A STUDY ABOUT THE INFORMATION RETRIEVAL SYSTEMS OF THE POLITICAL WORK BASED ON THE NEURAL NETWORK

摘要

As the political work has made many progresses in information retrieval technologies and several algorithms have been formed to verify the preliminary conclusions,the present study expects the result of the research in this paper to carry out an important role in political work of information technology and to play a positive role in promoting political work,.In this paper,we try to create mature political retrieval technologies,with state-of-the-art thinking.The development of the internet network has arisen a new type of problem for both users and makers: determining the relevancy of political work sources.In this paper,we propose a political work retrieval system employing relevance feedback retrieve documents.That employing of the back-propagation neural network learning mechanism as the political work retrieval knowledge encoding mechanism enables the system to handle a Query in a reasonably short time and return the searched results with high accuracy.Of course,this is only a preliminary exploration,some of the conclusions are not perfect enough,and some may not be optimal algorithm.We sincerely hope experts and scholars attend the center.