Our primary aim is to develop a classifier system that is capable of predicting the success or failure of hip prostheses on the basis of data from early radiological observations. The data set we employ (collected at The Royal London Hospital) records observations taken in the early years following fixation of the prosthesis and failure or otherwise after ten years. Many of the records contained in this data set have missing values. Recent work on the well-known Pima Indian data set has demonstrated the effectiveness of the Naive-Bayes (NB) method, coupled with boosting, on data with missing values. In this paper we investigate the performance of the NB method and boosting on the hip prosthesis data which contains a much greater proportion of missing values than the Pima Indian data. Our data set is additionally challenging in that it contains many more examples of one class (success) than the other.

In continuous speech recognition, the co-pronunciation between two successive phonemes seriously disturbs the recognition effect. It is difficult for pure hidden Markov model (HMM) methods to cope with co-pronunciation, because HMM methods consider that two successive frames of speech are independant. The hybrid HMM and artificial neural network (ANN) methods with feedback multilayer perceptron (MLP) (Bourlard and Wellekens, 1990; Bourlard and Morgan 1994) provide the ability to cope with co-pronunciation by means of feedback input. In this paper, we propose a new feedback method for feedback hybrid HMM/ANN methods on the basis of the original methods. The new feedback method provides more information of co-pronunciation to the feedback ANN. From HMM/ANN with feedback double MLP structure, we discuss the method that reduces the computation of the feedback MLP during recognition.

To compare to the pure hidden Markov model (HMM) methods, the hybrid artificial neural networks/hidden Markov models (ANN/HMM) methods(see [2], [3], [6]) have many advantages for speech recognition in theory. But, in practice, the hybrid ANN/HMM methods don't perform obviously better than HMM methods. In order to make use of the advantages of the hybrid ANN/HMM methods, in this paper, we proposes new feedforward and feedback hybrid methods that have double MLP structure on the basis of the original feedforward and feedback Hybrid methods. To compare to the original hybrid methods, new hybrid methods can more precisely estimate word model's posterior probabilities. In our experiments, new feedback hybrid method and new feedforward hybrid method fall error rate 42.6% and 25.8%, individually. New hybrid methods can perform Viterbi alignment more precise. Therefore, the training can converges farther.

A perfect (t, w, v; m)-threshold scheme is a type of combinatorial design that provides a way of distributing partial information (chosen from a set of v points called shadows) to w participants, so that any t of them can easily calculate one of m possible keys, but no subset of fewer than t participants can determine any partial information regarding the key. In this paper, we give a survey of recent constructions for perfect (t, w, v; m)-threshold schemes. In particular, we update results concerning perfect (3, 3, v; m)-threshold schemes.