Eduardo F. A. Silva

  • Citations Per Year
Learn More
This paper proposes a new method for binarization of digital documents. The proposed approach performs binarization by using a heuristic algorithm with two different thresholds and the combination of the thresholded images. The method is suitable for binarization of complex background document images. In experiments, it obtained better results than(More)
Information Extraction (IE) aims to extract from textual documents only the relevant data required by the user. In this paper, we propose a hybrid machine learning approach for IE on semi-structured texts that combines conventional text classification techniques and Hidden Markov Models (HMM). In this approach, a text classifier technique generates an(More)
This paper deals with automatic recognition of real bank checks. A new approach is proposed to read the numerical amount field from bank checks, considering the numeric value and the different delimiters that might exist in that field. The proposal combines different neural networks classifiers to perform the recognition. Experimental results have shown(More)
Geochemical mapping is the base knowledge to identify the regions of the planet with critical contents of potentially toxic elements from either natural or anthropogenic sources. Sediments, soils and waters are the vehicles which link the inorganic environment to life through the supply of essential macro and micro nutrients. The chemical composition of(More)
The Sn-W Panasqueira mine, in activity since the mid-1890s, is one of the most important economic deposits in the world. Arsenopyrite is the main mineral present as well as rejected waste sulphide. The long history is testified by the presence of a huge amount of tailings, which release considerable quantities of heavy metal(loid)s into the environment.(More)
The fast growth of electronic text collections (in particular, the Web) and the diversity of available documents immensely increased the difficulty to retrieve relevant documents in an efficient way. A variety of Web search engines have been built to help users in this task. These systems, however, lack precision in the retrieved documents. Different(More)
In this paper, we propose a hybrid machine learning approach to Information Extraction by combining conventional text classification techniques and Hidden Markov Models (HMM). A text classifier generates a (locally optimal) initial output, which is refined by an HMM, providing a globally optimal classification. The proposed approach was evaluated in two(More)
For increasing time values, isochrons can be regarded as expanding wavefronts and their perpendicular lines as the associated orthogonal isochron rays. The speed of the isochron movement depends on the medium velocity and the source-receiver position. We introduce the term equivalent-velocity to refer to the speed of isochron movement. In the particular(More)
Information extraction (IE) aims to extract from textual documents only the fragments which correspond to datafields required by the user. In this paper, we present new experiments evaluating a hybrid machine learning approach for IE that combines text classifiers and hidden Markov models (HMM). In this approach, a text classifier technique generates an(More)