Tasks:

The part of speech tagging training and testing were carried out using the 1998 People's Daily corpus.

Job input:

In 1998, the People's Daily Corpus (1998-01-105-tape.txt) used 80% data as training set and 20% data as verification set.

Operating environment:

Jupyter Notebook, Python3

Method of operation:

A simple statistical method is used to predict the part of speech of a word. N-gram language rules are not yet used.

Operational steps:

1. Processing corpus: Delete the pre-paragraph label.

# Read the original corpus file
in_path = '1998-01-105-Tape.txt'
file = open(in_path, encoding='gbk')
in_data = file.readlines()

# Pre-processed corpus
curpus_path = 'curpus.txt'
curpusfile = open(curpus_path, 'w', encoding='utf-8')

#Delete pre-paragraph labels, [], {}
for sentence in in_data:
    words = sentence.strip().split(' ')
    words.pop(0)
    
    for word in words:
        if word.strip() != '':
            if word.startswith('['):
                word = word[1:]
            elif ']' in word:
                word = word[0:word.index(']')]

            w_c = word.split('/')
            # Generating Corpus
            if(len(w_c) > 1):
                curpusfile.write(w_c[0] + ' ' + w_c[1] + '\n')

2. Random partition of training set 80% and verification set 20%.

from sklearn.model_selection import train_test_split

# Random partition
curpus = open(curpus_path, encoding='utf-8').readlines()
train_data, test_data = train_test_split(
    curpus, test_size=0.2, random_state=10)

# View the partitioned data size
print(len(curpus))
print(len(train_data) / len(curpus))
print(len(test_data) / len(curpus))

1114419
0.7999998205342874
0.20000017946571264

3. Statistical training set word frequency.

# Generating Word Frequency Recording Files
from tqdm import tqdm_notebook

doc = []

for sentence in tqdm_notebook(train_data):
    words = sentence.strip().split(' ')
    if len(words) > 1:
        temp = []
        temp.append(words[0])
        temp.append(words[1])
        flag = False
        for line in doc:
            if line[0] == temp[0] and line[1] == temp[1]:
                line[2] += 1
                flag = True
                break
        if not flag:
            temp.append(1)
            doc.append(temp)

4. Choose the most probable part of speech.

# Save the verification set
test_path = 'test.txt'
testfile = open(test_path, 'w', encoding='utf-8')
for sentence in test_data:
    words = sentence.strip().split(' ')
    if len(words) > 1:
        testfile.write(sentence)

# Save label results
result_path = 'result.txt'
resultfile = open(result_path, 'w', encoding='utf-8')

# Select the most probable part of speech for tagging
for sentence in tqdm_notebook(test_data):
    words = sentence.strip().split(' ')
    if len(words) > 1:
        words[1] = 'n'
        max = 0
        for line in doc:
            if line[0] == words[0] and line[2] > max:
                max = line[2]
                words[1] = line[1]
        resultfile.write(words[0] + ' ' + word[1] + '\n')

Performance evaluation: accuracy

def get_word(path):
    f = open(path, 'r', encoding='utf-8')
    lines = f.readlines()
    return lines

result_lines = get_word(result_path)
test_lines = get_word(test_path)

list_num = len(test_lines)
right_num = 0

for i in range(0, list_num):
    if result_lines[i][1] == test_lines[i][1]:
        right_num += 1

print("Accuracy:", right_num / list_num)

Accuracy: 0.23189316857201872