To help the authors better understand the aim and scope of the journal, we proposed new sections as follows; not all sections have to appear in one issue.
Philosophy of Data Science
The aim of this section is to highlight the role of critical scientific reasoning in data science by providing dedicated commentary by exemplars in the field. Instead of taking the typical approach of discussing how data science advances science, this section will discuss how the foundations of science underpin data science.
Statistical Data Science
This section is the home base of the reformed journal covering statistical methods that are motivated by real-world applications. It is not for papers with technical proofs that push the frontiers of theoretical developments. In addition to classic topics in Statistics, cutting-edge works on big data, visualization, machine learning, and artificial intelligence are also welcome.
Computing in Data Science
Computing is an indispensable component of all data science and big data applications. As more journals on data science and big data emerge, it is of great interest for the data science community to have a highly regarded outlet with a specialization in computing, covering a wide spectrum from methods, algorithms, software implementations, to case studies. Existing journals on statistical computing have their own traditions and may not meet the increasing demands. Some research works on cutting-edge problems may not fit well in any existing journal.
This section covers the following types of articles.
1. Software: Articles here are similar to those in the Journal of Statistical Software. They are not referencing manuals but vignettes that introduce the methods being implemented as well as the usage of the software with reproducible code chunks. The software implementation can be in any computer language with a sufficiently large user base.
2. Algorithms: Articles here focus on the performance side of the computing needs arising from domain applications. For example, one can propose algorithms that make infeasible tasks feasible or speed up existing algorithms.
3. Methods: Articles here are similar to those in Journal of Computational and Graphic Statistics or Statistics and Computing. The computing methods need to be motivated by a domain application with the properties carefully studied.
Data Science in Action
Data Science is applied by nature. Applied papers in all statistical journals have the tendency of becoming more and more theoretical despite the original design. Statisticians and data scientists should go out to the field to work on applications. Applications of data science in all domains are welcome.
In addition to all domain applications, we consider entries from data science competitions. There are many data science competitions such as Kaggle, Kesci, and DataCastle, among others. We welcome winning teams of data science competitions to share their work in the form of an academic paper, with our training.
This section is modeled after the Application and Case Study Section of the Journal of the American Statistical Association, the Biometric Practice Section of Biometrics, or papers in the Annals of Applied Statistics. Articles can present detailed case studies of data science or big data applications. The data and the computing code should be made available such that any interested reader can reproduce everything in the article from the data and the code. The problems can come from any research domain that involves data science and computing, including science, technology, engineering, medicine, health, social science, humanities, arts, among others.
Data Science Reviews
Reviews and tutorial articles on the latest data science techniques can be of interest to readers who want to get into a new field or pick up new skills. Experts of certain fields are invited to write tutorial articles. This section can be modeled after the Statistical Science or the Tutorials in Biostatistics Section in Statistics in Medicine.
Education in Data Science
Education in data science is ever-evolving along with new methodologies, technologies, and application fields. Discussion on the curriculum of data science and tools in teaching data science needs a good outlet. This section will be modeled after the special issue on data science of The American Statistician. In addition, a subsection on training practice may publish articles from outstanding student projects.
This section contains articles that do not clearly fall into the sections above. For example, the articles that are not directly motivated by data science applications, such as those on new probability distributions that the journal published, can be published in this section.