翻译：In this thesis, we use latent semantic indexing (LSI, an information retrieval technique) [7] to retrieve the relationship among proteins from protein interaction networks. LSI can solve synonymy (many ways to express a given concept) and polysemy (words have multiple meanings) problems in information retrieval (IR). This property is also useful for solving our problem. Our experimental results show that our method indeed overcomes the problems about predicting protein functions using protein interaction networks.

我们在这篇论文中使用潜在语义索引（LSI，一种信息检索技术）[7]从蛋白质互作网络中检索蛋白质之间的关系。 LSI可以解决信息检索（IR）中的同义词（表达给定概念的许多方式）和多义词（单词有多种含义）问题。 这个属性对于解决我们的问题也很有用。 我们的实验结果表明，我们的方法确实克服了使用蛋白质互作网络预测蛋白质功能的问题。