N-Gram，tf-idf和余弦相似度在Python中的简单实现

问题描述

我需要比较存储在数据库中的文档，并得出0到1之间的相似度分数.

I need to compare documents stored in a DB and come up with a similarity score between 0 and 1.

我需要使用的方法必须非常简单.实现n-gram的原始版本(可以定义要使用的克数)以及tf-idf和余弦相似度的简单实现.

The method I need to use has to be very simple. Implementing a vanilla version of n-grams (where it possible to define how many grams to use), along with a simple implementation of tf-idf and Cosine similarity.

是否有任何程序可以做到这一点?还是应该从头开始编写?

Is there any program that can do this? Or should I start writing this from scratch?

推荐答案

签出NLTK软件包: http://www.nltk .org 它具有您所需的一切

Check out NLTK package: http://www.nltk.org it has everything what you need

对于余弦相似度:

def cosine_distance(u, v):
    """
    Returns the cosine of the angle between vectors v and u. This is equal to
    u.v / |u||v|.
    """
    return numpy.dot(u, v) / (math.sqrt(numpy.dot(u, u)) * math.sqrt(numpy.dot(v, v)))

对于ngram:

def ngrams(sequence, n, pad_left=False, pad_right=False, pad_symbol=None):
    """
    A utility that produces a sequence of ngrams from a sequence of items.
    For example:

    >>> ngrams([1,2,3,4,5], 3)
    [(1, 2, 3), (2, 3, 4), (3, 4, 5)]

    Use ingram for an iterator version of this function.  Set pad_left
    or pad_right to true in order to get additional ngrams:

    >>> ngrams([1,2,3,4,5], 2, pad_right=True)
    [(1, 2), (2, 3), (3, 4), (4, 5), (5, None)]

    @param sequence: the source data to be converted into ngrams
    @type sequence: C{sequence} or C{iterator}
    @param n: the degree of the ngrams
    @type n: C{int}
    @param pad_left: whether the ngrams should be left-padded
    @type pad_left: C{boolean}
    @param pad_right: whether the ngrams should be right-padded
    @type pad_right: C{boolean}
    @param pad_symbol: the symbol to use for padding (default is None)
    @type pad_symbol: C{any}
    @return: The ngrams
    @rtype: C{list} of C{tuple}s
    """

    if pad_left:
        sequence = chain((pad_symbol,) * (n-1), sequence)
    if pad_right:
        sequence = chain(sequence, (pad_symbol,) * (n-1))
    sequence = list(sequence)

    count = max(0, len(sequence) - n + 1)
    return [tuple(sequence[i:i+n]) for i in range(count)]

对于tf-idf，您必须首先计算分布，我正在使用Lucene来做到这一点，但是您可能会使用NLTK做类似的事情，请使用FreqDist:

for tf-idf you will have to compute distribution first, I am using Lucene to do that, but you may very well do something similar with NLTK, use FreqDist:

http://nltk.googlecode.com/svn/trunk/doc/book/ch01.html#frequency_distribution_index_term

如果您喜欢pylucene，这将告诉您如何上下班tf.idf

if you like pylucene, this will tell you how to comute tf.idf

    # reader = lucene.IndexReader(FSDirectory.open(index_loc))
    docs = reader.numDocs()
    for i in xrange(docs):
        tfv = reader.getTermFreqVector(i, fieldname)
        if tfv:
            rec = {}
            terms = tfv.getTerms()
            frequencies = tfv.getTermFrequencies()
            for (t,f,x) in zip(terms,frequencies,xrange(maxtokensperdoc)):
                    df= searcher.docFreq(Term(fieldname, t)) # number of docs with the given term
                        tmap.setdefault(t, len(tmap))
                        rec[t] = sim.tf(f) * sim.idf(df, max_doc)  #compute TF.IDF
            # and normalize the values using cosine normalization
            if cosine_normalization:
                denom = sum([x**2 for x in rec.values()])**0.5
                for k,v in rec.items():
                    rec[k] = v / denom

这篇关于N-Gram，tf-idf和余弦相似度在Python中的简单实现的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持！