Data Analysis
Budget: $10 – $30 USD
You will be expected to compile the numerous individual .csv files that you are provided with into a
single dataset. You will be expected to use a Jupyter Notebook to integrate the datasets and dealt
with any issues that arise, e.g. differences in date/time formats, missing values, excluding any personally identifiable information.
You should formulate a set of questions about the data and then use appropriate analysis and
visualisation techniques to attempt to answer these questions. Your Jupyter Notebook should
include comments that explain the analysis you are performing and provide outputs such as data
visualisations that achieve your analysis results.
At the end of the project all of the below points should be completed:
Does the code process all of the .csv files provided, combining all of the
individual datasets into a single dataset for analysis
AND
Does the code/markup clearly demonstrate how you have chosen to deal with
any data wrangling issues (e.g. missing data, unrecognised activity codes,
incompatible date/time formats).
AND
Are these issues dealt with sensibly? (e.g. ensuring that different date/time
formats are resolved, returning error messages or handling exceptions if
incompatible .csv files are provided).
Does the Jupyter Notebook contain comments/markup that clearly explains
questions or hypotheses about the data, which were used to inform the
analysis?
Does the code demonstrate that appropriate analysis steps have been taken in
order to answer these questions or test hypotheses?
single dataset. You will be expected to use a Jupyter Notebook to integrate the datasets and dealt
with any issues that arise, e.g. differences in date/time formats, missing values, excluding any personally identifiable information.
You should formulate a set of questions about the data and then use appropriate analysis and
visualisation techniques to attempt to answer these questions. Your Jupyter Notebook should
include comments that explain the analysis you are performing and provide outputs such as data
visualisations that achieve your analysis results.
At the end of the project all of the below points should be completed:
Does the code process all of the .csv files provided, combining all of the
individual datasets into a single dataset for analysis
AND
Does the code/markup clearly demonstrate how you have chosen to deal with
any data wrangling issues (e.g. missing data, unrecognised activity codes,
incompatible date/time formats).
AND
Are these issues dealt with sensibly? (e.g. ensuring that different date/time
formats are resolved, returning error messages or handling exceptions if
incompatible .csv files are provided).
Does the Jupyter Notebook contain comments/markup that clearly explains
questions or hypotheses about the data, which were used to inform the
analysis?
Does the code demonstrate that appropriate analysis steps have been taken in
order to answer these questions or test hypotheses?