Algorithm Research about Textual Case Retrieval Based on Topic Words

Several shortages of Boolean retrieval, such as ignorance of the semantic relations among words and inability to rank the retrieval results in order of importance, are found by analyzing the essence of traditional text retrieval, in view of which an improvement of algorithm optimization based on topic words is proposed. Through enriching topic words to structure keywords library, the semantic distance and similarity of keywords are calculated on the basis of semantic retrieval framework. The improved algorithm is applied in the disaster case retrieval system at last, which retrieval results are then analyzed to detect performance. It is observed that the improved algorithm has a better improvement in retrieval both in precision rate and recall rate.