def create_tag_cloud(prm, filename_json):
if prm < 0:
prm_least_flag = True
else:
prm_least_flag = False
jsonData = format_json.loadjson(filename_json)
taglist = []
for key in jsonData.keys():
taglist.append(jsonData[key]["tags"])
tag_count = collections.Counter(itertools.chain.from_iterable(taglist))
deltemp = tag_count.most_common(1)
tag_count.pop(deltemp[0][0])
G = nx.Graph()
if prm == 0:
temp_most_common = tag_count.most_common()
str_prm = "all"
else:
if prm_least_flag:
temp_most_common = tag_count.most_common()[prm:]
else:
temp_most_common = tag_count.most_common(prm)
str_prm = str(prm)
G.add_nodes_from([(tag, {
"count": count
}) for tag, count in temp_most_common])
G = add_edges(G, taglist)
define_plt_pram(G)
filename_plot = get_filename_savefig(filename_json) + "_CLOUD_" + str_prm
plot_save_nx(G, filename_plot)
def loadjson(filename):
list_dict_results = []
dict_json = format_json.loadjson(filename)
if dict_json == -1:
print("error: Not Found file. Create new file")
if ".json" not in filename:
filename += ".json"
savejson(list_dict_results, filename)
else:
for key in dict_json.keys():
list_dict_results.append(dict_json[key])
return list_dict_results
def all_tags_list(filename_json):
taglist = []
jsonData = format_json.loadjson(filename_json)
for key in jsonData.keys():
taglist.append(jsonData[key]["tags"])
tag_count = collections.Counter(itertools.chain.from_iterable(taglist))
tag_most = tag_count.most_common()
result = []
for e in tag_most:
result.append(e[0])
return result
def create_year_map(filename_json):
jsonData = format_json.loadjson(filename_json)
taglist = []
for key in jsonData.keys():
temp = jsonData[key]["tags"]
year = jsonData[key]["year"][-4:]
if year != "None":
temp.append(int(year))
taglist.append(temp)
min_year = 3000
dict_taglist = {}
for e in taglist:
if e[-1] in dict_taglist.keys():
temp = dict_taglist[e[-1]]
set_temp = set(e[1:-1])
temp = temp | set_temp
dict_taglist[e[-1]] = temp
else:
temp = set(e[1:-1])
dict_taglist[e[-1]] = temp
if min_year > e[-1]:
min_year = e[-1]
tag_count = collections.Counter(itertools.chain.from_iterable(taglist))
main_tag = tag_count.most_common(2)[1][0]
G = nx.Graph()
all_tags = all_tags_list(filename_json)
temp = []
for e in dict_taglist.keys():
tags = dict_taglist[e]
for t in tags:
temp.append((str(e) + "_" + t, {"year": e, "tag": t}))
G.add_nodes_from(temp)
plt.figure(figsize=(50, 50))
#pos = nx.spring_layout(G, k=1.5)
pos = {}
for e in temp:
pos[e[0]] = ((e[1]["year"] - min_year) * 1000,
all_tags.index(e[1]["tag"]) * 100)
nx.draw_networkx_nodes(G,
pos,
node_color='b',
alpha=0.2,
node_size=30,
font_weight="bold",
font_family='VL Gothic')
nx.draw_networkx_labels(G, pos, fontsize=9)
keys = list(dict_taglist.keys())
b_tags = dict_taglist[keys[0]]
for i in range(1, len(keys)):
a_tags = dict_taglist[keys[i]]
same_tags = b_tags & a_tags
for s in same_tags:
G.add_edge(str(keys[i - 1]) + "_" + s,
str(keys[i]) + "_" + s,
weight=1)
b_tags = a_tags
nx.draw_networkx_edges(G, pos, alpha=0.4, edge_color='r', width=2)
filename_plot = get_filename_savefig(filename_json)
plot_save_nx(G, filename_plot)
for key in jsondata.keys():
sentence = jsondata[key]["summary"]
token = nltk.word_tokenize(sentence)
text = nltk.Text(token)
fdist = nltk.FreqDist(w.lower() for w in text
if w.lower() not in stopwords + symbols)
tagged = nltk.pos_tag(fdist)
temp = nltk.FreqDist(
stemmer.stem(w[0]) for w in tagged if w[1] not in exclude_list)
jsondata[key]["tags"] = temp
return jsondata
# ---- ----
# main
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument('Filename',
metavar='F',
type=str,
nargs='+',
help='Filename for saving data')
args = parser.parse_args()
FILENAME_JSON = args.Filename[0]
jsondata = format_json.loadjson(FILENAME_JSON)
jsondata = auto_tagging(jsondata)
format_json.savejson_dict(jsondata, FILENAME_JSON)