The Application Of Case-based Reasoning In Intelligent Vehicle Monitoring Data Processing System

Along with the rapid development of China's economy, the quantity of car ownership continues to rise. The rapid increase in car ownership add up lagging behind of urban transport construction, lead to the urban transportation deteriorating. Road traffic safety issues are becoming a serious threat to economic development and people's lives and property. Therefore, intelligent transportation system (ITS) as a new way to solve the traffic problems at home and abroad is developed rapidly. However, the system on the market today is basically just a vehicle has the function to track and monitor, most of them don't do further analysis for the vast amounts of data collected from the vehicle terminal, to solve this problem, and overcome the traditional limitations, this paper propose case-based reasoning processing system for road traffic accidents.With the help of vehicle information system terminal, do some work to the vast amounts of data to form a case base. On this basis, by means of the CBR mechanism of machine learning field, combined with information retrieval, text mining technology, to design a system to deal with road traffic accidents under the reasoning mechanism, which can offer the decision basis for road management department. The work of this paper is as follows:(1) The establishment of case base: combine application environment of the field, analysis the many factors collected through information collection systems, to establish a case base which can facilitate the CBR process.(2) Case-Based Reasoning: CBR accomplish through three sub-processes. Firstly feature identification, analysis the new problem and extract the relevant characteristics; secondly preliminary match, use the technology of text classification to find a class of candidate case which related to the current problem from the case base; thirdly is the best selection, use the vector space model method to select a couple of cases which was most similar with the current problem from the preliminary match result.(3) Case learning: by adding a new case, adjusting the weight of keywords via the feedback of user to accomplish the learning process.