def show_similarity(gid_x, gid_y):
"""
"""
novel_node_x = show_cluster_node(gid_x)
novel_node_y = show_cluster_node(gid_y)
cluster_similarity = NovelSimilarityModule()
similarity, match_list = cluster_similarity.novel_node_similarity_calculation(novel_node_x, novel_node_y)
print(similarity)
print(match_list)
def run(self):
"""
"""
self.logger.info('novel cluster edge module start')
similarity = NovelSimilarityModule()
process_gid_list = self.process_gid_collection()
for index, gid in enumerate(process_gid_list):
cluster_db = ClusterDBModule()
cluster_db.delete_novelclusteredgeinfo(gid)
cluster_node = self.cluster_node_collection(gid)
if not cluster_node:
continue
virtual_node = similarity.virtual_novel_node_generate(cluster_node)
related_gid_list = self.related_gid_collection(cluster_node)
if len(related_gid_list) == 0:
continue
book_name = cluster_node.book_name.encode('GBK')
pen_name = cluster_node.pen_name.encode('GBK')
self.logger.info('index: {0}/{1}'.format(index, len(process_gid_list)))
self.logger.info('novel_info: {0}@{1}@{2}, '
'chater_number: {3}, related_gid_number: {4}'.format(
gid, book_name, pen_name,
len(virtual_node.chapter_list), len(related_gid_list)
))
related_edge_list = []
for related_gid in related_gid_list:
related_cluster_node = self.cluster_node_collection(related_gid)
if not related_cluster_node:
continue
related_virtual_node = similarity.virtual_novel_node_generate(related_cluster_node)
cluster_similarity = similarity.novel_cluster_similarity_calculation(virtual_node, related_virtual_node)
if cluster_similarity >= 0.7:
cluster_edge = ClusterEdgeInfo(cluster_node.gid, related_cluster_node.gid, cluster_similarity)
related_edge_list.append(cluster_edge)
book_name = related_cluster_node.book_name.encode('GBK')
pen_name = related_cluster_node.pen_name.encode('GBK')
self.logger.info('novel_info: {0}@{1}@{2}, '
'chapter_number: {3}, similarity: {4}'.format(
related_gid, book_name, pen_name,
len(related_virtual_node.chapter_list), cluster_similarity
))
self.cluster_edge_update(gid, related_edge_list)
self.logger.info('novel cluster edge module end')
return True
def show_cluster_node(gid):
"""
"""
cluster_db = ClusterDBModule()
cluster_edge = ClusterEdgeModule()
cluster_similarity = NovelSimilarityModule()
cluster_node = cluster_edge.cluster_node_collection(gid)
virtual_novel_node = cluster_similarity.virtual_novel_node_generate(cluster_node)
book_name = cluster_node.book_name.encode('GBK', 'ignore')
pen_name = cluster_node.pen_name.encode('GBK', 'ignore')
print('gid: {0}, book_name: {1}, pen_name: {2}'.format(gid, book_name, pen_name))
print(', '.join('%s: %d' % (chapter.chapter_title.encode('GBK', 'ignore'), chapter.rank) for chapter in virtual_novel_node.chapter_list))
return virtual_novel_node
def run_test(self):
"""
跑评估数据
"""
gid_list = [int(line.strip()) for line in open('./data/rid.txt', 'r').readlines()]
similarity = NovelSimilarityModule()
for index, gid in enumerate(gid_list):
cluster_node = self.cluster_node_collection(gid)
if not cluster_node:
continue
print('gid: {0}, book_name: {1}, pen_name: {2}'.format(
gid,
cluster_node.book_name.encode('GBK', 'ignore'),
cluster_node.pen_name.encode('GBK', 'ignore')
))
novel_node = similarity.virtual_novel_node_generate(cluster_node)
print(', '.join('%s' % chapter.chapter_title.encode('GBK', 'ignore') for chapter in novel_node.chapter_list))
print('')
novel_node = get_novel_node(conn, dir_id)
id_list = []
for table_id in xrange(1, 256):
sql = 'SELECT dir_id FROM novel_cluster_dir_info{0} WHERE book_name = "{1}"'.format(table_id, novel_node.book_name.encode('GBK'))
cursor = conn.cursor()
cursor.execute(sql)
for (id, ) in cursor.fetchall():
id_list.append(id)
cursor.close()
if len(id_list) == 0:
continue
print_novel_node(novel_node, dir_id)
similarity = NovelSimilarityModule()
for id in id_list:
node = get_novel_node(conn, id)
print_novel_node(node, id)
print(similarity.novel_node_similarity_calculation(novel_node, node))
novel_node = node
conn.close()
here()
