Learn More
In real world information systems, data analysis and processing are usually needed to be done in an on-line, self-adaptive way. In this respect, neural algorithms of incremental learning and constructive network models are of increased interest. In this paper we present a new algorithm of evolving self-organizing map (ESOM), which features fast one-pass(More)
In tackling data mining and pattern recognition tasks, finding a compact but effective set of features is often a crucial step in the whole problem solving process. In this paper we present an empirical study on feature selection for classical instrument recognition, using machine learning techniques to select and evaluate features extracted from a number(More)
A modularised connectionist model, based on the mixture of experts (ME) algorithm for time series prediction, is introduced. A group of connectionist modules learn to be local experts over some commonly appeared states in a time series. The dynamics for combining the experts is a hidden Markov process, in which the states of a time series are regarded as(More)
Decision making in a complex, dynamic environment is a very difficult task that requires new techniques of computational intelligence for building adaptive, hybrid intelligent decision support systems (HIDSS). Here, a new approach is proposed based on evolving agents in a dynamic environment. Neural network and rule-based agents are evolved using methods(More)
The Department of Information Science is one of six departments that make up the School of Business at the University of Otago. The department offers courses of study leading to a major in Information Science within the BCom, BA and BSc degrees. In addition to undergraduate teaching, the department is also strongly involved in postgraduate research(More)
  • Da Deng
  • Int. J. Comput. Syst. Signal
  • 2004
Content-based image retrieval techniques have been under intensively research, focusing on extracting effective low level visual features for indexing and enabling fast and accurate retrieval of individual images by matching the feature indexes. In this paper we propose to extend the content-based approach towards the problem of multimedia collection(More)
RATIONALE Lymphomas take up about 14% of all head-neck malignancies, out of which 97% are non-Hodgkin lymphomas (NHL). The clinical courses, treatment responses, and prognoses of NHLs vary with different subtypes and anatomic sites. In the Chinese population (including the Taiwanese), head-neck NHLs are often seen with the tonsils, nasal cavity, nasal(More)
In this paper we present a case study of co-training to image classification. We consider two scene classification tasks: indoors vs. outdoors and animals vs. sports. The results show that co-training with Naïve Bayes using 8-10 labelled examples obtained only 1.2-1.5% lower classification accuracy than Naïve Bayes trained on the full(More)