Comments on: Data science, big data and statistics

Abstract

The invited paper by Peña and Galeano is a very inspiring one. The authors analyze how Big Data is changing statisticians' minds by describing new statistical methods in seven areas: the emergence of new sources of information; visualization in high dimensions; multiple testing problems; analysis of heterogeneity; automatic model selection; estimation methods for sparse models; and merging network information with statistical models. They also compare the statistical view with the ones of Computer Science and Machine Learning and put a lot of emphasis in the emergence of the new field of Data Science. Two nice examples of Big Data statistical analysis are also included.When the reader goes through the paper, plenty of interesting ideas appear. The problems dealt with are also very inspiring, suggesting interesting aspects that surely will need deep insights in the future. Five of them are mentioned in the following subsections.

Multivariate six sigma: A key improvement strategy in industry 4.0

“…Six Sigma was created in a data-scarce context: low number of variables with small sample sizes (usually more observations than variables) that in most of the cases had to be measured (or generated) (i.e., there were no data available at the beginning of the Six Sigma project). Once defined the project, the Measure phase begins not by measuring but by asking questions, following the typical Question-Data-Analysis (QDA) paradigm in Statistics (Cao 2019). However, in the data-rich context typical of the Big Data era, in most of the situations, a lot of data are already available before the project is even defined.…”

Discussion of “A review of data science in business and industry and a future view” by Grazia Vicario and Shirley Coleman

“…This is the classical approach in Statistics, following the Question‐Data‐Analysis (QDA) paradigm . Once the objective of the study is defined, questions give rise to collect (or generate) data and analyze it in an iterative scheme of several steps: model selection, model estimation, and model validation, until the model is considered satisfactory and some answers can be found from the questions posed.…”

Research on Evaluation and Improvement of Government Short Video Communication Effect Based on Big Data Statistics

“…With the vigorous development of new media for government affairs, it is very urgent and necessary to adapt to the form of policy publicity in the new era, improve the efficiency of information transmission, optimize the effect of public opinion guidance, and innovate the governance mode. Relying on the authority of the official account in the social platform, it is necessary to publish authoritative news, establish a good image, and transmit positive energy through text, pictures, videos and other forms [13]. This paper establishes a feature reconstruction model for the evaluation of the dissemination effect of government short video, combs the dissemination effect and influencing factors of government short video through content analysis, analyzes the main factors by using the linear regression of big data statistics, and extracts the fuzzy feature quantity of government short video.…”