Traditional relevance feedback technique could help improve retrieval performance. It usually utilize the most frequent terms in the relevant documents to enrich the user’s initial query. We re-examine this method and find that many expansion terms identified in traditional approaches are indeed unrelated to the query and harmful to the retrieval. This paper introduces a Support Vector Machines Based method to improve the retrieval results. Firstly, the classifier is trained on the feedback documents. Then, we can utilize this classifier to classify the rest of the documents and move relevant documents to the front of irrelevant documents. This new approach avoids modifying the initial query, so it’s a new direction for the relevance feedback techniques. Our Experiments on TREC dataset demonstrate that retrieval effectiveness can be improved more than 24.37% when our proposed approach is used.

The collected massive data during the breeding process usually cost plenty of human power. Therefore, it is urgent to create highly-efficient specific breeding data analysis software to assist breeders to screen fine varieties. For this reason, this paper developed a breeding data management analysis system by taking MyEclipse as development tool, which covers various statistical analysis techniques. In the development process of each algorithm module, JNI technology is adopted to realize Java-called DLL and complete algorithm call, improving the execution efficiency of a large number of arithmetical operations. This paper introduced Java-called DLL by aid of JNI technology in One-Way ANOVA in details and the parameter passing between Java and C++ and how to realize Java-called DLL in the operating system without C++ operating environment. Through the above-mentioned call, the advantages of C++ and Java are comprehensively considered, which not only lightens the burden of Java virtual machine and avoids the repeatability, but also improves the efficiency of arithmetical operation and the utilization rate of codes.