Study On Food Safety Incidents Based On Big Data

Currently,China has entered a period of frequent food safety incidents,inferior milk powder,cadmium rice,swill-cooked dirty oil and other major food safety incidents occur frequently.Emerge in endlessly the problem of food,not only seriously harm the health of consumers,but also affect the stability and development of social.With the rapid development of the news media and the growing popularity of the Internet,r and food safety incidents related to the massive news media reported by various news media and the rapid spread of the network.Processing and analysis of the huge amount of data,complex data structures in food safety news event data,traditional technology and methods is difficult to meet the demand.Big data and cloud computing technology,the rise and development,in-depth comprehensive analysis by exposure of food safety incidents news provides the possibility.With the help of Tianjin Food Safety&Low Carbon Manufacturing Collaborative Innovation Center,a new method based on big data and cloud computing is proposed to deal with the related news of food safety events in this paper.The method to "throw out the window" food safety incidents on the website to collect related news and events as the data source,the web crawler technology of Web data collector Octopus software based on acquisition related news and data on the site of food safety incidents and do some word segmentation and stop word processing data pre processing.Then the deployment on Hadoop cluster machine learning algorithms library in Mahout integrated algorithms of preprocessing the data serialization,to quantify the operation,through the coarse clustering on the algorithms library in the canopy clustering algorithm to find the optimal values of K and the initial cluster center,and then according to the results obtained using k-means clustering algorithm for secondary clustering and clustering description and Analysis on the final the clusters of food safety incidents news data on the site.The method proposed in this paper distributed text clustering experiments,"throw out the window" website on the number of food safety events in all kinds of food events distribution was obtained,which shows that the method is feasible and effective.