Automatic Bayesian Methods In Text Retrieval

本文提出了一种基于贝叶斯方法的新模型，它与G.索尔顿理论细节已经制定出来。理想情况下，文档必须通过它们所引用的“真实的”对象来索引，这些真实的对象成为称为特异性网络的多层次系统中的节点。每个层次结构都是由一个特殊性运算符产生的，并产生一个对象树，从最一般的根开始，随着向叶子的发展，对象树会移动到更特殊的位置。填充节点的对象由文本术语或短语表示。任何一个物体都可以有多种表现形式。如果每个文档都必须手工转换为适合表示其中讨论的对象的形式，则所描述的模型是劳动密集型的。因此，我们正在开发对象表示的自动提取方法。这将导致一个易于处理的任务来表示文档。我们的主要工作是开发机器学习方法，帮助构建这里描述的层次结构。该项目目前的工作重点是开发一个改进的贝叶斯分类模型，并开发新的方法来主动学习贝叶斯模型。

A new model based on the Bayesian approach has been developed which has interesting connections with the vector models of G. Salton. Theoretical details have been worked out. Ideally documents must be indexed by the "real" objects that they refer to and these real objects become nodes in a system of multiple hierarchies called a specificity network. Each hierarchy is produced by a specificity operator and results in a tree of objects starting at the root with the most general and moving to greater specificity as one progresses towards the leaves. The objects which populate nodes are represented by textual terms or phrases. There may be many representations of any single object. The model described is labor intensive to construct if each document must be converted by hand to a form suitable to represent the objects discussed within it. Thus we are developing methods of automatic extraction of object representations. This will lead to a tractable task to represent documents. Our major effort is to develop machine learning methods that can aid in the construction of the hierarchies described here. Current work on the project is focusing on developing an improved Bayesian classification model and developing new approaches to active learning with a Bayesian model.