Research On Ptimization Of Flowing Platform For Producing Vector Data Based On Analysing Users’ Behavior

The producing platform for vector data is an important platform for producing and editing vector data.With the rapid development of computer technology and internet technology,The producing platform for vector data provides functions,gradually from the basic functions,the "big and complete" functions,public oriented software platform to the provision of "special" functiosn,personalized and intelligent functions.The "user centric" thought used to optimize platform is an effective way to achieve this transformation.At the same time,the operating mode for producing vector data has changed from workshop to process.Therefore,this paper analyzes the importance of fast producing vector data that have high quality,and researchs status of users’ behavior analysis.Combine application of the producing platform for vector data with process – PLA that developed by Chinese Science Institute of earth remote sensing and digital(Point Line Area,referred to as PLA,the full name is the producing platform for vector data with process used to produce point,line and area),through the analysis collected users’ behavior and attribute data on the PLA platform to optimize the interface of PLA and achieve intelligent recommended mechanism for tasks,providing intelligent and personalized servicee for users and special function,so as to improve the efficiency and quality of producing vector data by users.this paper have several aspects of the work:(1)Analysis the importance of the producing platform for vector data with process,and the users’ behavior play the role in optimizing platform,then summarizes the current research situation and existing problems of analysising users’ behavior and optimizing software platform,put forward the research’s content and purpose of this paper;(2)Summarize the producing model of vector data on PLA platform,and analyze the problems of PLA,and propose the solutions based on analysising users’ behavior,the data of users’ behavior is classified and modeled,and different types of behavioral data are obtained,the data of users’ behavior is pre-processed,and stored in the relational database;(3)Aim at the existing problems in the interface of PLA,the interface is optimized by different analysis methods for users’ behavior are proposed to recommend and combine tools,the feasibility and practicability of the two methods are verified on the PLA platform,(4)Analyzes the disadvantages of the existing model for allocating task on PLA,and proposes an intelligent mechanism for recommending,according to the difficulty of task and the data of users’ behavior,respectively according to the historical behavior of users and new users(no or less history data on the PLA platform),put forward the methods for tasks that producing vector data by two step filtration and improved decision tree,the effectiveness of intelligent recommending method is verified,The average quality of vector data is improved;(5)Based on PLA and the experimental environment are described,and the conceptual design and logical design of the experimental system are introduced in detail,combined with the basic statistical analysis method,the analysis method for users’ behavior and software technology,the system for analysing users’ behavior based on PLA is realized.