Big data is used to describe and define the massive data generated in the era of information explosion. Statistical departments need to actively respond to the environmental changes and demand changes brought by big data, better carry out the utilization of statistical files, tap the potential value of files to a greater extent, provide file information services more comprehensively, and realize the leap-forward development of statistical file information management.
Walk into big data
With the rapid development of economy and society, especially in recent years, the economic structure and social layout are constantly changing, and the statistical files are increasing sharply. At the same time, the demand for statistical files from all walks of life has also increased greatly. Massive data and huge demand push the collection, collation and utilization of statistical files into the era of big data. The statistics department receives a large number of people who consult statistical files every year, and tens of thousands of files are consulted every year. How to understand and apply the law from the complicated personalized needs and a large number of files, and how to ensure the privacy and safety of enterprises and respondents in the large amount of data provided? This requires the use of statistical data, which can be regarded as a process of "adding value" to data through "processing". However, how can the statistical department improve the "processing capacity" of data? First of all, we must understand the changes brought by big data to statistical archives work, find contingency measures according to the changes, and improve the statistical analysis ability.
"Three Big" Trends
At present, the use of statistical files has achieved good social and economic benefits. However, in terms of utilization types, most of them are still limited to "actual utilization", that is, general query utilization with specific practical purposes, with only data such as the number of people and the number of visits, without comprehensive analysis, research, prediction and other data processing processes, and without using the processing results such as preference rules and utilization trends. Simple data is of little significance to ordinary users, and it has nothing to do with economic development, planning, forecasting and regulation. However, with the advent of the era of big data, statistical file management will gradually show the "three big" trends, which will also bring corresponding challenges to the existing statistical file utilization.
First, big files. The carrier forms are more extensive, and all documents, materials, audio-visual images and objects related to the economy and society obtained by the statistical department through investigation are regarded as statistical files. Digital files are larger, and the original paper statistical data and all non-digital statistical files will be continuously processed into electronic files through information technology. ?
The second is big service. In the era of big data, statistical file service will develop towards socialization, diversification, openness and advancement. File services are geared to the individual needs of users and provide networked and intelligent services. ?
The third is large-scale development. In the traditional management mode, file utilization only provides passive access service. In the era of big data, the utilization of archives will extend to the massive processing and secondary development of archives by using big data technology, paying more attention to analysis, discovery and prediction, and creating more value for users.
Positive response
In the face of the challenges brought by big data to the utilization of existing statistical files, statistical departments should actively respond and comprehensively improve their data processing capabilities. ?
Do a good job in the adjustment and use of statistical items. The first is to shift the focus of using statistics from quantity to benefit. It is suggested that the utilization benefit items should be added to the statistical archives utilization items, emphasizing the statistics of social benefits, such as increasing user satisfaction or utilization effect items, and the satisfaction of users' needs should be included in the statistical items in the form of serial data to reflect users' cognition and recognition of archives value. The second is to improve the statistics of digital archives utilization, and set up statistical items such as retrieval, reading and downloading of digital archives and electronic documents according to the particularity of digital archives resources utilization.
Do a good job in the informatization construction of statistical work. In the era of big data, the traditional working mode of manual registration and computer summary is losing its original meaning. We must speed up the application of user self-service terminal equipment, service quality evaluation equipment, statistical software and the construction of information system. After the user verifies his identity on the self-service terminal equipment, his usage records and the evaluation of service quality after obtaining the service will be transmitted to the information system in real time and automatically stored.
Do a good job in professional staffing. Big data will make the use of statistical work large-scale, complicated and professional, requiring staff to have certain statistical theory and practical literacy. Therefore, it is necessary to conduct professional training for relevant staff in time, and exchange experience, skills and achievements in the specific practice of statistical work among staff, so that staff can learn from each other and promote each other.
Make good use of statistical analysis methods. Utilization statistical analysis is the last stage of archives utilization statistics. At this stage, through the use of various professional statistical analysis methods, we can have a clear and comprehensive understanding of the utilization situation, correctly estimate the situation according to the statistical analysis, and provide a basis for decision-making. For example, for the utilization times and volume times over the years, we can use the descriptive analysis of time series to simply describe the changes of phenomena at different times by calculating the growth rate of quantity and the average growth rate, and we can also observe the changing laws and trends of data over time by making charts, providing a basic basis for further prediction. When forecasting, we should first distinguish whether the historical data model is a stationary sequence, a linear trend or a nonlinear trend, and then choose the forecasting method of moving average, linear regression or exponential model according to the data volume.
Do a good job in the use and publicity of statistical results. Through statistical analysis, we may find some rules of statistical file utilization, such as users' preferences, and perhaps we can reasonably predict the trend. Statistical departments should actively apply these statistical results to service work, adjust service priorities in time or make service plans in advance, and even make clear statistical forecasts and early warnings.
The above is what Bian Xiao shared about how to strengthen the information management of statistical files in the era of big data. For more information, you can pay attention to Global Ivy and share more dry goods.