Dynamic Network Embedding

• Btech S8 Project (2019)

System Specifications

• Python 3.6
• Operating Systems Tested : Windows 10 or Ubuntu 18.04 LTS

Packages

• Babel 2.6.0
• backcall 0.1.0
• conda 4.6.8
• conda-build 3.17.6
• Cython 0.29.2
• h5py 2.8.0
• Keras <= 2.2.4
• Keras-Applications 1.0.7
• Keras-Preprocessing 1.0.9
• networkx 1.11 (Only this Version)
• scikit-learn 0.20.1
• scipy 1.1.0
• seaborn 0.9.0
• six 1.12.0
• tensorboard 1.13.1
• tensorflow 1.13.1
• tensorflow-estimator 1.13.0
• Theano 1.0.3+2.g3e47d39ac.dirty
• tqdm 4.28.1

Running Dynamic Embedding Network Embedding

• Pickling Dataset Snapshots (To make it ready to load for embedding) $ python datasets/<name_of_dataset>/graph_pickler.py

• Performing Dynamic Embedding for a Specific Dataset $ python datasets/<name_of_dataset>/dyngem.py

• Performing Link Prediction on Embedded Graph $ python datasets/<name_of_dataset>/link_prediction.py

Dataset Plots

Dynamic Graph Embedding Architecture

Performance Comparisons with SDNE

Parameter Sensitivity