Text clustering to help knowledge acquisition from documents

Lapalut, Stéphane

doi:10.1007/3-540-61273-4_8

Cited by 13 publications

(7 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The combined text of responses from the 59 subjects was also examined using ALCESTE software [Analyse des Lexèmes Coocurrents dans un Ensemble de Segments de Textes, or Analysis of Co-occurring Lexemes in a Set of Text Segments (Image, 2007; Lapalut, 1995)]. This software uses an algorithm to construct an overview of patterns found in the text according to how often roots and word forms appear together (eliminating from the analysis common modifiers such as articles and prepositions).…”

Section: Methodsmentioning

confidence: 99%

Doing time: A qualitative study of long-term incarceration and the impact of mental illness

Yang

Kadouri

Révah-Lévy

et al. 2009

International Journal of Law and Psychiatry

View full text Add to dashboard Cite

Once convicted, the perpetrator of serious crime embarks upon a new journey: the challenge of adjusting to long-term imprisonment. Prisoners' views of incarceration and the meaning of this experience may affect their later adjustment to life in the community. On the basis of brief narrative responses collected during an epidemiological survey of the psychological health of prisoners in France, this study examined the impact of incarceration on psychological state in a group of 59 inmates serving long sentences. Qualitative content analysis and computer-assisted linguistic analysis (using ALCESTE software) were performed on the textual data of open responses to three standard questions. Using a combination of these two approaches, seven categories of the subjective experience of prisoners in the sample were identified: the Outside World, Others, Punishment, Time, Affects and Impulses, Self-Concept, and Speech. Further qualitative analyses were then performed to compare the responses of Severely Mentally Ill (SMI) subjects and subjects with no psychiatric disorder. These analyses revealed contrasting attitudes towards incarceration. SMI subjects spoke in more hostile and persecutory terms about their experience in prison, attributing suffering to external circumstances, while subjects with no psychiatric disorder evoked similar themes, but with an introspective attitude. The themes evoked by mentally ill individuals in our sample suggest that their reactions to the prison environment arise in part from aspects of their psychiatric symptoms, and this may have relevance to future mental health policy and practices in criminal corrections.

show abstract

Section: Methodsmentioning

confidence: 99%

Doing time: A qualitative study of long-term incarceration and the impact of mental illness

Yang

Kadouri

Révah-Lévy

et al. 2009

International Journal of Law and Psychiatry

View full text Add to dashboard Cite

show abstract

“…Com base nas frequências dos termos nas UCE, o algoritmo de análise realiza divisões sequenciais do conjunto de UCE em subgrupos, maximizando a diferença entre eles, o que é verificado por testes de qui-quadrado (Lapalut, 1995;Dransfield, Morrot, Martin & Ngapo, 2004). A Classificação Hierárquica Descendente implica uma primeira divisão do corpus em dois grandes conjuntos de UCE que compartilham vocabulário o menos possível, que foram denominados "Eixos" neste artigo e que serão retomados mais adiante.…”

Section: Método De Análiseunclassified

“…A estabilidade das classes é verificada pela realização de duas análises independentes com base em diferentes tamanhos de UCE. A Classificação Hierárquica Descendente é descrita com mais detalhes por exemplo em Lapalut (1995) e Dransfield et al (2004).…”

Section: Método De Análiseunclassified

Polifasia cognitiva nas representações sociais do alcoolismo

Souza

Menandro

2015

psi.saber.soc

View full text Add to dashboard Cite

RESUMO:Diferentes racionalidades agem na construção de representações sociais, por meio do processo de polifasia cognitiva. Parte-se da hipótese de que profissionais de saúde não se restringem ao saber técnico-científico ao construir suas representações. Esta pesquisa teve o objetivo de compreender as representações sociais do alcoolismo construídas por profissionais de Saúde da Família, em especial a dimensão da polifasia cognitiva. Participaram 40 profissionais, recrutados igualmente quanto a gênero e profissão (médicos e não-médicos). Eles responderam a entrevistas semiestruturadas e os dados foram analisados com Classificação Hierárquica Descendente com auxílio do software ALCESTE. Os resultados mostram que o alcoolismo foi representado simultaneamente como 1) doença multifatorial a ser tratada de forma integral: as causas são conhecidas e a equipe de saúde deve adotar procedimentos técnicos multiprofissionais; e 2) como problema social, relacionado à pobreza das comunidades, a ser tratado com os (poucos) recursos disponíveis e com força de vontade: as causas são obscuras, o alcoolista não recebe cuidados, há motivações contraditórias entre usuários e profissionais, o tratamento depende da força de vontade e de recursos alheios. Discute-se a presença de crenças que associam o alcoolismo à pobreza. Os profissionais recorreram implicitamente aos universos reificado e consensual para a construção das representações sociais, caracterizando a incidência da polifasia cognitiva. São realizadas sugestões metodológicas para uso do ALCESTE: um quadro para interpretação das Classes de trechos de discurso geradas com auxílio do software e o uso dos conceitos de universo reificado e universo consensual para interpretação de seus gráficos (dendrogramas).Palavras-chave: representação social; alcoolismo; atenção primária à saúde; polifasia cognitiva; ALCESTE.ABSTRACT: Different rationalities participate in the construction of social representations through the process of cognitive polyphasia. We considered the hypothesis that health professionals do not use only technical-scientific knowledge to construct their representations. This research aimed at understanding the social representations of alcoholism constructed by Brazilian Family Health professionals, especially the dimension of cognitive polyphasia. Forty health professionals equally sampled by gender and profession (physicians and non-physicians) participated in this research. They answered to semi-structured interviews. We analyzed the data with the Top-down Hierarchical Classification implemented in the software ALCESTE. Results show that professionals simultaneously represented alcoholism as: 1) a multifactorial disease to be treated with a comprehensive approach, its causes are known, the health team must apply multi-professional procedures. 2) A social problem related to the poverty of communities, to be treated with the (few) available resources and with willpower. Its causes are obscure, the alcoholic patient does not receive health care, there are contradict...

show abstract

“…The iterative algorithm continues on each subset until further sub classes are not significantly different (Lapalut, 1995). The stability of these classes is checked by performing two independent analyses with different length EUCs and comparing statistically the results of the occurrences of EUC in the two classifications.…”

Section: Statistical Analysesmentioning

confidence: 99%

“…The objective of the analysis was to determine the Ôcontext' behind the enunciations (Lapalut, 1995) and was derived by grouping the enunciations according to the occurrence of the words within them. Analyses were performed by the software ÔALCESTE 4.5' (Analyse Lexicale par Contexte of a document considered as an Ensemble de Segments de TExte.…”

Section: Statistical Analysesmentioning

confidence: 99%