In [1]:
import graphlab
In [2]:
products = graphlab.SFrame('amazon_baby.gl/')
In [3]:
graphlab.canvas.set_target('ipynb')
In [4]:
products.head()
Out[4]:
In [6]:
products['word_count'] = graphlab.text_analytics.count_words(products['review'])
In [7]:
products.head()
Out[7]:
In [8]:
products['name'].show()
In [9]:
giraffe_reviews = products[products['name'] == 'Vulli Sophie the Giraffe Teether']
In [10]:
len(giraffe_reviews)
Out[10]:
In [11]:
giraffe_reviews['rating'].show(view='Categorical')
In [12]:
products['rating'].show(view='Categorical')
In [13]:
#ignore all 3 star reviews
In [14]:
products = products[products['rating'] != 3]
In [15]:
len(products)
Out[15]:
In [16]:
#positive sentiment is 4/5 star ; negative sentiment is 1/2 star
In [17]:
products['sentiment'] = products['rating'] >= 4
In [18]:
products.head()
Out[18]:
In [19]:
train_data, test_data = products.random_split(0.8, seed=0)
In [22]:
sentiment_model = graphlab.logistic_classifier.create(train_data,
target='sentiment',
features=['word_count'],
validation_set=test_data)
In [23]:
sentiment_model.evaluate(test_data, metric='roc_curve')
Out[23]:
In [24]:
sentiment_model.show(view='Evaluation')
In [25]:
giraffe_reviews['predicted_sentiment'] = sentiment_model.predict(giraffe_reviews, output_type='probability')
In [26]:
giraffe_reviews.head()
Out[26]:
In [28]:
giraffe_reviews = giraffe_reviews.sort(
'predicted_sentiment',
ascending=False
)
In [29]:
giraffe_reviews.head()
Out[29]:
In [30]:
giraffe_reviews[0]['review']
Out[30]:
In [31]:
giraffe_reviews[1]['review']
Out[31]:
In [32]:
giraffe_reviews[-1]['review']
Out[32]:
In [33]:
giraffe_reviews[-2]['review']
Out[33]:
In [ ]: