fich = open("//home//idexlab//" + "TotalMeasuresBM25" + ".txt", "w")
fgraph = open("//home//idexlab//" + "GraphTotal" + ".txt", "w")
for field in dataset.fields:
logging.debug("Field = " + field)
if not knownAuthors:
dataset.learnModel()
dataset.overwriteVectoriser = False
dataset.overwriteModel = False
if not knownAuthors:
relAuthorsDocSimilarity, relAuthorsDocCitations = dataset.findSimilarDocuments(field)
else:
relAuthorsDocSimilarity, relAuthorsDocCitations = dataset.readKnownAuthors(field)
#Relevant experts with BM25 score
expertAuthors = dataset.readKnownAuthorsExpert(field)
relevantAuthors = set(relAuthorsDocSimilarity).union(set(relAuthorsDocCitations))
logging.debug("Total number of relevant authors : " + str(len(relevantAuthors)))
graph, authorIndexer = dataset.coauthorsGraph(field, relevantAuthors)
#f = open(dataset.getCoauthorsFilename(field),'rb') #Save the graph of co-authors
#storedlist=pickle.load(f)
#for i in storedlist:
#fgraph.write(str(i))
#fgraph.write("\n"+ "Author vertices")
#fgraph.write(str(authorIndexer.getIdDict()))
#fgraph.write(str(numpy.ones(graph.ecount())))
#print(i)
#fgraph.close()
