Research And Application Of Judgments Recommendation Algorithm Based On Multimodal Features

With the development of big data and artificial intelligence technology,the applicationof artificial intelligence technology to the legal field has attracted more and more attention.Using the existing information of judgments to push similar cases has become an important means to unify the judgment standard,promote judicial justice,and standardize sentencing.However,the existing judgment recommendation systems can not effectively use the multi-modal information of the judgment,ignoring the complementary advantages of different modal features,resulting in the unsatisfactory recommendation effect.In view of the above problems,this paper proposes a judgment recommendation algorithm based on multi-modal feature fusion.First,pre-process the judgments and extract the multi-modal features of the judgments respectively.Secondly,in order to achieve the multi-modal feature fusion,this paper proposes a multi-modal depth fusion method based on depth neural network,which combines the classification cross entropy loss,to obtain high-order fusion features of judgments.Finally,we use the learned fusion features to calculate the similarity between the judgments,and the Top-N with the highest similarity to the user.On this basis,based on the B / S architecture,using Vue.js,Spring boot,My SQL and other technologies,this paper designs and implements the recommendation system of judgment based on multi-modal features.According to the needs of users,the design and implementation of four modules are completed: data preprocessing module,judgment management module,judgment recommendation module and user information management module.The main work of this thesis is to take into account the complementary characteristics of the multi-modal feature of the judgment and implement the judgment recommendation.After testing,the platform has reached the design requirements.In addition,compared with the existing judgment recommendation system,the proposed judgment recommendation algorithm based on multi-modal feature fusion can effectively integrate the multi-modal features of the judgment and make the recommendation result more accurate,which is of great significance to avoid the phenomenon of "different judgments of similar cases" and to promote the application of artificial intelligence in the legal field.