Data Storage And Context Mining Based On Symbian Smartphone

With the improvement of the mobile phone's computing power and integration, the number of applications on it is increasing very fast, and as the mobile internet technology has also got significant development, which makes people can access the internet anytime and anywhere. So the mobile phones appear in more scenarios, and the users have more reliance on mobile phones. Therefore, the data generated in the process of using mobile phones can reflect the user's behavior accurately. So how to analysis these data and then mining the users'behavior pattern are of significance for mobile phone manufactures and service providers. This system aims to design and implemente a storage solution for storing data about the phone's state and sensor, and propose an efficient mining algorithm to mining context pattern from this data, and then analysis the user's behavior.This paper first introduces the background and significance of data storage and data mining on mobile platform. Then it introduces Symbian operating system which is related theories briefly and describes some basic knowledge about the development on Symbian OS to lay a solid foundation for the implementation of the system. In the third chapter, the significance of data mining is described briefly, and a detailed exposition of the concepts of data mining for the description of the mining algorithm which will be detailed in chapter five. Followed by Chapter three, the analysis overall system architecture and module of the project is detailed. The next chapter elaborates the algorithm proposed in this paper for mining context pattern according to the specific features of the data generated on the mobile platform. After the architecture of the system and the core algorithm has been determined, this paper expounds the implementation of this system in detail, including the data storage module and the mining plug in. After the implementation of all of them, a test about the solution of data storage, the efficiency and accuracy of the mining algorithm is carried on, and which has good results. Finally, this paper also looks forward to the work on the next prospect.The significance of data storage and context mining system is to propose an efficient and flexible storage solution to store the context data and propose an algorithm which has good performance to mining context pattern from these data. And the implementation of the system follows good software engineering practice and makes full use of the advantages of Symbian OS, which allows the system has good performance overall, the relationships between modules are clear and low coupling.