自然语言理解

浅浅的花香味﹌ 2021-09-18 11:02 542阅读 0赞

声明：代码的运行环境为Python3。Python3与Python2在一些细节上会有所不同，希望广大读者注意。本博客以代码为主，代码中会有详细的注释。相关文章将会发布在我的个人博客专栏[《Python自然语言处理》][Python]，欢迎大家关注。

--------------------

有了之前的学习之后，就可以进行自然语言的理解分析了，一起来看看吧~

**一、基于SQL数据库的问答系统**

1、首先输入一个特征文法

import nltk
    
    # 基于sql数据库的问答系统
    nltk.data.show_cfg('grammars/book_grammars/sql0.fcfg')  # 输入一个特征文法

输出结果为：

% start S
    S[SEM=(?np + WHERE + ?vp)] -> NP[SEM=?np] VP[SEM=?vp]
    VP[SEM=(?v + ?pp)] -> IV[SEM=?v] PP[SEM=?pp]
    VP[SEM=(?v + ?ap)] -> IV[SEM=?v] AP[SEM=?ap]
    NP[SEM=(?det + ?n)] -> Det[SEM=?det] N[SEM=?n]
    PP[SEM=(?p + ?np)] -> P[SEM=?p] NP[SEM=?np]
    AP[SEM=?pp] -> A[SEM=?a] PP[SEM=?pp]
    NP[SEM='Country="greece"'] -> 'Greece'
    NP[SEM='Country="china"'] -> 'China'
    Det[SEM='SELECT'] -> 'Which' | 'What'
    N[SEM='City FROM city_table'] -> 'cities'
    IV[SEM=''] -> 'are'
    A[SEM=''] -> 'located'
    P[SEM=''] -> 'in'

2、将询问的语句转换为一个可用于数据库查询的SQL语句

from nltk import load_parser
    # 将正常的询问语句转化为正常的SQL语句
    cp = load_parser('grammars/book_grammars/sql0.fcfg')
    query = 'What cities are located in China'
    trees = list(cp.parse(query.split()))
    answer = trees[0].label()['SEM']
    answer = [s for s in answer if s]
    q = ' '.join(answer)
    print(q)

输出结果为：

SELECT City FROM city_table WHERE Country="china"

3、测试

from nltk.sem import chat80
    
    rows = chat80.sql_query('corpora/city_database/city.db', q)
    for r in rows:
        print(r[0], " ")

输出结果为：

canton  
    chungking  
    dairen  
    harbin  
    kowloon  
    mukden  
    peking  
    shanghai  
    sian  
    tientsin

这里有必要说明一下，这个输出结果为粤语的拼音，故而感觉起来有点怪怪的~由上例可以看出，只要我们定义了一个文法，我们就可以将一个问题提取转换成一个正常的可用于数据库查询的SQL语句。

**二、命题逻辑**

1、使用boolean\_ops()方法进行表示

nltk.boolean_ops()
    
    negation       	-
    conjunction    	&
    disjunction    	|
    implication    	->
    equivalence    	<->

运算符为true的条件如下图所示：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2dka3l4eTIwMTM_size_16_color_FFFFFF_t_70][]

2、测试

read_expr = nltk.sem.Expression.fromstring
    read_expr('-(P & Q)')
    read_expr('P & Q')
    read_expr('P | (R -> Q)')
    read_expr('P <-> -- P')

最后一个的结果：

测试：

lp = nltk.sem.Expression.fromstring
    SnF = read_expr('SnF')
    NotFnS = read_expr('-FnS')
    R = read_expr('SnF -> -FnS')
    prover = nltk.Prover9()
    prover.config_prover9(r'D:\Program Files\LADR1007B-win\bin')
    prover.prove(NotFnS, [SnF, R])
    
    val = nltk.Valuation([('P', True), ('Q', True), ('R', False)])
    print(val['P'])

结果：

True

dom = set()
    g = nltk.Assignment(dom)
    m = nltk.Model(dom, val)
    print(m.evaluate('(P & Q)', g))
    print(m.evaluate('-(P & Q)', g))
    print(m.evaluate('(P & R)', g))
    print(m.evaluate('(P | R)', g))
    
    True
    False
    False
    True

**三、一阶逻辑**

# 一阶逻辑
    read_expr = nltk.sem.Expression.fromstring
    expr = read_expr('walk(angus)', type_check=True)
    print(expr.argument)
    print(expr.argument.type)
    print(expr.function)
    print(expr.function.type)

结果为：

angus
    e
    walk
    <e,?>

sig = {'walk': '<e, t>'}
    expr = read_expr('walk(angus)', signature=sig)
    print(expr.function.type)
    
    read_expr = nltk.sem.Expression.fromstring
    read_expr('dog(cyril)').free()
    read_expr('dog(x)').free()
    read_expr('own(angus, cyril)').free()
    read_expr('exists x.dog(x)').free()
    read_expr('((some x. walk(x)) -> sing(x))').free()
    read_expr('exists x.own(y, x)').free()

真值模型：

# 真值模型
    dom = {'b', 'o', 'c'}
    v = """
    bertie => b
    olive => o
    cyril => c
    boy => {b}
    girl => {o}
    dog => {c}
    walk => {o, c}
    see => {(b, o), (c, b), (o, c)}
    """
    val = nltk.Valuation.fromstring(v)
    print(val)
    print(('o', 'c') in val['see'])
    print(('b',) in val['boy'])

运算：

# 运算
    g = nltk.Assignment(dom, [('x', 'o'), ('y', 'c')])
    print(g)
    
    m = nltk.Model(dom, val)
    m.evaluate('see(olive, y)', g)
    print(g['y'])
    
    m.evaluate('see(y, x)', g)
    g.purge()
    print(g)
    m.evaluate('see(olive, y)', g)
    m.evaluate('see(bertie, olive) & boy(bertie) & -walk(bertie)', g)

**四、量化**

m.evaluate('exists x.(girl(x) & walk(x))', g)
    
    m.evaluate('girl(x) & walk(x)', g.add('x', 'o'))
    
    fmla1 = read_expr('girl(x) | boy(x)')
    m.satisfiers(fmla1, 'x', g)
    fmla2 = read_expr('girl(x) -> walk(x)')
    m.satisfiers(fmla2, 'x', g)
    fmla3 = read_expr('walk(x) -> girl(x)')
    m.satisfiers(fmla3, 'x', g)
    
    m.evaluate('all x.(girl(x) -> walk(x))', g)

**五、量化歧义**

v2 = """
    bruce => b
    elspeth => e
    julia => j
    matthew => m
    person => {b, e, j, m}
    admire => {(j, b), (b, b), (m, e), (e, m)}
    """
    val2 = nltk.Valuation.fromstring(v2)
    
    dom2 = val2.domain
    m2 = nltk.Model(dom2, val2)
    g2 = nltk.Assignment(dom2)
    fmla4 = read_expr('(person(x) -> exists y.(person(y) & admire(x, y)))')
    m2.satisfiers(fmla4, 'x', g2)
    
    fmla5 = read_expr('(person(y) & all x.(person(x) -> admire(x, y)))')
    m2.satisfiers(fmla5, 'y', g2)
    
    fmla6 = read_expr('(person(y) & all x.((x = bruce | x = julia) -> admire(x, y)))')
    m2.satisfiers(fmla6, 'y', g2)

**六、句子语义理解**

read_expr = nltk.sem.Expression.fromstring
    expr = read_expr(r'\x.(walk(x) & chew_gum(x))')
    print(expr)
    
    expr.free()
    
    print(read_expr(r'\x.(walk(x) & chew_gum(y))'))
    
    expr = read_expr(r'\x.(walk(x) & chew_gum(x))(gerald)')
    print(expr)
    print(expr.simplify())
    
    print(read_expr(r'\x.\y.(dog(x) & own(y, x))(cyril)').simplify())
    print(read_expr(r'\x y.(dog(x) & own(y, x))(cyril, angus)').simplify())
    
    expr1 = read_expr('exists x.P(x)')
    print(expr1)
    
    expr2 = expr1.alpha_convert(nltk.sem.Variable('z'))
    print(expr2)
    
    print(expr1 == expr2)
    
    expr3 = read_expr('\P.(exists x.P(x))(\y.see(y, x))')
    print(expr3)
    
    print(expr3.simplify())

**七、量词歧义**

from nltk.sem import cooper_storage as cs
    
    sentence = 'every girl chases a dog'
    trees = cs.parse_with_bindops(sentence, grammar='grammars/book_grammars/storage.fcfg')
    semrep = trees[0].label()['SEM']
    cs_semrep = cs.CooperStore(semrep)
    print(cs_semrep.core)
    
    for bo in cs_semrep.store:
        print(bo)
    
    cs_semrep.s_retrieve(trace=True)
    
    for reading in cs_semrep.readings:
        print(reading)

**八、段落语义理解**

read_dexpr = nltk.sem.DrtExpression.fromstring
    drs1 = read_dexpr('([x, y], [angus(x), dog(y), own(x, y)])')
    print(drs1)
    
    drs1.draw()
    
    print(drs1.fol())
    
    drs2 = read_dexpr('([x], [walk(x)]) + ([y], [run(y)])')
    print(drs2)
    
    print(drs2.simplify())
    
    drs3 = read_dexpr('([], [(([x], [dog(x)]) -> ([y],[ankle(y), bite(x, y)]))])')
    print(drs3.fol())
    
    drs4 = read_dexpr('([x, y], [angus(x), dog(y), own(x, y)])')
    drs5 = read_dexpr('([u, z], [PRO(u), irene(z), bite(u, z)])')
    drs6 = drs4 + drs5
    print(drs6.simplify())
    
    print(drs6.simplify().resolve_anaphora())
    
    from nltk import load_parser
    
    parser = load_parser('grammars/book_grammars/drt.fcfg', logic_parser=nltk.sem.drt.DrtParser())
    trees = list(parser.parse('Angus owns a dog'.split()))
    print(trees[0].label()['SEM'].simplify())

[Python]: https://blog.csdn.net/gdkyxy2013/column/info/30938
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2dka3l4eTIwMTM_size_16_color_FFFFFF_t_70]: /images/20210726/58c054ea000e4a61a9a6253999ccd87c.png