Text and position ranking algorithm based on sample weighted

To effectively solve results ranking of Meta- Search Engine problem, a text and position ranking algorithm based on sample weighted is proposed. On the full consideration of the structural information, the PageRank score is transformed into weight. Combined with text information and its position in the result list, adjustment of the local similarity is implemented, and relevant score of the result position is standardized. The algorithm presents two definitions of entry matching degree and entry relevancy. The experimental results illustrate that this algorithm is feasible and efficient.