Design And Implementation Of Clustering Ensemble Industrial Internet Method Based On Wisdom Of Crowds

With the popularization of computer and network technology,the information volume of today's society grows with the exponential situation.We have entered the era of information explosion,and there are huge amounts of data generated every moment,such as urban traffic data,commodity price data,and people's consumption.Data,Internet data,etc.,the potential value behind these data is very huge,then we need to use data mining technology to discover the hidden information and value of these data.Cluster analysis is a very important technical means in data mining technology.In this paper,the key technologies of clustering integration are studied,and a clustering integration method based on group intelligence is proposed.The specific work is summarized as follows:Firstly,this thesis deeply studies the clustering integration algorithm in the field of machine learning,and expounds the research status and related theoretical methods of clustering integration in detail,and analyzes the characteristics of each method and its shortcomings.Secondly,a clustering integration algorithm based on multi-link feature subsets is proposed.Firstly,clustering integration is carried out for multi-dimensional datasets.It is proposed to use independent feature subset selection algorithm,apply relevance to feature attribute selection,reduce dimension,improve the performance of clustering integration results,and then perform multi-link integration to get the final result.The experiment was carried out on the standard data set,and the method was validated and validated based on the experimental data.Thirdly,combined with the concept of group wisdom in social science,this paper proposes a cluster connection clustering algorithm based on group intelligence framework.Under the four criteria of meeting the group wisdom framework,the performance of clustering integration will be improved,and the experimental verification is carried out on the standard dataset.The experimental results show that the proposed algorithm can improve compared with other clustering integration algorithms.The performance of clustering integration results is superior.Fourthly,combined with service clustering,an intelligent service clustering algorithm based on group intelligence framework is proposed.Cluster integration and group intelligence are applied to service clustering.Through the climbed Web service dataset design experiment,the verification institute The feasibility and superiority of the proposed method show that compared with the existing service clustering algorithm,the intelligent service clustering algorithm based on the group intelligence framework can effectively improve the recall and purity of service clustering.