| With the development of Internet and the increasing of information, more and more uselessinformation causes great pressure on people in their life and people have to bear the burden ofinformation. As we know, it will be difficult to get the useful information than before with thesituation of information overload. To address the problems as list, many researchers start to focuson recommender that could recommend people useful information that they expect to obtain.In addition, social network has recently become more and more popular in people’s life suchas Facebook, Twitter and Blog. Users can make friends with others from direct or indirectconnection. With the development of social network, many people start to realize the value ofsocial network as the increasing the users.In recent years, lots of recommender systems are used by many companies but there are stillmany challenges with the expansion of the recommender system. The main research works in thedissertation include: the design of recommendation algorithm, the design of system, recommendersystem leveraging deep learning algorithm and social network information. The main study andcontributions in this dissertation are given as follows:①Study of the recommender systems based on social network and current situation of intelligentrecommender systems. Analyze the common problems that recommender systems have.②List some recommender system example on Android platform and compare the advantagesand disadvantages of some recommendation algorithms in mobile environment.③Discuss the feasibility of implementing the recommender system leveraging deep learn theoryand social network information.④Design and implement recommender system components, which include (1) the data miningmodule,(2) the recommendation engine module,(3) the feedback module.⑤Evaluate the system that I designed and verify the feasibility of the recommendation systemmodel. After analyzing the prototype, the paper summarizes the points and reaches theconclusions. |
PDF Full Text Request