Add comments to each topic graph

data/topic_charts.json

@@ -149,7 +149,8 @@
                 ],
                 "pctdistance": 1.2,
                 "labeldistance": 1.5
-            }
+            },
+            "comment": "As shown in the graph above, over 20% of the documents are related to Business & Economics & Finance, which makes it the largest topic group in dataset. On the contrary, the group of Culture & Cultural geography contains the smallest number of documents among all topics."
         }
     ],
     [
@@ -304,7 +305,8 @@
             "subplots_adjust": {
                 "left": 0.37,
                 "right": 0.98
-            }
+            },
+            "comment": "In average, documents related to Shopping & Commodity have larger fraction of words corrected in lines."
         }
     ],
     [
@@ -459,7 +461,8 @@
             "subplots_adjust": {
                 "left": 0.37,
                 "right": 0.98
-            }
+            },
+            "comment": "Compared with other topics, Personal Development & Human Resources & Career in average contain more lines ending with ellipsis."
         }
     ],
     [
@@ -614,7 +617,8 @@
             "subplots_adjust": {
                 "left": 0.37,
                 "right": 0.98
-            }
+            },
+            "comment": "Shopping & Commodity related documents have higher percentage of lines starting with bullet point."
         }
     ],
     [
@@ -769,7 +773,8 @@
             "subplots_adjust": {
                 "left": 0.37,
                 "right": 0.98
-            }
+            },
+            "comment": "Personal Development & Human Resources & Career in average has more lines with toxic words."
         }
     ],
     [
@@ -924,7 +929,8 @@
             "subplots_adjust": {
                 "left": 0.37,
                 "right": 0.98
-            }
+            },
+            "comment": "Daily Life & Home & Lifestyle in average has more toxic words."
         }
     ],
     [
@@ -1079,7 +1085,8 @@
             "subplots_adjust": {
                 "left": 0.37,
                 "right": 0.98
-            }
+            },
+            "comment": "Documents in the topic of Personal Development & Human Resources & Career in average contain more words than other topics."
         }
     ],
     [
@@ -1234,7 +1241,8 @@
             "subplots_adjust": {
                 "left": 0.37,
                 "right": 0.98
-            }
+            },
+            "comment": "There is no significant variance in the average word length for different topic groups. However, Education related data contain longer words than others in general."
         }
     ],
     [
@@ -1389,7 +1397,8 @@
             "subplots_adjust": {
                 "left": 0.37,
                 "right": 0.98
-            }
+            },
+            "comment": "Documents in the topic of Personal Development & Human Resources & Career usually contain more sentences."
         }
     ],
     [
@@ -1544,7 +1553,8 @@
             "subplots_adjust": {
                 "left": 0.37,
                 "right": 0.98
-            }
+            },
+            "comment": "Documents related to Daily Life & Home & Lifestyle usually have higher percentage of symbols."
         }
     ],
     [
@@ -1699,7 +1709,8 @@
             "subplots_adjust": {
                 "left": 0.37,
                 "right": 0.98
-            }
+            },
+            "comment": "The fraction of words with alpha character seems to be relatively consistent across different topics."
         }
     ],
     [
@@ -1854,7 +1865,8 @@
             "subplots_adjust": {
                 "left": 0.37,
                 "right": 0.98
-            }
+            },
+            "comment": "Culture & Cultural geography contains more stop words in average."
         }
     ],
     [
@@ -2009,7 +2021,8 @@
             "subplots_adjust": {
                 "left": 0.37,
                 "right": 0.98
-            }
+            },
+            "comment": "Natural Science & Formal Science & Technology has a significantly higher rate in percentage of documents that contain curly bracket. It might be related to the coding data."
         }
     ],
     [
@@ -2164,7 +2177,8 @@
             "subplots_adjust": {
                 "left": 0.37,
                 "right": 0.98
-            }
+            },
+            "comment": "Sports related documents have a higher number of duplication count."
         }
     ],
     [
@@ -2319,7 +2333,8 @@
             "subplots_adjust": {
                 "left": 0.37,
                 "right": 0.98
-            }
+            },
+            "comment": "In average, Culture & Cultural geography related documents are duplicated across a higher number of common crawl dumps. Duplication of Shopping & Commodity appears in less dumps than others."
         }
     ],
     [
@@ -2474,7 +2489,8 @@
             "subplots_adjust": {
                 "left": 0.37,
                 "right": 0.98
-            }
+            },
+            "comment": "In average, Culture & Cultural geography related documents are duplicated across more years than other topics."
         }
     ],
     [
@@ -2629,7 +2645,8 @@
             "subplots_adjust": {
                 "left": 0.37,
                 "right": 0.98
-            }
+            },
+            "comment": "In average, Culture & Cultural geography related documents are duplicated across a wider span of years."
         }
     ],
     [
@@ -2784,7 +2801,8 @@
             "subplots_adjust": {
                 "left": 0.37,
                 "right": 0.98
-            }
+            },
+            "comment": "Average language scores of different topic groups are mostly consistent. No significant differences are obeserved."
         }
     ],
     [
@@ -2939,7 +2957,8 @@
             "subplots_adjust": {
                 "left": 0.37,
                 "right": 0.98
-            }
+            },
+            "comment": "In average, Shopping & Commodity has a larger fraction of duplicate lines than others."
         }
     ],
     [
@@ -3094,7 +3113,8 @@
             "subplots_adjust": {
                 "left": 0.37,
                 "right": 0.98
-            }
+            },
+            "comment": "Shopping & Commodity usually has a larger fraction of characters in duplicate lines than others."
         }
     ],
     [

results.py

@@ -990,7 +990,7 @@ for title, data in topic_charts:
                 for rgb in data["kwargs"]["color"]
             ]
         )))
-    else:
+    elif data["type"] == "pie":
         topic_graphs.append(go.Figure(go.Pie(
             values=data["kwargs"]['x'],
             labels=data["kwargs"]["labels"],
@@ -1014,8 +1014,8 @@ cluster_div = Div(
         )),
         H3("Results Analysis"),
         *(
-            Section(H4(title), plotly2fasthtml(topic_graphs[i]))
-            for i, (title, _) in enumerate(topic_charts)
+            Section(H4(title), plotly2fasthtml(topic_graphs[i]), P(data.get("comment", '')))
+            for i, (title, data) in enumerate(topic_charts)
         )
     )
 )