INSTRUCTIONS: ————————————
- Open terminal at this folder location.
- Run command - “python solve_remedia.py”
DEPENDENCIES: ————————————
nltk numpy PyDictionary k-parser (output of k-parser is given in folder aspIO)
HOW TO READ LOG FILE: ————————————————————
Search Keywords in log file: “1st three answers:” -> To see answers at every question. “Checking in:” -> To see at which Rank a particular question is being answered. “-----Paragraph :” -> To see level and paragraph. 1st number is level, followed by paragraph number.
HOW TO READ RESULTS FILE (after execution): ——————————————————————————————————————————
AnswerMap is written in results file.txt in the same folder with three Ranks. Rank 1 being higher. Use these stats to generate percentages.
EXAMPLE: ———————
Per Paragraph:
2_1 : 4/5 : {1: {'What': 1, 'Where': 1}, 2: {'Who': 1, 'When': 1}}
2_1 -> level_paragraph
4/5 -> 4 questions answered correctly out of 5
1: {'What': 1, 'Where': 1} -> ’What’ and ‘where’ questions are answered by the Rank 1 sentence.
2: {'Who': 1, 'When': 1} -> ‘Who’ and ‘When’ questions are answered by the Rank 2 sentence.
Per Level:
2_P : 86/125 C1: 49 C2: 71 C3: 86
2_P -> All paragraphs in level 2.
86/125 -> 86 questions answered out of 125.
C1: 49 -> 49 questions answered correctly by Rank 1.
C2: 71 -> 71 questions answered correctly by Rank 1 and Rank 2.
C3: 86 -> 86 questions answered correctly by Rank 1 and Rank 2 and Rank 3.
Correct : {1: {'When': 15, 'What': 10, 'Who': 8, 'Where': 9, 'Why': 7},
2: {'Where': 9, 'What': 1, 'Who': 5, 'When': 3, 'Why': 4},
3: {'Where': 3, 'Who': 4, 'When': 7, 'Why': 1}}
{1: {'When': 15, 'What': 10, 'Who': 8, 'Where': 9, 'Why': 7} -> Rank 1 answered:
15 When questions, 10 What questions, 8 Who questions, 9 Where questions, 7 Why questions.
Similarly for Rank 2 and Rank 3.
Total Read: {'Where': 26, 'What': 23, 'Who': 26, 'When': 27, 'Why': 23}
Total no. of Where questions attempted : 26
Total no. of What questions attempted : 23
Total no. of Who questions attempted : 26
Total no. of When questions attempted : 27
Total no. of Why questions attempted : 23