This project is aimed at proving the theory and algorithmic feasibility in Motif-based PageRank(github: https://github.com/HKUST-KnowComp/Motif-based-PageRank.git) We make use of pettingZoo to create the similar environment according to the description in the essay "Ranking Users in Social Networks with Higher-Order Structures" for collecting the data from this environment. Furthermore, we reproduce the PageRank Algorithm Code and use the data to test its feasibility.
Setting up the environment - Collecting data - Validating the algorithm - Project evaluation
Lucy Cui, Yi Li, Eric Tan, Akaley Weng
Akaley Weng analyses the thesis find out necessary and tell Eric Tan what kind of environment he needs to make. After making environment by pettingZoo, Eric Tan collect the necessary data and convert to special data forms. YI LI use the data to test the PageRank algorithm he reproduce to test the result and create the network graph and evaluation graph based on the data given by Eric Tan After getting results from the experiment，Lucy Cui analyses and evaluates the results of the experiment and make prepare the presentation.
graph.py: make use of the data collected from the environment based on petting Zoo
levels.csv: data collected from petting Zoo which decides the data level of agent
main.py: code for collecting data from environment
message_history: record the sender and reciver message which makes up the citation network
reformatted_message_history: transfer the format of message_history data into algorithm format
remi_test2/remi_test3/remi_test4: simulate the petting zoo environment
Lucy Cui: Analyse and evaluate the results of the experiment and prepare the presentation Yi Li: reproduce the PageRank algorithm，use the data collected provided to make different graphs Eric Tan: create the environment by pettingZoo, collect necessary data according to requirements Akaley Weng: analyses the thesis and find the necessary environmental elements
Pagerank essay can be found here: https://aaai.org/papers/11287-ranking-users-in -social-networks-with-higher-order-structures/ Our code is accessible by: https://github.com/tanqiyuaneric/CSPL-- \602006final/blob/main/main.ipynb Pagerank code is available here: https://github.com/HKUST-KnowComp/Motif-based-PageRank.git