Vocabulary selection and organization for augmentative and alternative communication of children with speech impairment
Ano de defesa: | 2020 |
---|---|
Autor(a) principal: | |
Orientador(a): | |
Banca de defesa: | |
Tipo de documento: | Tese |
Tipo de acesso: | Acesso aberto |
Idioma: | eng |
Instituição de defesa: |
Universidade Federal de Pernambuco
UFPE Brasil Programa de Pos Graduacao em Ciencia da Computacao |
Programa de Pós-Graduação: |
Não Informado pela instituição
|
Departamento: |
Não Informado pela instituição
|
País: |
Não Informado pela instituição
|
Palavras-chave em Português: | |
Link de acesso: | https://repositorio.ufpe.br/handle/123456789/37884 |
Resumo: | Augmentative and Alternative Communication (AAC) systems are used to supplement or replace speech/writing of individuals with complex communication needs. These systems can be used for several applications beyond communication (e.g., inclusive education and serious games), and can be of two types: “low-tech” (i.e., non-technological systems based on pictures, objects, and communication books), or “high-tech” (i.e., technological systems such as speech-generating devices, and AAC software). To achieve good results, AAC systems must provide access to a large vocabulary, which is adequate to communication development and organized to facilitate the pictogram retrieval. Both, vocabulary selection and organization are essential tasks for successful AAC usage. To be easier, these tasks can be based on proposals of core word lists, and of vocabulary organization. However, there is no consensus among these proposals. In this thesis, we analyze the existing proposals of words and categories to propose a Core Vocabulary for children’s communication, which should also be organized to provide facilities for several high-tech AAC applications. In this context, we perform two Systematic Literature Review (SLR) to select, respectively, core word lists and category lists, and, based on these lists, we generate new core word lists using Natural Language Processing (NLP) techniques and new category lists using semantic clustering. Then, we conduct statistical analysis over corpora extracted from CHILDES Corpus Database to analyze the core word lists, and qualitative analysis with experts to analyze the category lists. Next, we organize the better results of words and categories into an ontology enriched with data from semantic databases (e.g., WordNet) and foundation ontologies (e.g., Ontolex). This thesis presents two main results: 1) a Core Vocabulary, formed by the best core word list organized in the categories that represent the essential concepts for children communication; and 2) an ontology (AACOnto), which organizes the core vocabulary and enriches it with syntactic and semantic data. Our Core Vocabulary supports vocabulary selection and organization tasks to implement a vocabulary on AAC systems that allows children to express themselves using different communicative functions. In addition, our ontology provides syntactic and semantic information about vocabulary instances, which allows the implementation of versatile high-tech AAC systems for several applications. |