A CROWDSOUCED SYSTEM FOR CITY ISSUES

The discovery and processing of city issues are important parts in urban governance，It is a development direction of urban informatization and intelligence to improve the precision of issue discovery and optimize the process flow. According to the city issue management process in the past, both the discovery and process of city issues are implemented manually, the distribution and process efficiency is low. Meanwhile, as the beneficiaries of urban management achievement, the advantages of citizens haven’t been better utilized, although they are the main component of a city. Therefore, how to integrate public participation and internet information into urban management, to build mutual urban management mode, to assign and process issues automatically, are the main research of this paper. The objective of this paper is to integrate information extracted from social networks and issues discovered by citizens; to classify issues in three ways: automatically, semi-automatically and manually; to ensure that issues are assigned quickly and accurately to the terminal personnel to deal with; to realize the transferring between different participants of urban management. Based on the combination of mobile Internet and traditional Internet, this paper designs and implements the City Issue Analytical System Based on Crowdsourcing, including the following work: (1) After researching on urban management mode based on crowdsourcing, city event discovery, processing mode and definition of urban management related roles are proposed. On this basis, requirements analysis of the city issue analytical system is make in order to determine the functional requirements, core business and workflow systems. (2) Researching on the classification and decision model of city issue, issues from social network and crowdsourcing are classified. The issues are quickly and accurately assigned to their belonging terminal department and personnel. The system uses vector classification algorithm based on KLD to classify the city issues. After comparing with several other classification algorithms, this algorithm shows higher accuracy and precsion in classification of city issue. (3) The design of City Issue Analytical System. In accordance with the idea of software engineering design, the overall architecture, functional module, database, classes and packages are designed. (4) Implementation and testing. Combining with the state transition diagrams, system business process diagrams and system interface to elaborate the implementation of each function. Finally, system testing is performed on the functions and performance of the system.