We will be using the u_mass and c_v coherence for two different LDA models: a "good" and a "bad" LDA model. The good LDA model will be trained over 50 iterations and the bad one for 1 iteration. Hence in theory, the good LDA model will be able come up with better or more human-understandable topics. Therefore the coherence measure output for the good LDA model should be more (better) than that for the bad LDA model. This is because, simply, the good LDA model usually comes up with better topics that are more human interpretable.
In [1]:
import numpy as np
import logging
import pyLDAvis.gensim
import json
import warnings
warnings.filterwarnings('ignore') # To ignore all warnings that arise here to enhance clarity
from gensim.models.coherencemodel import CoherenceModel
from gensim.models.ldamodel import LdaModel
from gensim.models.wrappers import LdaVowpalWabbit, LdaMallet
from gensim.corpora.dictionary import Dictionary
from numpy import array
In [2]:
logger = logging.getLogger()
logger.setLevel(logging.DEBUG)
logging.debug("test")
As stated in table 2 from this paper, this corpus essentially has two classes of documents. First five are about human-computer interaction and the other four are about graphs. We will be setting up two LDA models. One with 50 iterations of training and the other with just 1. Hence the one with 50 iterations ("better" model) should be able to capture this underlying pattern of the corpus better than the "bad" LDA model. Therefore, in theory, our topic coherence for the good LDA model should be greater than the one for the bad LDA model.
In [3]:
texts = [['human', 'interface', 'computer'],
['survey', 'user', 'computer', 'system', 'response', 'time'],
['eps', 'user', 'interface', 'system'],
['system', 'human', 'system', 'eps'],
['user', 'response', 'time'],
['trees'],
['graph', 'trees'],
['graph', 'minors', 'trees'],
['graph', 'minors', 'survey']]
In [4]:
dictionary = Dictionary(texts)
corpus = [dictionary.doc2bow(text) for text in texts]
We'll be setting up two different LDA Topic models. A good one and bad one. To build a "good" topic model, we'll simply train it using more iterations than the bad one. Therefore the u_mass coherence should in theory be better for the good model than the bad one since it would be producing more "human-interpretable" topics.
In [5]:
goodLdaModel = LdaModel(corpus=corpus, id2word=dictionary, iterations=50, num_topics=2)
badLdaModel = LdaModel(corpus=corpus, id2word=dictionary, iterations=1, num_topics=2)
In [14]:
goodcm = CoherenceModel(model=goodLdaModel, corpus=corpus, dictionary=dictionary, coherence='u_mass')
In [15]:
badcm = CoherenceModel(model=badLdaModel, corpus=corpus, dictionary=dictionary, coherence='u_mass')
Following are the pipeline parameters for u_mass coherence. By pipeline parameters, we mean the functions being used to calculate segmentation, probability estimation, confirmation measure and aggregation as shown in figure 1 in this paper.
In [16]:
print goodcm
As we will see below using LDA visualization, the better model comes up with two topics composed of the following words:
Therefore, the topic coherence for the goodLdaModel should be greater for this than the badLdaModel since the topics it comes up with are more human-interpretable. We will see this using u_mass and c_v topic coherence measures.
In [17]:
pyLDAvis.enable_notebook()
In [18]:
pyLDAvis.gensim.prepare(goodLdaModel, corpus, dictionary)
Out[18]:
In [19]:
pyLDAvis.gensim.prepare(badLdaModel, corpus, dictionary)
Out[19]:
In [20]:
print goodcm.get_coherence()
In [21]:
print badcm.get_coherence()
In [25]:
goodcm = CoherenceModel(model=goodLdaModel, texts=texts, dictionary=dictionary, coherence='c_v')
In [26]:
badcm = CoherenceModel(model=badLdaModel, texts=texts, dictionary=dictionary, coherence='c_v')
In [27]:
print goodcm
In [28]:
print goodcm.get_coherence()
In [29]:
print badcm.get_coherence()
This API supports gensim's ldavowpalwabbit and ldamallet wrappers as input parameter to model.
In [5]:
model1 = LdaVowpalWabbit('/home/devashish/vw-8', corpus=corpus, num_topics=2, id2word=dictionary, passes=50)
model2 = LdaVowpalWabbit('/home/devashish/vw-8', corpus=corpus, num_topics=2, id2word=dictionary, passes=1)
In [7]:
cm1 = CoherenceModel(model=model1, corpus=corpus, coherence='u_mass')
cm2 = CoherenceModel(model=model2, corpus=corpus, coherence='u_mass')
In [8]:
print cm1.get_coherence()
print cm2.get_coherence()
In [20]:
model1 = LdaMallet('/home/devashish/mallet-2.0.8RC3/bin/mallet',corpus=corpus , num_topics=2, id2word=dictionary, iterations=50)
model2 = LdaMallet('/home/devashish/mallet-2.0.8RC3/bin/mallet',corpus=corpus , num_topics=2, id2word=dictionary, iterations=1)
In [21]:
cm1 = CoherenceModel(model=model1, texts=texts, coherence='c_v')
cm2 = CoherenceModel(model=model2, texts=texts, coherence='c_v')
In [22]:
print cm1.get_coherence()
print cm2.get_coherence()
Hence as we can see, the u_mass and c_v coherence for the good LDA model is much more (better) than that for the bad LDA model. This is because, simply, the good LDA model usually comes up with better topics that are more human interpretable. The badLdaModel however fails to decipher between these two topics and comes up with topics which are not clear to a human. The u_mass and c_v topic coherences capture this wonderfully by giving the interpretability of these topics a number as we can see above. Hence this coherence measure can be used to compare difference topic models based on their human-interpretability.