Wolfram Data Repository

Immediate Computable Access to Curated Contributed Data

Sample Data: Movie Review Sentence Polarity

Movie review data

This dataset consists of 10,662 snippets of movie reviews obtained from the review aggregator Rotten Tomatoes. Each review was labeled either positive or negative based on whether Rotten Tomatoes gave the movie a Fresh or Rotten rating, respectively. The test and training sets were constructed by stratified random sampling using 30% of the data for the test set and the rest for the training set.

Examples

Basic Examples

Retrieve the resource:

In[1]:=
ResourceObject["Sample Data: Movie Review Sentence Polarity"]
Out[1]=

Retrieve a sample of the dataset:

In[2]:=
RandomSample[ ResourceData["Sample Data: Movie Review Sentence Polarity"], 5]
Out[2]=

Analysis

Train a classifier:

In[3]:=
classifier = Classify[ResourceData["Sample Data: Movie Review Sentence Polarity", "TrainingData"]]
Out[3]=

Obtain general information about the classifier:

In[4]:=
Information[classifier]
Out[4]=

Visualize the accuracy of the classifier on the test dataset:

In[5]:=
ClassifierMeasurements[classifier, ResourceData["Sample Data: Movie Review Sentence Polarity", "TestData"], "ConfusionMatrixPlot"]
Out[5]=

Wolfram Research, "Sample Data: Movie Review Sentence Polarity" from the Wolfram Data Repository (2019)  

Data Resource History

  • Date Created:

Source Metadata

  • Creator: Bo Pang, Lillian Lee
  • Date: 2005
  • Language: English
  • Citation: "Seeing Stars: Exploiting Class Relationships for Sentiment Categorization with Respect to Rating Scales." Proceedings of the ACL, 2005.

Data Downloads

Publisher Information

  • Publisher of Record: Wolfram Research