Design And Development Of Network Security Autonomous Learning Platform

With the rapid development and progress of Internet technology,human beings have gradually entered the era of Internet.With the continuous integration of the Internet and human life,human beings rely more and more on the networked way of life.However,the related network security problems emerge in endlessly,which have more and more impact on the whole society.Therefore,the training of network security talents is very urgent.Meanwhile,the "Internet plus" initiative has attracted wide attention.The Internet and education are deeply integrated in different degrees and different aspects,which makes more and more educational resources go to the Internet.There are many web learning platforms similar to MOOC.The emergence of network learning platform enables learners to learn anytime and anywhere,which makes learning more convenient and rich in learning resources.To sum up,it is of great significance to design and develop an online learning platform for network security and train network security talents through network learning.However,at present,most online learning platforms are only limited to the sharing and use of learning materials.The more learning materials,the larger the data,for beginners without clue,to find a suitable learning route and related learning materials is undoubtedly a very severe test.This kind of "blind people touch the elephant" learning method is difficult to achieve good learning results,which will eventually lead to the loss of learning enthusiasm and interest of learners.In view of the above problems,this thesis introduces the personalized recommendation mechanism into the online learning platform of network security,through obtaining the learning basis and learning preference of learners,and designs a recommendation algorithm to recommend the learning route and related learning materials suitable for each learner,so as to build an autonomous platform of network security based on personalized recommendation.First,the calculation method of user similarity is studied,a new weight coefficient is proposed,and an "improved modified cosine similarity" is proposed.Based on this improved modified cosine similarity,"anew user based collaborative filtering recommendation algorithm" is further proposed(A Novel User-based Collaborative Filtering Algorithm,NUCF for short);Second,based on the collaborative filtering algorithm NUCF,a network security autonomous learning platform is designed and developed.The platform recommends learning routes and materials for learners through NUCF algorithm,and gradually guides learners to carry out the learning of related network security knowledge,so as to increase learners' interest in learning and improve their learning efficiency.The main research work of this thesis is as follows:(1)A new user based collaborative filtering recommendation algorithm,NUCF,is proposed.The traditional modified cosine similarity only considers the items that users evaluate together.Once the data is sparse,it will affect the calculation accuracy of user similarity.To solve the above problems,this thesis improves the traditional modified cosine similarity,and proposes a new weight coefficient weight.Through combining the advantages of modified cosine similarity,Jaccard coefficient and weight coefficient,an improved modified cosine similarity is obtained.Based on this improved modified cosine similarity,a new user based collaborative filtering recommendation algorithm,NUCF,is proposed.Experiments on the MovieLens data set show that the NUCF algorithm can achieve very good recommendation performance.(2)According to the principle of software engineering,this thesis analyzes the requirement of network security autonomous learning platform.First,the overall goal of network security autonomous learning platform is proposed;Second,we analyze the feasibility of learning platform,including: technical feasibility,economic feasibility,and legal feasibility;Finally,the functional requirements and nonfunctional requirements of learning platform are analyzed.(3)On the basis of requirement analysis,we design the network security autonomous learning platform in detail.Through the functional module design,the whole platform is divided into multiple sub-modules with low coupling and high cohesion.We analyze the design method of each sub-module in detail.The whole platform is divided into four main modules: personalized learning recommendation module,main learning module,information management module,and user management module.Different modules cooperate with each other,which makes the running and operation of our platform simple and efficient,thus reducing the unnecessary finding and search time of users,and achieving efficient learning.In addition,the table structure in the database is designed in detail.(4)Based on the analysis and design of system,the network security autonomous learning platform is further implemented.Based on the development language Python,web application framework Flask and MySQL database,the development of whole platform is completed.The platform adopts the NUCF algorithm proposed in(1)to implement the personalized recommendation function,and provides autonomous learning services with multiple specific topics.Users can select one or several specific topics for autonomous learning,and the platform will recommend learning routes,books and technical posts to the users,under the specific topic.In addition,we have carried out a detailed functional test to the platform,thus ensuring the correctness of the platform.