Font Size: a A A

Research And Application Of LBS Recommendation System For Converting Socialized Networks In Mobile Applications

Posted on:2018-05-13Degree:MasterType:Thesis
Country:ChinaCandidate:H Y ZhangFull Text:PDF
GTID:2348330536468746Subject:Engineering
Abstract/Summary:PDF Full Text Request
Now,smart phones have been very common,users using mobile Internet have accounted for the vast majority of all Internet users.And mobile applications are more and more rich,mobile application market can provide a variety of third-party software for users to use.Mobile e-commerce applications are also particularly popular,including Taobao,Meituan and other mobile phone applications,and they greatly meet the user’s purchase of goods and services needs.Because smart phone has the disadvantages of the screen shows limited,calculation and storage restrictions,they can not make mobile users can see the information or services with their own satisfaction,that is,mobile smart phones also exist in the traditional Internet information overload problem,and the problem in the smart phone side has been further amplified.How to solve this problem,as the traditional Internet products,using recommended systems eliminate a large number of information bring the user to choose the difficulties,in time provide users with the information and services they need.Of course,smart phones also have their own advantages,and they can be anywhere to obtain the user’s location information,and the user’s location information is indeed very important information.Because it is possible to determine the position of the user based on the location information,we can find the merchant around the user,and determine the weather of the user’s current location based on the location information,and the information is about the user’s consumption of the merchant and the merchant.In addition to the smart phone to obtain the user’s location,it can also get to the user’s communication information,which includes the user and who made a number of text messages,users and who played the phone and so on,and this information can also for our recommended system to provide the original data support,effective to prevent the system cold start problem.Based on the research on the existing achievements and techniques of LBS system and recommendation system,this paper finds out some specialities based on geographic location service,and this special place is the best point of combination with recommendation system.Therefore,information service recommendation system.We through the grasp of the public comment on the network of businesses and users of the data analysis and mining,build the user’s long-term preference model,which is the user’s long-term business preferences model.And the user’s short-term preferences,that is,the user’s current situation to make the preferences of choice,we introduced here the concept of situational information,in order to be based on different scenarios to determine the user’s choice of differentiated behavior.And the acquisition of the scene information is based on the advantages of smart phones to get,such as the user’s current location can use GPS and base station positioning to obtain the user’s current location weather data can be based on the current latitude and longitude to the weather website,the season can be based on the time of the mobile phone system to get to.Finally,in order to effectively solve the problem of cold start system,we joined the concept of social networking,according to the smart phone-side communication data to build the user’s social network,and then select the mobile users from the network of neighbors,registered users to provide friends like the business or service.We also designed a convergence of the social network of LBS business recommendation system demonstration program.Through the implementation of the above recommended system framework,from the forecast accuracy,system sensitivity and diversity and other indicators to evaluate our recommendation system,which in the above indicators of experimental results analysis,the ideal to meet our own of the many requirements of the recommended system.Of course,there may be some of our algorithms,hope that in the follow-up work we can continue to improve.
Keywords/Search Tags:Mobile Applications, LBS, Scenario information, Social relations network, Long-term preference
PDF Full Text Request
Related items