P4A - Social Network Interaction

P4A Social Network Interaction is a module for clustering large number of social networks’ data based on their common features. By encoding the
activities of the entities within the network as multi-modal and/or multi-parametric objects (the object-user may
have many attributes), and by modelling the social network data into k-partite graphs we manage to efficiently
categorize the entities into certain behavioral groups – provided that the positioning of the objects over the
graph is based on the relevance of their attributes.

The repository contains the below directories:

1. [GGraph] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/tree/master/Code/Visualizations/GGraph(link is external)): This includes the source code of the Social Network Interaction Module.
2. [Supplementary] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/tree/master/Supplementary(link is external)):
   

   1. [App] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/tree/master/Supplementary/App(link is external)) : This includes the source [code] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/blob/master/Supplementary/App/main.cpp(link is external)) of sample application that uses the Social Network Interaction Module and the [pro] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/blob/master/Supplementary/App/KPGraph.pro(link is external)) file.

   2. [Configuration Files] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/tree/master/Supplementary/Configuration%20Files(link is external)) : This includes two configuration files. The [configuration.ini] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/blob/master/Supplementary/Configuration%20Files/configuration.ini(link is external)) is used for the configuration of the Twitter's parsers, while the [parameters.ini] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/blob/master/Supplementary/Configuration%20Files/parameters.ini(link is external)) is responsible for configuring the sample application.

   3. [Twitter Parsers] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/tree/master/Supplementary/Twitter%20Parsers(link is external)) : This includes the source code of the twitter parsers, written in Python.
      
      i. [getTwitterTrendingTopicsPerCity.py] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/blob/master/Supplementary/Twitter%20Parsers/getTwitterTrendingTopicsPerCity.py(link is external)) : Twitter parser for collecting the trending topics for specified cities.
      
      Inputs: The following parameters are passed as inputs to the script via the [configuration.ini] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/blob/master/Supplementary/Configuration%20Files/configuration.ini(link is external))
      
      - User's authentication params
      
      - Cities : A list with the cities (separated with commas) that we are interested in collecting their trending topics.
      
      - Cities WOEID: A list with the corresponding cities' WOEID (separated with commas). You can find a city's WOEID [here] (http://woeid.rosselliot.co.nz/(link is external)).
      
      - Continent : A list with the corresponding cities' continent (separated with commas).
      
      Sample Output:

      ii. [getTweetsCity.py] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/blob/master/Supplementary/Twitter%20Parsers/getTweetsCity.py(link is external)) : Twitter parser for collecting tweets containing certain keywords around specified areas.
      
      Inputs: The following parameters are passed as inputs to the script via the [configuration.ini] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/blob/master/Supplementary/Configuration%20Files/configuration.ini(link is external))
      
      - User's authentication params
      
      - KeywordsList: A list with key-words (separated with comma), for searching tweets containing these key-words.
      
      - Cities : A list with the cities (separated with commas) that we are interested in collecting the relative tweets.
      
      - Cities_lat_lng : List with the latitude, longitude and the searching area for each of the corresponding cities, that we are interested in collecting the relative tweets.
      
      - lang : language
      
      - recordsPerCity : Max number of records per city
      
      Sample Output:

      iii. [getUsersIdsSpecificTweets_v4.py] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/blob/master/Supplementary/Twitter%20Parsers/getUsersIdsSpecificTweets_v4.py(link is external)) : Twitter parser for collecting the Following and the Followings for a number of users.
      
      Inputs: The following parameters are passed as inputs to the script via the [configuration.ini] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/blob/master/Supplementary/Configuration%20Files/configuration.ini(link is external))
      
      - User's authentication params
      
      - keyword: Searching users who have posted tweets containing the specified keyword.
      
      - users_limit : Max number of users
      
      Sample Output:

   4. [Sample Data] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/tree/master/Supplementary/SampleData(link is external)) contains:
      
      i. sample output data for each Twitter Parser.

      ii. [Results] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/tree/master/Supplementary/SampleData/Results(link is external)) : contains the final output of the 1st example. The directory contains two csv files describing topology of the graph ([graph_structure.csv] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/blob/master/Supplementary/SampleData/Results/graph_structure.csv(link is external)) and the [vertex_positions.csv] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/blob/master/Supplementary/SampleData/Results/vertex_positions.csv(link is external)) ) and an [image] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/blob/master/Supplementary/SampleData/Results/trending_topics.png(link is external)) view of the results.

Dependencies

The following libraries are necessary for the execution of Social Network Interaction module. Older subversions may also be compatible:

1. [Qt 5.5.1] (http://www.qt.io/download-open-source/(link is external)) : Used to build the sample App.

2. [Tweepy] (http://www.tweepy.org/(link is external)) : Used to access the Twitter API.

The Twitter parsers were tested in Python 3.4. The app was built and ran using [Qt Creator] (https://www.qt.io/download/(link is external)) 3.5.1

App usage

Step 1: Set the [configuration.ini] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/blob/master/Supplementary/Configuration%20Files/configuration.ini(link is external)) file

Step 2: Set the [parameters.ini] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/blob/master/Supplementary/Configuration%20Files/parameters.ini(link is external)) file and pass it as input to the app

1. Set the number of the example in order to to run the corresponding twitter parser (e.g. runExample=1: Runs the [1st Parser] (https://github.com/P4ALLcerthiti/P4ALL_Social_Network_Interaction/blob/master/Supplementary/Twitter%20Parsers/getTwitterTrendingTopicsPerCity.py(link is external)))

Funding Acknowledgement

The research leading to these results has received funding from the European Union's Seventh Framework Programme (FP7) under grant agreement No.610510