Abstract:

An Analysis of Big Data Impact on Scientific Production: Books, Videos, Conference Papers in Google Books, YouTube and IEEE Explore® Digital Library is the subject of present laboratory seminar. Queries append public records to two sets of data sets (research paradigms): Big Data (incl. Big Data: Analysis, Engineering, Architecture, Governance, Management, Frameworks) and Big Data interdisciplinary fields: Data Science, Data Mining, Deep Learning, Machine Learning, Artificial Intelligence. Metadata from more than 25 000 conference papers, 2000 books over the past 50 years and 4 000 videos for last 12 years, matching the searching criteria, has been stored and analyzed. The outputs are summarized in statistic, forecasting, rating key findings by various attributes: title, author, publisher, research field, category, subject, publication year, description, abstract, view count, and a combination of indicated metadata in cross-tables. The concept of experimental session lays on a case-based learning approach designed for graduate students in software engineering. This paper aims to reveal a guided knowledge discovery process in quantitative analysis facilitating the skill-building in data-driven research and assisting a master's thesis preparation. Practical strategies and activities are developed to examine the capabilities of Application Programming Interface (API) technology in scientific projects. The methodological foundation grounds on Big Data Interoperability Framework (National Institute of Standards and Technology, NIST, USA) through adopting concepts and templates in accordance with the educational purposes. The problem-solving focus serves to define data-driven hypotheses motivating the need for analysis and outputs interpretation. Big Data Knowledge Domain in the role of indicative medium extends learner’s critical thinking about interdisciplinary nature of software processes. The paper outlines a modern framework of dynamic instructional environment. This strategy engages students in communicative and collaborative activities that require to sort out actual data in order to make a decision on given task or find a solution to determined case. The session interconnects Software Engineering - Big Data - Analytics in a foundational learning paradigm. Close scientific relation illustrates the idea behind the present laboratory seminar: applying a skill-building interdisciplinary approach covering a wide range of knowledge to solve specific research needs using quantitate data analysis. Tools, practical guideline and a sample of findings are also presented.

Keywords:

Lab Session, Data Analysis, API, Scientific Publishing, Big Data.