The objective of the research presented in this thesis was to systematically investigate the computational structure for cross-database emotion recognition. The research consisted of evaluating the stability of acoustic features, particularly the glottal and Teager Energy based features, and investigating three normalization methods and two data fusion techniques. One of the challenges of cross-database training and testing is accounting for the potential variation in the types of emotions expressed as well as the recording conditions. In an attempt to alleviate the impact of these types of variations, three normalization methods on the acoustic data were studied. Motivated by the lack of large and diverse enough emotional database to train the classifier, using multiple databases to train posed another challenge: data fusion. This thesis proposed two data fusion techniques, pre-classification SDS and post-classification ROVER to study the issue. Using the glottal, TEO and TECC features, of which the stability of emotion distinguishing ability has been highlighted on multiple databases, the systematic computational structure proposed in this thesis could improve the performance of cross-database binary-emotion recognition by up to 23% for neutral vs. emotional and 10% for positive vs. negative.

Submitted by Aelson Maciera (aelsoncm@terra.com.br) on 2017-04-26T20:25:16Z No. of bitstreams: 1 DissTMS.pdf: 4210213 bytes, checksum: de15a2cacf15c1179f470589ca57169b (MD5) / Approved for entry into archive by Ronildo Prado (ronisp@ufscar.br) on 2017-05-02T20:20:12Z (GMT) No. of bitstreams: 1 DissTMS.pdf: 4210213 bytes, checksum: de15a2cacf15c1179f470589ca57169b (MD5) / Approved for entry into archive by Ronildo Prado (ronisp@ufscar.br) on 2017-05-02T20:20:25Z (GMT) No. of bitstreams: 1 DissTMS.pdf: 4210213 bytes, checksum: de15a2cacf15c1179f470589ca57169b (MD5) / Made available in DSpace on 2017-05-15T18:02:44Z (GMT). No. of bitstreams: 1 DissTMS.pdf: 4210213 bytes, checksum: de15a2cacf15c1179f470589ca57169b (MD5) Previous issue date: 2012-02-28 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Patent document is a technological information source, internationally standardized and of great relevance to increase the business potential of competitiveness and innovation. The present research analyses the patent databases Derwent Innovations Index e Espacenet in order to contribute to studies in competitive intelligence through patent analysis related to nanotechnology applications in cosmetics and personal care products. Derwent Innovations Index and Espacenet databases are web accessible databases, with international coverage in all fields of technology. Thus, this study intends to elucidate the features and resources available in each of those databases and contribute to the optimization of the information search and retrieval for patent documents analysis. Based on the proposal, the research is exploratory and descriptive and quantitative and qualitative approach, which aims identify the features offered by each databases and then elaborate patents analysis using softwares and resources for this purpose. As results, Derwent Innovations Index and Espacenet are different in some aspects and those two databases can be used alone or in a complementary manner depending on the objectives, and ultimately, concludes that Derwent Innovations Index is the most suitable database according to the criteria evaluated. / O documento de patente é uma fonte de informação tecnológica padronizada internacionalmente e de grande relevância para o aumento do potencial competitivo e de inovação das empresas. Esta pesquisa analisa as bases de dados de patentes Derwent Innovations Index e Espacenet com o intuito de contribuir com estudos de inteligência competitiva por meio da análise de patentes sobre aplicações da nanotecnologia em cosméticos e produtos de higiene pessoal. A Derwent Innovations Index e a Espacenet são bases de dados acessíveis por meio eletrônico, com abrangência internacional e cobertura de todos os campos tecnológicos. Assim, o presente estudo visa delinear as características e recursos disponíveis em cada uma das bases de acordo com critérios de avaliação adotados e contribuir com a otimização da busca e recuperação da informação para análise dos documentos de patentes. Com base no objetivo proposto, a pesquisa é exploratória e descritiva e com abordagem quantitativa e qualitativa, no qual visa identificar os recursos oferecidos por cada uma das bases de dados e, em etapa posterior, elaborar análises de patentes com a utilização de softwares e recursos para esta finalidade. Com base no desenvolvimento desta pesquisa, pode-se afirmar que a Derwent Innovations Index e a Espacenet são bases de dados diferentes em alguns aspectos e estas duas bases de dados podem ser utilizadas de maneira complementar ou isoladamente dependendo dos objetivos e, por fim conclui-se que a a Derwent Innovations Index é a base de dados a mais indicada de acordo com os critérios avaliados.

The Master’s thesis deals with the topic of the use of artificial inteligence for managerial decision making in the firm. This thesis contains an aplication of fuzzy logic system for firm’s supplier evaluation that provides informations for more efficient collaboration with suppliers.