Aquatic resource conservation - The first Yangtze finless porpoise successfully born in captivity

Background. As the sole freshwater subspecies of finless porpoise (Neophocaena phocaenoides), the Yangtze finless porpoise (N. p. asiaeorientalis) lives only in the middle and lower reaches of the Yangtze River and its appended Poyang and Dongting Lakes. As a result of human activity on the river, including over and illegal fishing, pollution, transportation and dam construction, the population of Yangtze finless porpoises has been steadily and rapidly decreasing during the past several decades, which leads the animal to be endangered. Methods. For saving this unique animal from extinction, three corresponding measures, in situ conservation, ex situ conservation, and intensifying breeding research in captivity, were proposed and have been implemented since the 1980s. Results. After successfully rearing the animals in captivity for almost nine years, the first Yangtze finless porpoise was successfully born in captivity on July 5, 2005. The calf is male, with a body length of 69 cm. This is the first freshwater cetacean ever born in captivity. Conclusion. The successful birth of this calf confirms that it is possible to breed the Yangtze finless porpoise in captivity. Furthermore, this will greatly benefit the conservation efforts, and also greatly bolster our on-going efforts to study the reproductive biology of these animals. Recommendation. More studies and efforts are expected to establish a sustainable, captive colony of the Yangtze finless porpoise, which will not only greatly expand our knowledge about the reproduction biology of this animal, but also help to redeem the wild population through a careful yearly 'soft releasing' process.

面向Blog的兴趣挖掘和推荐系统

Blog作为Web2.0的重要应用以其个性化的信息发布平台、多元化的内容载体等特点吸引网络用户。撰写和浏览Blog已经成为网络文化的流行热点,推动了Blog搜索服务的发展。目前的Blog搜索服务大都是基于对查询关键词的匹配来实现的,缺乏自动提取用户兴趣并进行推荐的能力。该文设计和实现了一个面向Blog的兴趣挖掘和推荐系统Blog-digger,该系统采用兴趣挖掘技术,能自动识别用户的兴趣,并主动推荐主题相关的Blog。实验结果证明了该系统的有效性。

基于CoP建模的协作过滤推荐方法

传统的协作过滤推荐方法主要基于个人兴趣特征来实现推荐。在组织内部协作场景下,为实现知识共享与重用,推荐系统不仅要考虑用户兴趣,还应考虑用户和用户组的任务。传统的协作过滤推荐方法已不能满足要求。CoP是组织内部人员管理的主要形式,它的特征是其成员任务的反映。基于已有的协作过滤推荐研究与D-S理论,提出了一种CoP特征构建算法,并以此为基础研究了面向CoP的协作过滤推荐。

基于特征提取的服装外形设计草图推荐技术

草图是服装外形设计中一类高效的信息载体,为了提高服装外形设计的效率,减少大量的重复操作,本文设计并实现了一种基于草图特征提取的二级推荐技术,在设计过程中整合草图轮廓的几何特征和草图勾画过程中的过程特征,同时结合相关反馈技术和用户交互历史数据,并结合笔式界面技术,开发了笔式服装外形设计原型,改善了人机交互方式,能够在用户设计过程中满足草图推荐的目的。

AN EXPERT SYSTEM FOR SOLVENT-EXTRACTION OF RARE-EARTHS

An expert system for solvent extraction of rare earths has been developed using LISP. The goal of this project was to mimic the chemists' inferential abilities to assist in the process of solvent extraction of rare earths. The system includes frequently used extractants, separation of specific rare earths, recommendation of procedures for the separation of mixtures of rare earths using (2-ethylhexyl)phosphonic acid 2-ethylhexyl monoester, selection of parameters for counter-current extraction and methods for evaluation of the technique, and the economics of the processing. The expert system runs on an IBM-PC/XT.