Research On Data Mining Service Model Of Big Data Alliance

With the rapid development of social networks,sensor technologies,and mobile Internet technologies,and the amount of data storage has reached the PB level.Become an important strategic resource for enterprises and even the country.Driven by the active promotion of national policies,big data service platforms continue to emerge,laying a good foundation for meeting the needs of resources in the "data ocean" and mining the potential value behind the data.However,at present,the big data service market is still in its infancy,and its development is not yet complete.So,when faced with the diverse data mining service needs of users,a single enterprise still has deficiencies in service capabilities.To this end,relying on the big data industry chain,with the purpose of improving data mining service capabilities and service quality,by integrating the resource advantages,technical advantages and servicce advantages of enterprises in the big data industry chain,middle and downstream,the Big Data Alliance provides data mining services to the society in the form of an alliance,which has played an important role in improving the big data service market.The service model adopted by the Big Data Alliance will directly affect the quality and level of data mining services.Building a data mining service model that meets the characteristics of big data service requirements and is conducive to integrating the data resources of alliance members is an urgent issue that data companies need to solve urgently.On the basis of tracking and summarizing the research status,development trends and practical applications of big data,big data alliances,data mining and service models at home and abroad,the related concepts and characteristics of big data alliances and data mining services are analyzed and defined.From the perspective of cooperation and collaboration,using core competence theory,synergy theory,game theory,etc.,to reveal the cooperation mechanism of bigdata alliances.Based on analyzing the data evolution process and data mining level of big data alliances,a data mining service model framework for big data alliances is constructed.On this basis,the standardized data mining service model,customized data mining service model and intelligent data mining service model of the Big Data Alliance are proposed.The standardized data mining service is a kind of data mining service provided in response to user common demand.This service model combines the types of users' needs to encapsulate the resources obtained by the alliance through data mining for service products,and to publish and deliver them in the form of data products.According to the different levels of users' awareness and understanding of services,the service methods are divided into two self-service and consulting service methods,and standardized data mining service guarantees are proposed from the aspects of after-sales service and operation supervision.Customized data mining service is a data mining service based on the user's personalized demand.This service model builds a customized data mining service process model for the big data alliance from the description of alliance user needs,service capabilities mining,alliance data preparation,model construction,model evaluation,and service delivery,according to the degree of user participation.And the service method is divided into two types of service methods,commissioned and follow-up,and from the three aspects of coordination management,resource allocation,collaborative scheduling,put forward customized data mining service guarantee measures.Intelligent data mining service is a kind of data mining service based on user mixed demand.The service model is a fusion of intelligent decision theory and knowledge reasoning methods,from service demand knowledge representation,service intelligent retrieval,service reuse and adjustment,service payment,service knowledge learning and storage.And the Big Data Alliance's intelligent data mining service process model is divided into two types of service methods: intelligent interactive delegation and intelligent collaborative follow-up,and proposes intelligent data mining service guarantee measures from the aspects of multi-case experience integration and swarm intelligence.In the empirical research section,the Zhongguancun Big Data Industry Alliance is selected,and the ZWJ Big Data Alliance whose main business is providing data mining services is selected as the research object.The data mining service modelproposed in this paper is applied to the actual alliance to verify the data.Verify the scientificity and rationality of the data mining service model,and improve the data mining service model management system.Research on the data mining service model of the Big Data Alliance is beneficial to the circulation,sharing and value creation of data resources.While meeting the growing demand for data mining services,it will give full play to the resource and technical advantages of the alliance to mine the "knowledge wealth" behind data resources.In order to promote the vigorous development of China's big data service industry,improving the vitality of data resource value and the ability to innovate digital technologies has an important role.At the same time,the research results are helpful to enrich the enterprise resource management theory and service model theory system,and have important practical significance for the development of big data industry and the application of big data services.