Download link:# ASSIGNMENT 1 Social network of TV shows actors
Description
5/5 – (2 votes)
Goal:
- develop a Python-application that builds a social network and computes a graph of relations between actors from TV shows.
-
- the nodes of the graph will be the actors,
-
- and the edges indicate that one work with another in the same TV show,
-
- the edges will carry a weight representing the number of times the actors have worked together.
Given a corpus of data, you will extract who has worked with whom, and whenever a connection is found, an edge is added to the graph of a social network, or an existing edge is strengthened.
- Extract data about the first 8 TV shows from JSON-file “tvshows.json”
- Create the class Actor to store data about an actor of TV show: actorID, actorName, actorBirthday, actorCharacterName, actorShows.
actorShows – is a dictionary of show’s IDs (where the actor participated). Initially – empty. It stores TV show data as follows: keys are show’s IDs, values – TV shows dates (year of premiere).
Actor class has a method to fill the actorShows dictionary.
- Create the class ActorGraph to store and maintain a dictionary of actors (Actor), these are the nodes of our graph.
- Actor will in turn maintain a list of shows. Actor will also maintain a dictionary in which the keys are other Actors (actorIDs) instances, and the values indicate the weight of the relationship. This thus makes up the edges of our graph ActorGraph.
- The weight of the relationship between actors is calculated as follows:
If actors participated in the same show the default weight of the relation is 1. If actors were together in more than one TV show the weight increases appropriately (+1).
- Design appropriate methos to calculate the “influence” of each actor. “Influence” is the average of relationships weights.
- Build the graph of actors and their relationships.
- Visualize “Influences” of the top 5 the most “influential” actors by using Seaborn bar plot.
- Visualize the graph of actors and their relationships by using NetworkX.
- Customize the first version of the graph visualization by adding colors for nodes, colors and widths for edges according to “influence” of actors, and weights of relationships.
