the workflow we proposed requires human
effort for data analysis and interpretation. This is necessary
because our purpose is to achieve deeper understanding of
the student experiences. To the best of our knowledge, there
is currently no unsupervised automatic natural language
processing technique that can achieve the depth of understanding
that we were able to achieve. There is a trade off
between the amount of human effort and the depth of the
understanding. The labels generated can be applied to any
similar data sets in other institutions to detect engineering
student problems without extra human effort. Often times,
manual analysis is time-consuming not only because of the
time spent on analyzing the actual data, but also the time
spent on cleaning, organizing the data, and adapting the
format to fit the algorithms. We plan to build a tool based
on the workflow proposed here combining social media
data and possibly student academic performance data. This
tool can assist in identification of students at risk. This tool
will provide a friendly user interface and integration
between qualitative analysis and the classification and
detection algorithms [61]. Therefore, educators and
researchers using this tool can focus on the actual data analysis
and investigate the types of learning issues that they
perceive as critical to their institutions and students. This
tool can also facilitate collaboration among researchers and
educators on data analysis. Advanced natural language
processing techniques can be applied in the future to provide
topic recommendations and further augment the
human analysis results, but cannot completely rule out the
human effort.