Research On Multi-attribute Ranking Strategy Based On Feedback Weights In Internet Of Things

With the development of sensing equipment and mobile communication technology, the application of mobile Internet of Things has been increasingly valued by all sectors of the community. It has been widely used in industrial agricultural production, intelligent transportation, intelligent home, logistics monitoring and other fields.The Internet of Things search services also came into being in this context and mobile applications become the key of Internet of things services. In this context,the sorting algorithm as the core of the Internet of Things search application is the main content of this paper. Based on the traditional sorting algorithm of the Internet of Things, this paper proposes a multi-attribute sorting strategy with the feedback weight, and simulates this sorting strategy, which proves the validity of the sorting strategy. The main contents and main achievements of this paper are as follows:(1) This paper researched the three-tier classification of the Internet of Things and proposed a distributed two-tier service architecture including geography and type domain, which is concise and suitable for large-scale Service deployment. Based on the traditional sensor sorting strategy, this study introduced the multi-type and multi-dimension context-aware information, including user context awareness and sensor perception information, which can be used to accurately understand the search scene and finally got the optimal ranking result under different situations. The multi-attribute ranking strategy in this paper includes attribute clustering, fast selection and accurate ranking model. This ranking strategy can sort the multi-dimensional attributes quickly and efficiently.(2) In order to solve the problem of users' lack of knowledge in the field of Internet of Things and the subjective assumption of input weight,a dynamic attribute weighting model based on user's feedback was proposed. The feedback model can comprehensively analyze the results of feedback and dynamically adjust the weight of attributes. It has wide applicability and can solve the problem of attribute weight change in different situations.(3) In this paper, the scheduling strategy was simulated and tested and the multi-attribute sorting strategy based on feedback weight was designed and implemented. The experimental results show that the algorithm can sort a large number of elements with multi-dimensional attributes in a given search context and then dynamically adjust the results with user's feedback to achieve the optimal sorting results step by step.