of subject: "non-coincidence with himself or herself" approaches us to the hypothesis of the unconscious, the subject introduced by Freud in Psychoanalysis. In this thesis, the psychoanalysis will be present. Dialogue, in this case, involves love - it is affected by the transference, concept that will be discussed from the reading of three psychoanalysts (Freud, Lacan and Pommier) to then it will be discussed in the Language Clinic. The case of Mr. Eurico, propellant of this debate, is going to appear in the final chapter to drive the discussion about transference in the Language Clinic, as theoretically constructed within the Research Group CNPq "Aquisição, Patologias e Clínica de Linguagem", directed by Maria Francisca Lier-DeVitto and Lucia Arantes, in-PUCSP Lael / Este trabalho nasceu de um duplo desejo: aprofundar reflexões feitas em minha dissertação de mestrado sobre o diálogo e poder falar sobre um atendimento cujas lembranças retornavam a cada leitura sobre transferência. No que se refere ao diálogo, esta tese recusa qualquer aproximação entre comunicação ou intersubjetividade e diálogo, aqui assumido como pautado numa irremediável dissimetria (não coincidência) entre falantes. O diálogo permite a relação faltosa (não toda) entre diferentes: o outro será sempre o outro nunca haverá comunhão , união ou comunicação entre falantes. Isso não significa, contudo, que inexista diálogo. O diálogo não tampona diferenças nelas se ancora, ao mesmo tempo que as sustenta. Sendo assim, falar em diálogo clínico é tomar partido da indeterminação e da imprevisibilidade. É afirmar que não se sabe o que o paciente vai dizer até que ele diga. Opõe-se à ideia de que, para haver diálogo, seja preciso chegar ao conhecimento comum . No diálogo clínico com pacientes com falas sintomáticas, o que vem como experiência mais forte é uma fratura na ilusão de controle sobre a fala falas sintomáticas, como conta Lier-DeVitto (2011), retiram da sombra a verdade da não-coincidência . Indeterminação e imprevisibilidade de falas remetem, portanto, a uma noção específica de sujeito: não-coincidência consigo próprio nos aproxima da hipótese do inconsciente, do sujeito introduzido por Freud na Psicanálise. Nesta tese, a Psicanálise fará presença. Diálogo, aí, envolve amor ele está em causa na transferência, conceito que será discutido a partir da leitura de três psicanalistas (Freud, Lacan e Pommier) para, em seguida, ser discutido na Clínica de Linguagem. O atendimento de Sr. Eurico, propulsor desse debate, aparecerá no último capítulo para movimentar a discussão sobre transferência na Clínica de Linguagem, conforme teoricamente construída no âmbito do Grupo de Pesquisa CNPq Aquisição, Patologias e Clínica de Linguagem, liderado por Maria Francisca Lier-DeVitto e Lúcia Arantes, no LAEL-PUCSP.

An approach to biomedical language processing is to apply existing natural language processing (NLP) solutions to biomedical texts. Often, existing NLP solutions are less successful in the biomedical domain relative to their non-biomedical domain performance (e.g., relative to newspaper text). Biomedical NLP is likely best served by methods, information and tools that account for its particular challenges. In this thesis, I describe an NLP system specifically engineered for sentinel event extraction from clinical documents. The NLP system's design accounts for several biomedical NLP challenges. The specific contributions are as follows. - Biomedical tokenizers differ, lack consensus over output tokens and are difficult to extend. I developed an extensible tokenizer, providing a tokenizer design pattern and implementation guidelines. It evaluated as equivalent to a leading biomedical tokenizer (MedPost). - Biomedical part-of-speech (POS) taggers are often trained on non-biomedical corpora and applied to biomedical corpora. This results in a decrease in tagging accuracy. I built a token centric POS tagger, TcT, that is more accurate than three existing POS taggers (mxpost, TnT and Brill) when trained on a non-biomedical corpus and evaluated on biomedical corpora. TcT achieves this increase in tagging accuracy by ignoring previously assigned POS tags and restricting the tagger's scope to the current token, previous token and following token. - Two parsers, MST and Malt, have been evaluated using perfect POS tag input. Given that perfect input is unlikely in biomedical NLP tasks, I evaluated these two parsers on imperfect POS tag input and compared their results. MST was most affected by imperfectly POS tagged biomedical text. I attributed MST's drop in performance to verbs and adjectives where MST had more potential for performance loss than Malt. I attributed Malt's resilience to POS tagging errors to its use of a rich feature set and a local scope in decision making. - Previous automated clinical coding (ACC) research focuses on mapping narrative phrases to terminological descriptions (e.g., concept descriptions). These methods make little or no use of the additional semantic information available through topology. I developed a token-based ACC approach that encodes tokens and manipulates token-level encodings by mapping linguistic structures to topological operations in SNOMED CT. My ACC method recalled most concepts given their descriptions and performed significantly better than MetaMap. I extended my contributions for the purpose of sentinel event extraction from clinical letters. The extensions account for negation in text, use medication brand names during ACC and model (coarse) temporal information. My software system's performance is similar to state-of-the-art results. Given all of the above, my thesis is a blueprint for building a biomedical NLP system. Furthermore, my contributions likely apply to NLP systems in general. / Graduate