Recently, Shanghai TV has a news report about Xuhui District Civil Affairs Bureau’s assistance to families in need:
We can see from the news and traditional way of relief through the often difficult family application, review, examination and approval procedures, meet needs across different departments, hierarchy, cross blocks resolve the case, you also need to hold the meeting at all levels in the form of a negotiated solution, like hu old man take the initiative to find this kind of situation is hard to civil administration branch.
Now through the way of “precision”, civil affairs departments in the usual MoPai to hu old man’s house, to enter information into the “one network over” data center, and then according to the data model to identify the hu old man’s house is “families”, then formed on the repair order to social workers to help, so as to improve rescue efficiency and do it now.
For this reason, “One Network Management” precise rescue scenes have also been named and praised by the State Council:
The General Office of the State Council issued a circular praising 43 typical practices, of which three were listed
As a partner of Shanghai Xuhui District Civil Affairs Bureau, SmartBi is responsible for the development of accurate rescue scenarios from the underlying data to the front-end application, which is also fully recognized by customers.
01. Collect data on people’s livelihood from multiple departments and build a basic database
The precise rescue scenario of “One Network Management” takes the actual population database of public security and GIS map of the Institute of Surveying and Mapping as the lowest level data, superposition about 16 million pieces of data closely related to people’s livelihood from 14 departments including Civil Affairs, Human Resources and Social Security, Disabled Persons’ Federation, Veterans Affairs Bureau, Housing Administration, Health and Justice, etc., and forms the basic data source for people’s livelihood.
Then, the data source is collected, shared and managed by the data center platform to build the basic database of the DMCS data center.
Through SmartBi ETL function and manual processing, the basic data of 14 government departments in the region are collected to provide support for the construction of applications that can meet the needs of various livelihood services. Data sharing is a data sharing mechanism built under security control, which provides flexible, convenient, safe and controllable data access channels and interfaces for the projects planned now and in the future. Data governance norms Big data system, including data cataloging, data management norms, data security norms, etc., for the collection of big data and data analysis application to lay a reliable data foundation.
02. Data mining, establishing family tags and carrying out data modeling
Data is the starting point. After collection and management, the business data of 14 departments and 45 types of business label data are very useful basic assets for data mining practice. Enough diversified and unbiased data can ensure accurate and unbiased service applications in subsequent development.
Through careful analysis of people’s livelihood business and selecting and looking for appropriate formulas, statistical models or forecasting algorithms to build data models, dozens of front-end applications of the precise rescue scenario under the “one network management” can accurately obtain service objects from the complicated data.
In the design of data model, and xuhui district civil affairs bureau and the east China university of science and technology cooperation Smartbi common set up relief model for accurate research topic, through research and development with independent intellectual property rights of mathematical algorithms and models, should make full use of its advantages in huge amounts of data and can assign xuhui district people’s livelihood service, provide the valuable data for the civil affairs department and business advice.
In terms of the landing of the data model, based on the SmartBI data mining platform, the implementation threshold can be greatly reduced. The platform integrates several commonly used big data mining algorithms, including classification algorithm, clustering algorithm, timing algorithm and so on. According to the six aspects of family employment, housing, pension, medical care, education and assistance, family labels are established and data modeling is carried out to form a large livelihood indicator system, a family evaluation model and a focused family model, so as to provide guiding data mining and analysis services for front-end business applications
03. Analysis and application to meet the different needs of precise rescue scenarios
Family business on the basis of data mining platform labels and data analysis, through Smartbi domestic production function key developed signs of the people’s livelihood, family portraits, accurate rescue and wisdom endowment and community is closely related to the basic livelihood of the people, such as the front end of the service application subsystem and scene, also provides a number of community collaborative application system, match the community residents in the basic livelihood of the people demand.
Among them, the characteristics of people’s livelihood are documented for communities, residents and families from three dimensions, and 45 personal and family attribute labels are formed to display the basic characteristics of people’s livelihood in the whole area.
Family portrait interpretation of the basic information data package and social workers and service data, by helping family core index analysis, index analysis and family type index of supply and demand analysis for tag, business and derivative, two dimensions in time and area calculation infer focuses on helping families, to support the family, three types of families generally support family, Provide guidance and suggestion for rescue and pension and other follow-up business.
For example, in the intelligent discovery process, based on the basic database of people’s livelihood, portrait analysis is carried out from seven dimensions: the young have good education, the weak have public support, the students have excellent education, the sick have good medical treatment, the old have good care, and the living have livable. Each dimension can be subdivided into small categories, and each small category of indicators can be cross-combined analysis. Through the combination of indicators of different dimensions and family labels, it can flexibly and accurately locate the matched population, family and relevant basic information under different screening conditions.
Shanghai Xuhui District, based on the actual needs of families in need, makes full use of big data technology to create a “one network management” precision relief scene, which greatly improves the scientific and accurate nature of social assistance and makes the city warmer. SMARTBI is honored to participate in the design and development of the project. In the future, SMARTBI will continue to give full play to its professional advantages and continue to explore in the field of data analysis, so that big data can better serve people’s livelihood.