为解决在信息过滤（ｉｎｆｏｒｍ ａｔｉｏｎ ｆｉｌｔｅｒｉｎｇ）中表达用户的信息需求困难的问题，提出了一种新方法——固定文章集法（ｆｉｘｅｄ ｄｏｃｕｍ ｅｎｔｓｅｔ， Ｆ Ｄ Ｓ）。在这种方法中，用户可以通过评价一些专门选择的文章来表达自己的信息需求。这避免了用户选择关键词的困难，扩大了用户信息需求表达的空间，同时可以应用于处理多语种信息过滤和社会化过滤问题。选择固定文章集使用选择文章基因法（ｓｅｌｅｃｔｄｏｃｕｍ ｅｎｔ＇ｓｇｅｎｅ， Ｓ Ｄ Ｇ），实验证明， Ｓ Ｄ Ｇ 的性能较其它方法有明显优势 In order to solve the problem that it is difficult to express user’s information in inform ation filtering, a new method named fixed docum entset (FDS) is proposed. In this approach, users can express their information needs by evaluating some specially selected articles. This avoids the user’s difficulty in selecting keywords and expands the space for expressing user information needs, and can also be applied to deal with multilingual information filtering and social filtering issues. The selection of a fixed essay using select article gene method (selectdocum ent’sgene, S D G), experiments show that the performance of S D G has obvious advantages over other methods