Automatic Generation of a Metric Report: A Case Study of Scientometric Analytics

Abstract

Automatic report generation is an emerging technology that mechanically generates documents in the form of a report consisting of text, tables, and figures about a specific topic. This paper proposes a model for automated generation of a scientometric research analysis report for any selected country. The scholarly database utilized in this study is Microsoft Academic Graph. Given the two-letter code of the selected country, starting study year, and ending study year data concepts to the employed database, the model extracts the datasets, including the scientific research information for the selected country in the specified period, and generates an in-depth analysis report about the country’s research publications. The model consists of two main stages. The first stage extracts the datasets for the selected country from the utilized database using Azure Databricks and Azure Blob Storage services. The second stage utilizes a predefined scientometric research analysis report for generating a new report for the selected country. A case study on big data analysis for Saudi Arabian research publications was conducted. An evaluation was performed on the report within 10 evaluators to understand the practicability of the proposed model. They evaluated the report through four-point criteria on the user perceptions. The results indicated that the model was able to successfully create quite a pleasing scientific report in terms of factuality, coherency, sufficiency, and its ability to impart new knowledge for the readers.

Keywords

• Automated report generation
• big data
• coefficient collaboration
• collaborative research
• growth rate
• scientometrics