Data science combined with static

written by Kim Raon 

In the current era of data revolution in the 21st century, the inundation of big data and public data has brought about a significant change in the perception of statistics. Rather than being an exclusive domain of experts, statistics have become a crucial factor that intimately impacts the lives of individuals. In response to these changes, there has been an innovation in statistical production methods, which incorporates data science.

This innovation has enabled the development of new techniques to analyze and interpret data, leading to a better understanding of the underlying patterns and trends. As a result, businesses and academic institutions are now able to make more informed decisions based on the insights derived from these advanced statistical methods, which also offer more accurate and user-friendly statistics while minimizing the burden on respondents.

Data science is an interdisciplinary field that blends computer science, statistics, and business analysis to extract insights from data. It involves the use of various technologies and tools to collect, process, and analyze large sets of structured and unstructured data in order to identify patterns, trends, and relationships.

The work process of data science is

1.  Understanding the problem

2.  Preparing a data sample

3.  Creating a model

4.  Applying a model to know how a model works in the field

5.  Placing on the site

This technology harnesses the power of big data to achieve its objectives.

Big data is an advanced technology that enables the processing of vast amounts of data, characterized by high volume, velocity, and variety.

 In comparison to traditional statistics, data science including big data offers the advantages of being timely and cost-effective. However, due to its collection methods, such as “data crawling(a method which involves data mining from different web sources)” and “aggregation sensor(a composite type sensor which serves to summarize or to average the performance of other sensors.)”, big data lacks representativeness of the population.

Additionally, the analysis methods differ from traditional statistical production methods, such as data mining, machine learning, and optimization. Therefore, significant supplementation of technology development is required to use big data for statistical production purposes.

The utilization of algorithms in data science for statistical purposes can often result in issues. While the algorithmic approach offers an advantage over the existing parameter approach by allowing the application of complex data, it also has the disadvantage of being difficult to interpret the results, as only machines can recognize them.

In addition, due to limited involvement in the data collection process, researchers have little control over the data that is collected. This lack of control creates conflicts with personal information protection when dealing with big data that covers search patterns, access records, location information, and more. To address these issues, a legal review is necessary, as well as exploration of ways to improve the legal system.