+1443 776-2705 panelessays@gmail.com

Finish it in 24 hours

No reference needed

During the course of your project, you will apply the techniques learnt in class
to the study of a real social media analytics dataset. This first step in the
project is intended to help you form a team and define the problem along with
an associated dataset(s).

For now, develop and submit a project report (in PDF) containing the following

1. Your team composition. The project can be performed in groups of 1-3
people. If you decide to do it solo (i.e., a group of 1), note that the expectations
of a complete project are still the same as for groups with more members.

2. Identify a suitable social media related dataset that you would like to work
on for your project and provide a short summary of the dataset (e.g., what the
dataset is about, how big it is, when was it collected). It would be ideal if the
dataset contains i) some type of network information, ii) some type of text
information, and iii) some type of endorsement/recommendation information.
For example, a collection of tweets will satisfy this requirement since you will
have Twitter network follower structure, textual content (of tweets), and
retweeting/liking behavior. Similarly, you can find other datasets that contain
similar facets of information, e.g., from Youtube, or Facebook, or other social
media. You can choose datasets from public resources such as:

http://snap.stanford.edu/data/index.html (Links to an external site.)
(Links to an external site.)
datasets#socialnetworks (Links to an external site.)
https://www.kaggle.com/datasets?search=network (Links to an external site.)

or you can choose other datasets that you have access to that are not
necessarily public.

3. Define a suitable “problem” over this dataset. For instance, a problem for a
Twitter dataset could be to understand popularity of different automotive
products. A problem over a movie rating dataset could be to recommend
movies to users based on other movies they have liked, and so on. Make sure
the problem is realistic/meaningful.

At this stage, we just need to know the team composition, what dataset you
will use, and what problem you will study. No actual analysis or computations
are necessary.