The process of monitoring academic publishing performance in higher education: A scientometry-based business intelligence model

Scientific publications are communication tools that can be expressed as the output and performance indicator of scientific activities, through which scientists convey their work, share their working methods, benchmark their qualifications. While the records of the publications of higher education institutions submitted to the literature can be revealed with certain metrics and statistics, the effectiveness of the current processes adopted for monitoring the publication performance in higher education institutions is a debatable phenomenon. In order to measure publication performance, institutions can use academic CV web applications created within the institution, or YÖKSİS CV system as well as benefit from the reports they can obtain from databases such as Web of Science, Scopus, PubMed that provide bibliometric data to researchers. With the help of the analysis to be made on the bibliometric data to be obtained through these databases, the publication performances of countries, institutions such as universities / institutes, institution employees and journals belonging to the institution can be measured, monitored and publication qualities can be checked. In this way, policies can be established to increase scientific productivity and quality. Since the reports that can be obtained through the relevant databases have limited content and are specific to that database, a holistic monitoring cannot be provided, and it may be difficult for universities to monitor publications, to determine researcher publication activity in national and international journals, and to obtain detailed parametric reports. At this point, it seems that performance monitoring processes and systems that meet this need in a generally accepted standard, systematically meet this need and comply with international criteria, have not yet been implemented, and it is thought that this need should be addressed as a priority research topic. In this direction, the purpose of the application carried out in the thesis is a process model that includes the integrated use of database management, data mining and scientometric methods in order to effectively monitor the performance of scientific publications in higher education, primarily through publications in generally accepted databases. It is the development and implementation of a business intelligence model that will enable effective monitoring of the process. Within the scope of the subject, first of all, in order to determine a generalizable process and performance criteria, interviews were made on a structured form with the higher education institutions to be selected as representative. Through enhanced forms of business processes created for academic publishing performance of universities in Turkey and the problems they face in this process have been analyzed and recorded. An ideal business model has been demonstrated for universities with process modeling techniques. Common performance indicators have been developed based on bibliometric data belonging to publications scanned in Web of Science, Scopus and ULAKBİM TR Index databases. An integrated electronic system based on business intelligence has been designed for all analysis and visualization techniques. This system can be used as an academic performance evaluation console that can be updated in line with the needs in the periods determined for the institution managers. During the implementation phase, Dokuz Eylül University was chosen as the pilot institution. In the first stage, the application is limited to Web of Science, Scopus and ULAKBİM TR Index databases. Converting these databases, which have different data structures and can also contain common records, into a single data warehouse by overlapping each other is among the primary outputs of the project. In this regard, the first research and development activities will be carried out on the selected databases, and if the association problems are solved and consistent data is obtained, other databases will be easily added in the future. As a result, the developed application, through the parametric reports to be produced can collect many information such as universities, academicians' publication performances, publication qualifications, publications with academicians from different universities, funding status of the publications, citation status and journals where the publications are published. In the study, by using data and, in particular, text mining methods, on bibliometric databases, reports can also be produced on the content of the publication, so that the work areas where the institution is effective can be monitored, while the implementation of research strategies can also be monitored. Thanks to the developed application, data can be collected from many data sources on a data warehouse system created. Screens are shaped according to the research questions determined and key performance indicators. The technologies of the developed application and the developed business model have got the potential for the integration of the data from different systems of higher education institutions. This situation demonstrates that the potential value of the study is very high. The study integrates many different disciplines. The study presents an upto-date and effective solution for public institutions. Furthermore, it is a different performance management approach for Turkish higher institution system.