Semiparametric Preference Learning

Unlike traditional supervised learning problems,preference learning learns from data available in the form of pairwise preference relations between instances.Existing preference learning methods are either parametric or nonparametric in nature.We propose in this paper a semiparametric preference learning model,abbreviated as SPPL,with the aim of combining the strengths of the parametric and nonparametric approaches.SPPL uses multiple Gaussian processes which are linearly coupled to determine the preference relations between instances.SPPL is more powerful than previous models while keeping the computational complexity low （linear in the number of distinct instances）.We devise an efficient algorithm for model learning.Empirical studies have been conducted on two real-world data sets showing that SPPL outperforms related preference learning methods.     

利用糠醛废渣生料栽培平菇

糖醛废渣是污染环境的三废之一。为变废为宝,作者取废渣进行分析化验,并研究出糖醛废渣的中和剂及PH缓冲剂等,从理论上说明了生料栽培平菇的可行性,同时用一年多的实验,总结出最佳配方及操作方法,最后归纳了用糖醛废渣生产平菇的优点及发展前景和推广价值。     

Adjuvanticity and delayed-type hypersensitivity by Bacillus Calmette-Guérin (BCG)

Résumé Le BCG, lavé à l'eau distillée, puis à l'etherethanol, puis au chloroforme, est traité par l'anhydride acétique. Le bacille ainsi traité par acétylation perd, à poids égal, par rapport au bacille vivant, de son pouvoir d'induire une hypersensibilité retardée spécifique, mais accroit son pouvoir de stimulation non spécifique des réactions immunitaires.

---

---

Acknowledgments. The authors wish to thank ProfessorG. Mathé for his interest in this investigation and MrsF. Kosloff from Pasteur Institute, Paris, France, for her generous gift of BCG.     

A multiple intelligent agent system for credit risk prediction via an optimization of localized generalization error with diversity

Company bankruptcies cost billions of dollars in losses to banks each year. Thus credit risk prediction is a critical part of a bank＇s loan approval decision process. Traditional financial models for credit risk prediction are no longer adequate for describing today＇s complex relationship between the financial health and potential bankruptcy of a company. In this work, a multiple classifier system （embedded in a multiple intelligent agent system） is proposed to predict the financial health of a company. In our model, each individual agent （classifier） makes a prediction on the likelihood of credit risk based on only partial information of the company. Each of the agents is an expert, but has limited knowledge （represented by features） about the company. The decisions of all agents are combined together to form a final credit risk prediction. Experiments show that our model out-performs other existing methods using the benchmarking Compustat American Corporations dataset.     

MATHEMATICAL MEASURES OF SHAPE CONFORMITY

This paper has developed a single mathematical measure for the evaluation of the shape conformity of the curved garment objects such as moulded brassiere cups. The measurements are based on the sectional analysis of the fringe pattern generated by means of moire topography. They include the area ratio, shape ratio and root mean square (RMS) of the height deviation of the moulded cup from the standard shape profile. Ultimately, the RMS was found to be a satisfactory method of assessing the shape conformity and a guideline for determining a good fit from the RMS value was then derived.     

Mycobacterial adjuvant and its carrier

Résumé Une fraction peptidoglycolipidique peut être extraite à partir du BCG par décapage acide dans un mélange de solvants fortement lipophiles. Cette fraction appelée ?LF? peut reproduire l'activité adjuvante du BCG entier dans les réactions immunitaires à médiation cellulaire et humorale. L'activité adjuvante du BCG appara?t comme liée à ce composant actif, LF, car après son extraction, le résidu bacillaire (BB) est dépourvu d'activité adjuvante. Néanmoins lorsque BB est utilisé pour adsorber (?véhiculer?) LF, le complexe BB-LF s'est révélé le meilleur adjuvant.

---

---

This work was performed at the Laboratory of Experimental Hematology, Department of Zoology, the University of Nebraska-Lincoln, Lincoln, Nebraska 68508, USA.     

Ultrastructural correlation of water reabsorption in isolated rat cauda epididymidis

Summary Electron microscopic study was made on the water reabsorption of the epithelial cells of the rat cauda epididymidis. It was shown that when the epididymal duct was reabsorbing water at a maximal rate, widely dilated intercellular spaces were seen. It is suggested that the standing gradient model of water reabsorption first proposed for the gall bladder may also operate in the cauda epididymidis.This work was supported in part by the World Health Organization and grant number 158/254 of Hong Kong University.     

Evaluation of Woven Fabrics for Comfort Properties in Children’s Summer Wear

This paper evaluates the comfort properties of eightkinds of woven fabric commonly used for children'ssummer wear in Hong Kong,China.Due to the expan-sion of children's wear market and the quality require-ment for parents,objective and subjective measurementsare the main methods used to evaluate the chosen fabricsfor garment deslgners and consumers as reference. Objective measurement was carried out through ex-periments by means of the Kawabata Evaluation Systemfor Fabrics(KES-F).Experimental tests on fabrics in-clude compression,surface,bending,air permeability,thermal and tensile.Subjective measurement was con-ducted by a market research on consumers' opinions onfabrics for children's summer wear in the form of ques-tionnaire survey.These two ways of measurement onfabrics were analyzed and compared. It was found that Fabric Sample 6 which is com-posed of polyester and cotton fibre has the best perfor-mance in terms of its comfort properties.Besides,Sam-ple 4 composes of rayon and cotton fibre also ha