一种新的基于多启发式的特征选择算法

在查询扩展方法中，如果通过查询结果中关键词的上下文来计算候选关键词的权重，将权重大的词作为查询扩展词，其候选关键词来源于文档中关键词的上下文，这种方法存在主题漂移的问题。为了解决这个问题，提出一种将初始查询结果过滤，只选择与源文档语境相似的搜索结果，来帮助选择查询扩展词的方法。实验结果表明该方法能获得更合适的查询扩展词。

New feature selection algorithm based on multiple heuristics

In order to enhance the operation speed,reduce the memory space occupied and filter out irrelevant or low-degree relevant features,feature selection algorithm must be used.One novel feature selection algorithm named as selection method using multiple heuristic rules was presented.The algorithm comprehensively considered features,so features could be selected more effectively and accurately.Experimental results show that the algorithm is better than mutual information,chi-square statistic and document freque...