Design And Development Of Data Knowledge Resource Management System Based On Cloud Service Platform

Nowadays,digital working mode with information technology as the core is becoming more and more popular.Enterprises are also following the trend to constantly innovate.However,the traditional way of resource management has the problems of inconvenient sharing,inconvenient consulting,inconvenient management and easy loss,so the demand of online intelligent knowledge resource management is emerging.This thesis designs a data knowledge resource management system based on cloud service platform,which not only solves the problems of traditional management,but also ensures the internal sharing of enterprise knowledge.It provides a convenient and effective resource management platform.The main research content of the dissertation is divided into the recommendation of resources,the implementation of search engine,the recognition of handwritten notes,and synchronous message feedback.The details of these four aspects are as follows:(1)To solve the problem of inconvenient sharing,this article combines user-based collaborative filtering recommendation algorithm and content-based recommendation algorithm to realize the recommendation of interest resources.First,according to the logs of users and the corresponding attention degree,the resource information that users may be interested in is calculated,and then the content-based recommendation algorithm is concluded to complete the recommendation.The user-based collaborative filtering recommendation algorithm uses the method called Pearson similarity,which makes the prediction result more accurate than traditional calculation methods.The content-based collaborative algorithm adopts recommendation principle with higher correlation of the same type of resources.Upload resources occupy corresponding attention,which can effectively solve the problem of "cold start of items".In addition,according to the recommendation algorithm based on popularity,the recommendation service of hot resources is provided.(2)To solve the problem of inconvenient consulting,this thesis takes the system knowledge resources as the data source,and realizes the search engine based on the cloud service platform.Through the data processing and distribution platform Apache Ni Fi,the database resource information and corresponding documents are obtained,transcoded and pushed to the cloud service platform.In the cloud platform,the search server Elasticsearch is used to build the index.It uses the document extraction plugin,Ingest-Attachment,to preprocess the text,and IK Analyzer to specify the type of word segmentation in the search field.Han language processing package does stop word processing for user input.Finally,full text retrieval is realized.(3)Aiming at the difficulty of management,this dissertation adds recognition of handwritten notes to help users manage personal resources.Baidu character recognition processing technology is adopted in this article.First,baidu test paper and analysis interface is used to locate the table area in the note image,cut and save,and record the relevant information of the table.After that,baidu handwritten character recognition interface is used to recognize the note images and process the recognition results.When the result is outside the table area,the result of character recognition is written into the result document.When the result is in a table area,baidu table character recognition is used and the recognition result is written into the document.The recognition of multi images needs to use the template to record the historical content.Apache POI is used for document processing.(4)In order to solve the problem that the manager can not convey the instructions in time,this dissertation uses Web Socket communication protocol to realize the realtime reminder of announcement message,that is,synchronous message feedback.It is convenient for message transmission between client and server.Heartbeat detection and reconnection mechanism make announcement notification data push to online users in time.Through the research of the above key content,this dissertation realizes an intelligent knowledge resource management system.It can effectively avoid the loss of resources.