• What are the places frequently visited by you?
• How often do you visit the same place?
• how does it correlate with the location of you house/ college?
  Did you notice a pattern here?

Description

As a facebook-archive user,
I need to have EXIF Data (at least Date and Time (original)) from my photos archived,
so that I can upload / store this photos in an organized manner.

Explanation

Facebook archive doesn't provide EXIF data from your photos.
There is a way to obtain it by going to you_archive/photos_and_videos folder/album .
Here, there is all your albums in .html format.
In each .html, there is each photos and a link with photo timestamp.

Acceptance Criteria

Update [Required]

• All photos in your_archive_/photos_and_videos/* should have an EXIF data (at least Date and time (original) attribute) extracted from your_archive_/photos_and_videos/album/* html files.

Definition of Done

• All of the required items are completed.
• Approval by 1 mentor.

Estimation

1 Day

What all events you have created so far?

• Do they correspond to some particular page that you're handling?
• Are they independent events

How to correspond to an event being invited to?

• Do you generally accept the invite and never go
• Do you mark it as interested?
• Did you observe a pattern here?

• How often do you search for yourself?
• How often does a user search for a person, vs a group?
• Which friends do you search again and again?

Extensions to this could be:

• which people do you search for and immediately send a friend request to ? (Possible to get since time stamps logged)
• Bonus: Did you react / comment on any of their posts recently before sending a friend request?

This is an addition over the current feature.
How many friends you make, based on received friend requests, and requests you sent. For each day, you have 2 entries, friends you made who sent you requests, and friends you sent a request to. (ed

• How often do you use love reacts?
• What friends' posts are you most likely to react to?
• Is their a friend for whose posts you use a single react?

Description

Extend functionality of plot_friends.py to specify from and to period. Currently doing for all time.
Something of the type: python plot_friends.py --from 2018-12-01, should support both --from and --to flags, both optional.

Mocks

Steps to reproduce (Incase of a bug)

NA

Acceptance Criteria

Update [Required]

• Support --from and --to flags in plot_friends.py

Enhancement to Update [Optional]

• [LIST ITEMS]

Definition of Done

• All of the required items are completed.
• Approval by 1 mentor.

Estimation

4-5 hours after having an idea of original code

This task could be a bit harder, since you have to process both posts and comments, and then aggregate words from both, to form a wordcloud.

• What hashtags do you frequently use?
• What language do you often use?
• Skip proper nouns, esp. all your friend's names.

Classify a facebook user's ad interests into categories, say sports, engineering, humor, developing, etc.

This issue has a lot of potential, and is currently only in ideation phase. Please discuss your vision about how / what can be done here.

Plot number of messages with each friend as a function of time. Since this could get messy, plot only top 10-20 friends.

This could include treating any message exchange as a message, or
it could have two plots:

• top 10 friends by who message you
• top 10 friends who you message

We have few scripts but they are not documented enough, a lot of functions are missing docstring, I feel it will be a good addition to have a description regarding what are they doing.

Description

There appears to be a bug in the plot_messages.py script due to a discrepancy between the naming convention currently used by Facebook vs. the name expected by the script for .json files inside the messages directory. The script expects those files to be named "message.json". However, in a download that I did this week of my Facebook archives, the files were named "message_1.json". This naming discrepancy causes the plot_messages.py script to fail to recognize the files and subsequently throw a fatal error.

I also noticed some documentation issues in the project's README.md file related to plot_messages.py. Currently the README.md file says there is an "-a" option for generating plots of all messages, but no such option exists. Also, the instructions for specifying a user's ID when plotting messages for a specific user don't work.

Acceptance Criteria

Update [Required]

• Running plot_messages.py will not throw fatal errors.
• The script can successfully find and process message files that match either a pattern of "message.json" or "message_\d+.json".
• The README.md file does not give incorrect information about how to run the plot_messages command.

Definition of Done

• All of the required items are completed.
• Approval by 1 mentor.

Estimation

1 hour

Hi man! Great work on creating exactly what I wanted to do with my Facebook archive! I'm trying to run your scripts, but I'm afraid some libraries (or the right versions) are missing. Could you make a requirements.txt file, so that people can easily run it?

Check out this link to see how:
http://www.idiotinside.com/2015/05/10/python-auto-generate-requirements-txt/

Description

The README file indicates the code uses python 2.7.

However, when I tried to run plot_friends.py, the given example gives an error.

$ python plot_friends.py
Enter facebook archive extracted location: ./examples
Traceback (most recent call last):
  File "plot_friends.py", line 87, in <module>
    friends(loc)
  File "plot_friends.py", line 22, in friends
    loc = input('Enter facebook archive extracted location: ')
  File "<string>", line 1
    ./examples
    ^
SyntaxError: invalid syntax

This is because the python 2.7 input function expects a python expression that can be executed.

Mocks

N/A

Steps to reproduce (Incase of a bug)

• Use python 2.7
• Install required dependency pip install -r requirements.txt
• run python plot_friends
• input ./exmaples

Acceptance Criteria

Update [Required]

Choose Either one of the solutions:

• Change to python 3.x on the README file
• Change to raw_input instead of input
• Use command line argument instead of having the user input it manually (recommended)

Definition of Done

• All of the required items are completed.
• Approval by 1 mentor.

Estimation

2-3 hours

• What time does the user post?
• What months the user is most active?
• Do posts with images get more likes? (ML-esque)
• What times do other people post on your wall (Obvious birthday posts hike)

More additions can be suggested.

Create a map which tells how often you checked in ,
and the circle size for any place would be proportional to the number of times you were in the city.
Example: If I was 5 times in Bangalore, 2 times in Kolkata, 1 time in Pune; I would get a map of India (world?) that shows these places, and the circle over Bangalore would obviously be the biggest.

Example: onemilliontweetmap.com

• Analyse what time user most likely to comment
• How frequently do you comment on a particular friend's post? Which friends' posts are you most likely to comment on?

None of these are straight answers, all of them involve graphs.

The goal is to develop a webapp so that the user does not have to use command line at all.

However, this webapp has to explicitly mention that no data is being uploaded anywhere, and if the user feels, they can download the code and run it on their local machines just to be sure.

I'm not sure how many people would be willing to use a webapp, but this could open up an avenue of opportunities in terms of visualizations.

Work on this will only start once sufficient features are supported in the command line, until then this issue will be marked wontfix.

• For a given chat thread, what percentage of messages do you send?
• How many conversations does a person initiate?
• Are there any friends who never reply?

You can use this to find which friend is most needy.

• How many chats are just You're now connected on Messenger. ?

The naming pattern for files and organisation for current archive is radically different as compared to the archive available at the time of writing this code first.

Since facebook now allows you to download data in html OR json format, we'll be using JSON for the following reasons, as suggested by @xprilion as well :

• JSON easier to work with
• JSON more accurate, as html tag search (current approach) might go wrong.
• JSON less susceptible to change.

This issue is more of a guideline, so won't be closed any time soon.