The Design And Implementation Of A Personal Note Management System Based On Text Recommendation

Informationization has been an important topic in the modern society.In such a time,the information people can obtain is severely exploding.Therefore,how to store and organize the information we get has been a problem that must be solved.Electronic note management system has been developed and upgraded for a long time,which benefits people’s work and study.Because of its simplicity and practicability,Markdown language has been adapted by more and more note management system.Meanwhile,a more convenient way to sort notes is also an expectation of users.In view of the above situation,the paper designs a complete web application using MVC pattern,conforming to the RESTful API specification.It has complete user authentication system,supports the editing and preview of notes written in Markdown language,and features notes sorting based on text similarity calculation.Firstly makes the general system design,uses different frontend technics for specific use cases.Django’s built-in template is used in the simple user page.Uses single page application technics based on React.js in the notes editing page in order to handle the complex communication between frontend and backend.Uses nature language process technics to calculate the similarity between texts.Segment the texts into isolated words,uses bag-of-words model to extract the feature vector,then use the cosine of two vector as the similarity value.In order to solve the performance limitation,the system uses timed task to compare notes in the database,designs a way of measuring similarity value between note and notebook,and implements the corresponding backend APIs.After a set of functional test and performance test,the system has been proved that it nearly archives the original goal,can be run in mainstream browsers.