pagerank_res.write\
.format('bigquery')\
.mode('overwrite')\
.option('table',pagerank_table)\
.save()
logging.info('Done!')
# .option("query", "MATCH (n:User) WHERE n.screen_name CONTAINS '' RETURN n.screen_name,n.id_str,n.description,n.followers_count")\
"""
Community Detection
Triangle indicates that two of a node's neighbours are also neighbours value of 1 would mean the user is a part of a triangle
"""
#
logging.info('Triangle count algorithm')
result = g.triangleCount()
triangle_count = result.sort("count", ascending=False).filter('count > 0')
#triangle_count.show()
#
#
# ## save triangle count results to BQ
#
triangle_count_table = 'project_data.triangle_count_results'
logging.info('Writing Triangle Count data to BQ')
#triangle_count.write.csv('triangle_count.csv')
triangle_count.write\
.format('bigquery') \
.mode('overwrite')\
.option('table',triangle_count_table)\
# +---+---+------------+
# | a| b| friend|
# | b| c| follow|
# | c| b| follow|
# | f| c| follow|
# | e| f| follow|
# | e| d| friend|
# | d| a| friend|
# | a| e| friend|
# +---+---+------------+
# Count Triangles
# To count triangles, we use the `GraphFrame.triangleCount()`
# method, which counts the number of triangles passing through
# each vertex in this graph.
results = graph.triangleCount()
results.show()
# +-----+---+-------+---+
# |count| id| name|age|
# +-----+---+-------+---+
# | 0| g| Gabby| 60|
# | 0| f| Fanny| 36|
# | 1| e| Esther| 32|
# | 1| d| David| 29|
# | 0| c|Charlie| 30|
# | 0| b| Bob| 36|
# | 1| a| Alice| 34|
# +-----+---+-------+---+
# To show only vertex ID and the number of triangles
# passing through each vertex, we can write:
