The products and services of our company of interest, the Datateam, are mainly focused on data analysis and visualization through the graph database approach, and implementation of machine learning in data analysis technologies. One interesting usage of such technologies is seen in one of the latest projects the Datateam has been working on as of the writing of this project: the determination of work efficiency of employees in a company through the analysis of their work related data. Such methods of big data usage can enable corporations to learn more about their workforce, increase productivity of their business, and introduce revolutionary business processes in their areas of interest. While the use of this type of technology can seem to be harmless and beneficial only, given its promises on organizational efficiency of corporations, it has certain drawbacks. The use of this type of technology requires the tracking of the employees extensively and measuring their performance against industry benchmarks. This in turn can create an organization in which a high level of supervision is in place, and this can be overwhelming for workers. While such a level of monitoring can be in the best interest of a corporation through the use of data analysis methods mentioned above, it is arguable whether such an environment would be desirable for the employees of that corporation. Therefore, the need for privacy of the employees shall be addressed in the design of such analysis tools, while also providing enough information such that the analysis can be done properly. These concerns correlate with the confidentiality and transparency of the FACT model of responsible data science approach. However, a high level of monitoring may not be the only concern with this type of data analysis usage. Given the conditions in which this technology may be used, one can deduce that the dismissal of employees in a corporation may possibly be based on the results of the data analysis methods introduced by the technology. Therefore, the accuracy of the results obtained through the analysis of work data of employees becomes even more crucial in the sense that they can cause the dismissal of an employee unfairly, if the performed analysis is inaccurate. These concerns are addressed with the fairness and accuracy of the FACT model of responsible data science approach. All in all, the development of such type of data analysis technology in accordance with the FACT model of responsible data science approach is of crucial importance, because the lack of implementation of this approach in data technologies can result in undesired consequences for society, for the convenience of which the data technologies were developed in the first place.

From our findings, it is evident that innovation should be done in a responsible manner, and data science is no different in this aspect. There are risks and concerns that need to be addressed to make data science innovations more socially desirable, and this requires public interests to be taken into consideration. Responsible data science approach is precisely about considering the risks and concerns of the data technologies and inscribing the associated values with these concerns in the innovation of new data science developments. The four fundamental values that need to be implemented in a responsible data science approach are fairness, accuracy, confidentiality and transparency, all of which have been explained in previous sections of this project. As our research and findings demonstrate, these values are indeed considered and implemented in the design of new data technologies, because users of these technologies demand these values from data technologies, and data scientists are then obliged to implement these values. Thus we can say that data technologies are user driven and they are shaped by the demands of the users of the technology, since developers cannot ignore the demands of their customers.