Storing data

[akshika47]

We have to come up with a strategy for storing data in a secure manner that can be used for data analysis. At the same time, we should come up with a policy framework for analysing our data. Looking for suggestions!

[mahfoos]

We can use Anonymization techniques that can help protect the privacy of individuals whose data is being collected and analyzed. Use techniques such as de-identification and differential privacy to ensure that sensitive information is not disclosed.

[mahfoos]

There are a few libraries in Python that you can use for de-identification of data

1. Faker
2. Anonymizer
3. Deidentify

[akshika47]

This sounds wonderful. Do you want to start coding up a solution to de-identify data_set in a CSV file?

We can start with the ScholarX 2021 student application data set and the CSV file would have the following columns

Timestamp | Full Name | Email address | Please share your LinkedIn profile | University/ Institution | Department/ Major | Grade Point Average (GPA) | CV Submission | Describe your achievements to us | Describe any professional experience you have gained in your field to us | Please let us know your top areas of interest (maximum 5 areas) | Please elaborate your career aspirations/ goals to us | Please select the skills/ related fields that you are most interested in developing/ pursuing | First Choice | What do you expect to gain from the ScholarX program through this mentor choice? | Second Choice | What do you expect to gain from the ScholarX program through this mentor choice? | Third Choice | What do you expect to gain from the ScholarX program through this mentor choice? | Would you like to join the ScholarX Management Team as a volunteer to help make the program successful for next year?

Let me know if you need further information. We can set up a call when you have some code developed and I we can run the code to anonymize the data_set.

[mahfoos]

lets start from cleaning the data set put some analyse such as graph or something , then there are some preprocess step available for train the de-identify model can you add the data set in this repo ?

reference for the model - https://bmcmedinformdecismak.biomedcentral.com/articles/10.1186/s12911-019-0935-4