Research On Personalized Retrieval And Recommendation Methods Of Judgement

With the rapid development of information technology in the Internet era,the informationization degree of the Judicial Platforms is getting higher and higher.With the construction of "Sunshine Justice" and "Smart Justice",electronic judgement documents have been published efficiently and timely.One of the most important issues ofj udgement disclosure platform is personalized retrieval and recommendation of judgement,which intends to retrieve and recommend the most valuable judgements to the users.Information retrieval and recommendation are effective methods to solve the information overload problem.Therefore,in this context,our research on personalized retrieval and recommendation methods of judgement is a topic with fatal research significance and application value.On the basis of summarizing the relevant research of document retrieval and recommendation from home and abroad,we make a thorough study on personalized retrieval and recommendation problems of judgement.This paper has a certain research purpose and significance in theory,method and application level.We propose a retrieval method of judgement based on ILSA(Improved-Latent Semantic Analysis)model,a personalized recommendation method of judgement based on the user profile.In the practice aspect,a specific example of personalized judgement retrieval and recommendation for "ZG Judgements Online" has been given.It provides a beneficial reference method to judgement retrieval and recommendation.This paper carries on a thorough research to the personalized retrieval and recommendation methods of judgement,the main research contents are as follows.(1)The problem description and research framework of personalized judgement retrieval and recommendation have been proposed.Firstly,we give the problem description of the judgment retrieval and recommendation,and refine the research problem from the perspective of practical application background.Then,the relevant mathematical notations are explained on the basis of the description of the problem,and the corresponding research framework has been constructed.The work mentioned above has laid a foundation to the subsequent work.(2)The personalized retrieval method of judgement based on I-LSA model has been proposed.First,the methods of acquiring and preprocessing judgement are given by analyzing the contents and characteristics of judgement documents.Then,the methods of preprocessing users’ queries and forming users’ query profile are proposed on the basis of analyzing the characteristics of personalized retrieval of judgement.Moreover,we present how to determine candidate judgement documents based on users’ query profile.Finally,the judgement retrieval method based on I-LS A model is explained.(3)The personalized recommendation method of judgement based on the users’ interest profile has been proposed.Firstly,a judgement network is established according the law-citation relationship between judgement documents.And the user’s interest profile is generated by analyzing the judgement network.Secondly,based on the analysis of the characteristics of the judgement documents,the methods of calculating the content similarity score,the authority score and the popularity score of candidat e judgement are demonstrated,and then we sort the candidate judgement documents according the final score.(4)An application study has been presented.Our application on "ZG Judgements Online"shows that the proposed method is practical and feasible.