Cássia Trojahn dos Santos

Integrating knowledge through cooperative negotiation: a case study in bioinformatics

Data Mining techniques have been used for knowledge extraction from large volumes of data. A recent practice is to combine Data Mining and Multi-Agent Systems approaches. In this paper we propose the use of cooperative negotiation to construct an integrated and coherent domain model from several sources. Agents encapsule different symbolic machine learning algorithms to induce their individual models. After this, a global model yields from the interaction via cooperative negotiation of these agents. The results shows that the proposed approach improves the accuracy of the individual models, integrating the best representations of each one.

A multi-agent approach to question answering

In this paper we present a multi-agent approach to question answering for the Portuguese language. Our proposal is composed by three modules: (1) document and query processing; (2) ontology construction; and (3) answer generation. Each module is composed by multiple cooperative agents which adopt distinct strategies to generate its outputs and cooperate to create a global result. This approach allows the use of different strategies and the reduction of errors introduced by individual methods. The cooperation among the agents aims to reach better solutions in each step of the processing.

Automatic Classification of Enzyme Family in Protein Annotation

Most of the tasks in genome annotation can be at least partially automated. Since this annotation is time-consuming, facilitating some parts of the process --- thus freeing the specialist to carry out more valuable tasks --- has been the motivation of many tools and annotation environments. In particular, annotation of protein function can benefit from knowledge about enzymatic processes. The use of sequence homology alone is not a good approach to derive this knowledge when there are only a few homologues of the sequence to be annotated. The alternative is to use motifs. This paper uses a symbolic machine learning approach to derive rules for the classification of enzymes according to the Enzyme Commission (EC). Our results show that, for the top class, the average global classification error is 3.13%. Our technique also produces a set of rules relating structural to functional information, which is important to understand the protein tridimensional structure and determine its biological function.

Selection of data sets of motifs as attributes in the process of automating the annotation of proteins' keywords

Automatic annotation tools are becoming popular since the biologists and curators of databases cannot cope with the volume of sequences to be annotated manually. One way to automate the annotation is to use techniques of symbolic machine learning to derive rules to guide this annotation. However, the training instances tend to have too many attributes, turning the machine learning process difficult and time consuming.