python中的循环索引问题

Question

I'm having trouble indexing a result in a list.

The function "Similar" calculates the similarity between two strings in a range from 0 to 1. (Example: similar('dino','bino') = 0.75)

What I want to do is to iterate all items in each sublist of x with all items in list y. And find the domain with the highest similarity for each sublist of x.

My expected output would be:

['smart phones', NaN, 'fruits']

Here's my code so far:

x = [['phones', 'galaxy samsung', 'iphone'],[],['fruit', 'food']] ##each sublist refers to one user
y = ['fruits', 'smart phones', 'fishing', 'cars']                 ##domains

point = [0] * len(x)
best_dom = ['n'] * len(x)

for list in x:
i=0
  for query in list:
    for dom in y:
        sim = similar(query,dom)
        if sim > point[i]:
            point[i] = sim
            best_dom[i] = dom
i = i+1

print(best_dom)

But this is the output I'm getting:

['fruits', 'n', 'n']

Answer 1

My solution script consist of two parts:

1. Generating a list similar to x with its sublists.
   These hold tuples according to (<item>, <similarity_value>) storing the item and its highest similarity value to any item in y .
2. Getting the max-value of each sublist or 'NaN' if the sublist is empty.

-> afterwards printing the output list

# this will  be similar to x but storing the similarity_value with alongside each item
similarity_values = []

# iterating over x to generate values of similarity_values
for i, sublist in enumerate(x):
    similarity_values.append([])  # creating new sublist inside
    for item in sublist:  # iterating over each item in a sublist
        similarity_values[i].append((item, similar(item)))

# outputting the max values
output_list = []
for sublist in similarity_values:
    if sublist == []:  # 'NaN' if list is empty
        output_list.append('NaN')
    else:  # otherwise item with highest similarity_value
        output_list.append(max(sublist, key=lambda e: e[1])[0])
        # key for max() is set to the second element of the tuple

print(output_list)

I hope this will solve your problem!

Answer 2

I got it! I was not addressing the index in the correct way.

x = [['phones', 'galaxy samsung', 'iphone'],[],['fruit', 'food']] ##each sublist refers to one user
y = ['fruits', 'smart phones', 'fishing', 'cars']                 ##domains

point = [0] * len(x)
best_dom = ['n'] * len(x)

for user in x:
  for query in user:
    for dom in y:
      i = x.index(user)
      sim = similar(query,dom)
      if sim > point[i]:
            point[i] = sim
            best_dom[i] = dom
                
print(best_dom)