We use cookies to ensure you have the best browsing experience on our website. Please read our cookie policy for more information about how we use cookies.
  1. Prepare
  2. Artificial Intelligence
  3. Natural Language Processing
  4. T9 Predictions Based on Unigram Frequencies
  5. Discussions

T9 Predictions Based on Unigram Frequencies

Sort by

recency

|

7 Discussions

|

  • jonathan_r
     + 0 comments

    Currently the maximum score is 15 but the top submissions on the leaderboard all have 30 points. How is that possible?

  • jonathan_r
     + 0 comments

    The description is a bit ambiguous regarding dashes and apostrophes. No key can be used to input these characters, but the description explicitly says that these characters are valid. What I took away is that these characters should be ignored from the perspective of the prediction system but replicated in full when outputting the suggestions. This begs the question how would the user of the phose input these characters if predictive typing were disabled.

  • jitendarbafna
     + 2 comments

    Can somebody tell me how to approach this question. I am trying with dictionary but getting time out. Here have a look at my code and let me know if I can do any optimization? Or if I need to use another technique than please provide some resource related to that techinque.

    import re
def solve(s,loc,word):
    if loc == len(s):
        wo = ''.join(word);
        for key, value in dictionary.items():
            if len(key) >= len(s) and wo == key[:len(s)]:
                ans.append([value,key]);
    else:
        for i in range(len(mobile[s[loc]-2])):
            word.append(mobile[s[loc]-2][i]);
            solve(s,loc+1,word);
            word.pop();

mobile = ["abc","def","ghi","jkl","mno","pqrs","tuv","wxyz"];

file = open("t9Dictionary",'r');

dictionary = {}
ans = []
for line in file:
    dictionary[line.split("\n")[0]] = 0;
#print(len(dictionary));   
file = open("t9TextCorpus",'r');
for line in file:
    for i in line.split():
        kk = re.findall(r'^[a-zA-Z]*[a-zA-Z]$',i)
        if len(kk) > 0:
            if i in dictionary:
                dictionary[i] = dictionary[i]+1;

#print(dictionary["over"]);               
t = int(input());
for i in range(t):
    test = input().strip();
    test1 = [int(x) for x in test];
    if test1.count(1) > 0:
        print("No Suggestions");
    else:
        solve(test1,0,[]);
        ans.sort(reverse=True);
        if len(ans) == 0:
            print("No Suggestions");
        else:
            for i in range(5):    
                print(ans[i][1],end=';') 
            print("")
    ans.clear();

  • L3O
     + 1 comment

    In description mapping has been mentioned for 2-9 but not 1. If only 2-9 has to be considerred , implies word with character ' and - will never be the output. Anyone please clear the query. Thanks.

  • sc659987
     + 6 comments

    file not found ,