Research On The Constructing Of Food Quality Analysis Model In The Internet Of Things Environment

The food industry plays an indispensable role in people’s livelihood. With thegrowing of food industry, the quality and safety of food attract more attention thanever before. However, there are many factors that influence food product, which arechangeable and uncontrollable; the food industry’s regulatory system has been in thedownstream of all areas. Common methods to identify food quality is observing colorand tasting taste by people. It’s obviously lack of scientific theories and easily affectedby human factors.This paper is the food quality analysis module of a science and technology projectcalled "food quality control and traceability of things demonstration platform", whichis also a municipal science and technology development program of Changchun. Mainwork of this paper is creating a body model by using the ontology modeling methodfood production line for the Dingfengzhen food company, in order to make a foodquality analysis. Traditional neural network can not be achieved due to the specialnature of the field of food. Latent Semantic ontology reasoning is able to discover thefood field, and automated analysis of food quality.In this paper, upper body concept was firstly, built in the field of food availablefor different applications of the field of food to share. We analysis of various factorsand abandon which uncontrollable factors, such as clues to the food production chain,purchase of raw materials, storage, material processing, and the factors that impact onfood quality to the finished goods storage and sale of the entire food productionenterprises within the food processing process, the establishment of food qualityanalysis underlying concept system of the body. We then add data attributes andrelationship attributes and the hierarchical relationship of the body to establish foodquality analysis. We add all this factors to the body with the help of experts instance,analysis of experiments to determine the range of the data attributes, then establishrules of inference and reasoning protégé which comes from Pallet inference engine thatfirstly established ontology reasoning verification body and appears contradictoryconcepts and examples, reasoning and make corrections. This article uses the protégé ontology modeling tools in SWRL food qualityanalysis ontology reasoning rules, Jess ontology inference engine, use Jena2in, theinference engine to retrieve the ontology model. OWL rules can not be added to foodquality analysis of the weight in the analysis, the reasoning requirements is high, sothe SWRL establish the rules of inference. Conversion from OWL ontologies andSWRL rules to Jess Protégé JessSWRL Tab plug-in can be achieved, so this articletacit knowledge Jess inference engine ontology reasoning get written back to the bodyin the same JessSWRL Tab plug. This article to use Jena2ontology retrieval the goodcompatibility Jena2Java program and its strong body retrieval ability.The analysis in food quality and the body foundation build rely on the layer in theregulatory platform Dingfengzhen food quality food quality analysis model, the modelconsists of information layer, quality analysis layer, food quality analysis ontologylayer. Information acquisition layer is used for food production information, qualityanalysis layer is used to analyze the quality of the food, and the food quality analysisontology analysis provide the basis for the food quality.The food quality analysis model provide a reliable scientific data to supportanalysis model for food quality and food quality control and traceability.Jena2of javagood support, makes this model can be well compatible with the actual food qualitysupervision retrospective platform to provide a reliable basis for food quality appraisal.Given the main users of the ontology model established large-scale food productionenterprises which abandoned factors that can not be obtained by the automatedmonitoring, leading to food quality analysis comprehensive analysis, the model is onlyadvisory.