app/difficulty.py

@@ -7,7 +7,7 @@
 
 import pickle
 import math
-from wordfreqCMD import remove_punctuation, freq, sort_in_descending_order, sort_in_ascending_order, map_percentages_to_levels
+from wordfreqCMD import remove_punctuation, freq, sort_in_descending_order, sort_in_ascending_order
 import snowballstemmer
 
 
@@ -94,58 +94,30 @@ def revert_dict(d):
     return d2
 
 
-def user_difficulty_level(d_user, d, calc_func=0):
+def user_difficulty_level(d_user, d):
-    '''
-    two ways to calculate difficulty_level
-    set calc_func!=0 to use sqrt, otherwise use weighted average
-    '''
-    if calc_func != 0:
-        #  calculation function 1: sqrt
-        d_user2 = revert_dict(d_user)  # key is date, and value is a list of words added in that date
-        geometric = 0
-        count = 0
-        for date in sorted(d_user2.keys(),
-                           reverse=True):  # most recently added words are more important while determining user's level
-            lst = d_user2[date]  # a list of words
-            lst2 = []  # a list of tuples, (word, difficulty level)
-            for word in lst:
-                if word in d:
-                    lst2.append((word, d[word]))
-
-            lst3 = sort_in_ascending_order(lst2)  # easiest tuple first
-            # print(lst3)
-            for t in lst3:
-                word = t[0]
-                hard = t[1]
-                # print('WORD %s HARD %4.2f' % (word, hard))
-                geometric = geometric + math.log(hard)
-                count += 1
-        return math.exp(geometric / max(count, 1))
-
-    #  calculation function 2: weighted average
     d_user2 = revert_dict(d_user)  # key is date, and value is a list of words added in that date
-    count = {}  # number of all kinds of words
+    count = 0
-    percentages = {}  # percentages of all kinds of difficulties
+    geometric = 1
-    total = 0  # total words
+    for date in sorted(d_user2.keys(),
-    for date in d_user2.keys():
+                       reverse=True):  # most recently added words are more important while determining user's level
         lst = d_user2[date]  # a list of words
+        lst2 = []  # a list of tuples, (word, difficulty level)
         for word in lst:
             if word in d:
-                if d[word] not in count:
+                lst2.append((word, d[word]))
-                    count[d[word]] = 0
-                count[d[word]] += 1
-                total += 1
 
-    if total == 0:
+        lst3 = sort_in_ascending_order(lst2)  # easiest tuple first
-        return 1
+        # print(lst3)
-    for k in count.keys():
+        for t in lst3:
-        percentages[k] = count[k] / total
+            word = t[0]
-    weight = map_percentages_to_levels(percentages)
+            hard = t[1]
-    sum = 0
+            # print('WORD %s HARD %4.2f' % (word, hard))
-    for k in weight.keys():
+            geometric = geometric * (hard)
-        sum += weight[k] * k
+            count += 1
-    return sum
+            if count >= 10:
+                return geometric ** (1 / count)
 
+    return geometric ** (1 / max(count, 1))
 
 
 def text_difficulty_level(s, d):

app/templates/userpage_get.html

@@ -73,7 +73,6 @@
 	<button type="button" class="btn-close" data-bs-dismiss="alert" aria-label="Close"></button>
     </div>
     {% endfor %}
-
         <div class="pagination">
           <button class="arrow" id="load_pre_article" onclick="load_pre_article();Reader.stopRead()" title="Previous Article">
             <i class="fas fa-chevron-left"></i> 上一篇

app/wordfreqCMD.py

@@ -10,32 +10,6 @@ import operator
 import os, sys # 引入模块sys，因为我要用里面的sys.argv列表中的信息来读取命令行参数。
 import pickle_idea
 
-
-def map_percentages_to_levels(percentages):
-    '''
-    功能：按照加权平均难度，给生词本计算难度分，计算权重的规则是(10 - 该词汇难度) * 该难度词汇占总词汇的比例，再进行归一化处理
-    输入：难度占比字典，键代表难度3~8，值代表每种难度的单词的占比
-    输出：权重字典，键代表难度3~8，值代表每种难度的单词的权重
-    '''
-    # 已排序的键
-    sorted_keys = sorted(percentages.keys())
-
-    # 计算权重和权重总和
-    sum = 0  # 总和
-    levels_proportions = {}
-    for k in sorted_keys:
-        levels_proportions[k] = 10 - k
-    for k in sorted_keys:
-        levels_proportions[k] *= percentages[k]
-        sum += levels_proportions[k]
-
-    # 归一化权重到权重总和为1
-    for k in sorted_keys:
-        levels_proportions[k] /= sum
-
-    return levels_proportions
-
-
 def freq(fruit):
     '''
     功能： 把字符串转成列表。 目的是得到每个单词的频率。