Random walk scaling on networks
Aditya Mohapatra
Random_walk_scaling_on_networks
Overview
This project implements a Markov Chain scaling analysis in Python, focusing on the computation of path probabilities and their scaling behaviors. The analysis includes polynomial and exponential scaling methods. This was implemented as a course project for the courese PH5490 at IIT Madras. Link to the refered paper is: https://www.nature.com/articles/ncomms6121
Key Features
Path Checking: Utilizes breadth-first search (BFS) to identify paths from the start state to the target state.
Node Filtering: Removes nodes that are not involved in the identified paths.
Graph Analysis: Identifies strongly connected components and classifies them as acyclic, monocyclic, or multicyclic.
Scaling Types
Polynomial Scaling: This method computes the growth of paths in polynomial terms.
Exponential Scaling: This method analyzes the growth of paths in exponential terms.
Python Packages Used
NumPy: For numerical operations and handling arrays.
SciPy: For sparse matrix operations and linear algebra.
NetworkX: For graph-based computations and pathfinding.
Matplotlib: For visualizing results through plots.
Example Usage
To use this implementation, you can define a transition matrix for a simple Markov chain and specify start and target states. The scaling analysis can then be computed based on these parameters.
Like this project
Posted Dec 23, 2024
Analysis of random walks on different structures of networks. It used graph algorithms like Breadth Fast Search (BFS), Detection of SCC.
Likes
0
Views
5
