- Download the gremlin server and console from https://tinkerpop.apache.org/downloads.html
- add graphml file to
path_to_server_folder/data - add path of final graphml file here
path_to_server_folder/scripts/school-sample.groovy - add path of script file in conf folder
path_to_server_folder/conf/gremlin-server-school.yamlin ScriptEngines value
apache-tinkerpop-gremlin-server-3.4.10/bin/
./gremlin-server.sh console
A graph can be created using create_graph(self, nodes, edges) function of the class Import.
- Parameters of create_graph : list of nodes and edges between the nodes of the graph you want to create.
- G = nx.Graph() : Using NetwrokX module imported as nx, nx.Graph() creates a empty graph instance of name G.
- G.add_nodes_from(nodes) : This command adds Nodes from the list 'nodes' to the graph instance G.
- G.add_edges_from(edges) : Adds edges between the nodes specified in the 'edges' list.
- returns the graph instance G to the called block.
Example calling Function Block:
nodes = [(1, {'labelV': 'node1', 'node_id': 'INT'}), (2, {'labelV': 'node2', 'node_id':
'INT'})]
edges = [(1,2, {'labelV' : 'has'})]
create_graph(self,nodes,edges)
Creates a graph instance with 2 node and 1 edge
2. Created graph is converted into graphml file using nx.write_graphml(filename) which is a method in networkx lib.
- join_graphs(graph1,graph2) is a method of the class Algo()
- Parameters : Two graph instances graph1 and graph2.
- Returns : Combined graph with bridge nodes.
- Description : Visits each nodes in the two graph and similarity between each nodes of the two graphs is found with the help of wordnet from nltk libraray, and bridge node is added between those two nodes with suitable similarity range. The bridge node is the lowest comman hypernym of the two node.
4. Returned graph isntance from join_graph() is converted into graphml file using nx.write_graphml().
g = traversal().withRemote(DriverRemoteConnection('ws:https://localhost:8182/gremlin','g'))
- A GraphTraversalSource is created from the anonymous traversal() method where the "g" provided to the DriverRemoteConnection corresponds to the name of a GraphTraversalSource on the remote end.
- Note : Whenever changes are made in the local Knowledge graph, server must be re-started so that the changes are applied to the graph at remote end. This graph at the remote end is the main Knowledge Graph.