| With the development of Chinese economy and the growth of the personal disposable income,the travel consumption of Chinese people has been increased rapidly including the overseas-travel consumption.As the result of the asymmetric information problem caused by the unfamiliar shopping environment,the desire for the overseas-travel shopping information service has also been increased.On the other hand,the popularity of mobile devices has provided a solid hardware foundation for an overseas-travel shopping system.Under such background,the design and implementation of an overseas-travel shopping system has both its practical significance and technical support.As for the system,the main purpose is to provide accurate and reliable shopping information,and the most essential part is the back-end data service.This thesis will focus on it and use multiple technics including data crawler,notification-pushing,etc.The main purpose of the thesis is to provide the shopping information for overseas-travel consumers through data digging,collecting,matching and notification-pushing.The data will be provided to front-end mobiles from back-end data service system through API gateway.The thesis elaborates the research background at the beginning,then analyses the specific requirement,and expounds the design and implementation of the system from three aspects of crawler,notification-pushing and mobile API gateway.In the end of the thesis,the system has been tested,deployed and run.The thesis offers a full solution of the data service and has completed the following researching tasks:1.Use web crawler to collect data automatically.Python language has been used to create the clawer to dig and collect the public related information.And the information has been extracted,classified and stored based on specific data structures.Also a specific strategy has been designed to solve the anti-crawler mechanism.2.Design a location-based information recommendation module with a high concurrency and high reliable notification-pushing function.This module uses integrated locating system of the mobile OS.This module decouples the generating and sending of notification into two phases.Notification generating is achieved by an SOA service which is exposed to the recommendation module.These two phases are asynchronous to avoid the potential problems and the hardware requirement caused by the synchronized high-concurrency operation.The thesis discusses the notificationpushing module in both Android and i OS systems.Different strategies have been made according the two systems.3.Use the API gateway to solve the problems during the mobile service development,such as the high coupling problem with the back-end service,the difficulty to upgrade rapidly and the lack of unified security protection.The biggest innovation is to use the hot deployment which provides both reliability and iteration.The usage of script language has not only insured the smooth operation of essential functions,but also provided a non-stop release opportunity while the system keeps operating.Besides,the downgrade and fallback strategy has also been considered and designed in case system crashes.4.Test and deploy the whole system including the crawler,notification-pushing and mobile API gateway.The crawler is designed with Python,the others use Java and Golang to fit different scenes.Each module has passed the operation test and been doubled checked on high-concurrency and sustainability.The research creatively enables the back-end data service system to dig,collect and push information based on the comprehensive application of multiple technologies,including the quick locating and notification-pushing technology based on GPS and Wi-Fi,the automatic digging technology of crawler and a high reliable gateway service with high performance.However,the whole solution is only a basic one for back-end data service.The problems may occur in the real online environment.So it still needs to be optimized and updated continuously after the system is brought online. |
PDF Full Text Request