Code

Code base of our KDD 2018 work on route recommendations for taxi drivers.

Code for our WWW 20 paper tiled "GraphGen: A Scalable Approach to Domain-agnostic Labeled Graph Generation".

Code for our SIGMOD 2019 paper tiled "Efficiently Answering Regular Simple Path Queries on Large Labeled Networks".

Executables for our WWW 2019 paper tiled "RAQ: Relationship-Aware Graph Querying in Large Networks".

Code for our WWW 2019 paper tiled "Beyond Shortest Paths: Route Recommendations for Ride-sharing".

Code for our fingerprint indexing technique called "RISC: Rapid Inverted-Index Based Search of Chemical Fingerprints". We believe this is one of the fastest indexing techniques for chemical fingerprints with dimensions of 2048 and above. Our tool is available for free. If you notice any dataset, where this claim is not true, please send us a note. We would like to study such datasets and explore strategies that can help us improve further.

Code for our Reaction Pathway Prediction algorithm published in Bioinformatics.

Executable and datasets for the SkyGraph paper published in VLDB 2017.

Code and datasets for our Influence Maximization benchmarking study, published in SIGMOD 2017.

Executables for RESLING, published in ICDM 2016.

Executables for MANTRA, published in KDD 2016.

Executables for COMMIT, published in SIGMOD 2015.

Executables for GeoScop algorithm published in ICDM 2015.

Code for the EDwP distance measure for trajectories. The work was published in ICDE 2015.

Code and dataset for Answering top-k representative queries in graph databases. The work was published in SIGMOD 2014.

Mining discriminative subgraphs from global-state networks. The tool and dataset for the work published in KDD 2013.

Answering top-k queries over a mixture of attractive and repulsive dimensions. The tool and dataset for the work published in PVLDB 2011

GraphSig - A significant subgraph mining tool: The tool mines statistically significant subgraphs from large graph databases. The statistical significance of a graph is quantified by measuring its p-value. The work has been published in ICDE'2009. The tool has been extended to classify graphs and finds application in analyzing chemical libraries. Significant subgraphs can be employed to construct chemical descriptors and can then be classified in the feature space. The work on classification has been published in ACS Journal of Chemical Information and Modeling.