Correct issues from peer review

[aysterberi]

1. Are data accurately analysed?
   Under section 4.2 in your thesis, the provided data for the two algorithms, Fisherface and Eigenface, is accurately represented in two separate tables. Later in the thesis, under section 5.2, the analysed data is also accurately stated in both numeric and percentage form.

2. Is the discussion presented in a clear and logical way?
   In the beginning of section 5.1, you briefly mention that there were some small issues with the artefact. You do mention that there are known workarounds as well to these issues. You then proceed to describe the issues that you personally encountered during your research which is really good. Further down you do discuss again your choice between the OpenCV library and a finished API. You do back up your choice with good points regarding possible implications of data integrity if you decided to chose to the API over the OpenCV library. Here however, you can go one step further in my opinion by discussing the possible integrity issues of OpenCV as well. While security might not be dependent on the library itself, it can be dependent on the computer that is being used. There are many security issues which can be discussed, if not mention briefly. I think section 5.2 - The test results, should be moved out of the Discussion chapter of the thesis and put under the section Result. Afterwards you can refer to and discuss the results in the discussion chapter instead. It creates a better flow rather then analysing the data and discussing it under the same title. One thing I felt missing here was that there was description of the environments that you used to test the two face detection algorithms. You mention them very briefly under section 3.2 - Application of method, but never describe them in detail. I see this as a minus as you say these two environments are drastically different from each other. It would be great if the readers were informed in what way they differ other than the employees that work there. This would also provide information which can be discussed in the Discussions chapter of the thesis.

3. Are the results discussed in relation to previous research?
   The results discussed lacks information regarding previous research of the two algorithms. In your results discussion you can definitely bring up previous reports about hits, misses and unrecognisable data. You can also discuss your results further and talk about the difference (if any) between the data you have collected and the data from any other previous research.

4. Is the methodological approach discussed?
   There is a detailed discussion about the methodological approach in this research.

5. Are conclusions justified?
   Yes. I like the discussion around the use of the artefact in real life scenarios and its implications. However, I do believe that you should bring up the research question again in your conclusion together with the results you obtained to formulate your answer. This will wrap the thesis a lot better. Overall the conclusion reflects over the research question, the limitations and possible ways to combat these limitations have been suggested.

6. Is the aim of the study fulfilled?
   I believe the aim of this study has been fulfilled as the results which were obtained provided some interesting insight of how the two algorithms worked. In regards to the research question, there was always a discussion around the constraints of implementing such algorithmic surveillances. Furthermore, there were also suggestions and discussions about how such constraints can be solved.

7. Are implications of the study well addressed?
   There were some brief mentions of certain implications in the thesis. I suggest that you address these implications in a paragraph or two under a separate subtitle.

8. Are suggestions for future research reasonable?
   While you discuss a working artefact after your conclusion, there are currently no suggestions for future researches. It should be mentioned for a possible future research based on the results obtained by conducting this research. As mentioned in the thesis instructions (can be found on the scipro website), in order to be rewarded with 2 points the limitations of the study must be thoroughly discussed and that a profound reasoning about possible and relevant future studies is made. After your conclusion you mention a working artefact, here you can also suggest what the future studies can focus on, in-order to improve face recognition in office environments. You can refer to the implications you encountered during your research and suggest how a future study can be performed better in-order to obtain better and more accurate data. Limitations can also be brought up by suggesting how a future study can have a broader reach.

[aysterberi]

Added section "Previous studies" under section 5.2