数据驱动的量化信任评估模型及其在农业物联网中的应用

发布时间：2018-07-15 13:21

【摘要】：近年来,食品安全事故的频频发生致使食品安全问题得到了社会各界越来越多的关注。农业物联网的提出实现了农产品从生产到销售信息的全程追溯,从而使得消费者能够便捷地获取农产品在整个供应链中的大量真实信息,从而提高食品的安全性。然而,这也导致了消费者在挑选产品时需要面对大量产品信息。面对这些信息,让一位普通消费者在短时间内分析这些信息并且判断出该产品的可信度是困难的。本论文设计了一种面向农业物联网的信任评估模型,可以用来解决这一问题。该模型由数据驱动,评估过程不涉及人工干预,能分析大量数据信息并计算出一个量化数值。在应用到农业物联网中时,该模型从时间、地点、签名、供应链、电子化覆盖率以及商家声誉六个方面对产品的信任度进行考虑,计算出一个在[0,1]范围内的数值作为产品信任度的参考,从而帮助消费者方便快捷地挑选出值得信任的农产品。其次,论文结合农产品电子履历系统对该模型进行了系统开发。我们从农产品电子履历系统中自动获取待评估产品的所有电子履历信息并对其进行数据分析,在提取所需数据后通过预设公式计算出每一个影响因子的值,然后利用贝叶斯网络计算出待评估产品的信任度。论文以一个应用实例验证了我们提出方法的可行性。最后,我们通过实验研究了各个影响因子对信任度的影响以及先验概率对信任度的影响,并且将本模型与其他方法以及模型进行比较,说明了本模型在应用于农业物联网时的优势。
[Abstract]:In recent years, the frequent occurrence of food safety accidents has caused more and more attention to food safety problems. The proposal of agricultural Internet of things (IOT) realizes the tracing of agricultural product information from production to sale, which makes it easy for consumers to obtain a large amount of real information of agricultural products in the whole supply chain, thus improving the safety of food. However, this also causes consumers to face a lot of product information when selecting products. In the face of this information, it is difficult for an average consumer to analyze the information and determine the reliability of the product in a short period of time. In this paper, a trust assessment model for the agricultural Internet of things is designed, which can be used to solve this problem. The model is data-driven and does not involve human intervention. It can analyze a large amount of data and calculate a quantitative value. When applied to the agricultural Internet of things, the model takes into account the trust of the product in terms of time, location, signature, supply chain, electronic coverage and reputation. Calculate a value in the range of [0] as a reference to product trust, thus helping consumers to quickly and easily select trustworthy agricultural products. Secondly, this model is systematically developed in combination with the agricultural product electronic resume system. We automatically acquire and analyze all the electronic resume information of the product to be evaluated from the agricultural product electronic resume system. After extracting the required data, we calculate the value of each influence factor through the preset formula. Then the trust degree of the product to be evaluated is calculated by using Bayesian network. An application example is given to verify the feasibility of the proposed method. Finally, we study the influence of various factors on trust and the influence of priori probability on trust through experiments, and compare this model with other methods and models. The advantages of this model in agricultural Internet of things are explained.
【学位授予单位】：复旦大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP391.44;TN929.5

【相似文献】