Create a ML model
Budget: $10 – $30 CAD
Description:
Sentiment Analysis is a common NLP task that data scientists need to
perform. In this project, we want to do sentiment analysis on IMDB
movie reviews to figure out if a comment is positive or negative.
To do so, you need to create a model using either traditional
classification models or deep learning algorithms, to predict the number
of positive and negative reviews.
Your training accuracy should not be less than 87%, and the difference
between training and test accuracy should not exceed 2.5%.
By creating and implementing such a model, you would prevent
overfitting your model. You may use plotting to check whether your
model is overfitted (optional).
Note: you cannot predict whether your model has overfitted by
defining a difference threshold between your training and testing
accuracy. In this model, we experimented that this difference
threshold may show overfitting in the model.
Dataset:
IMDB dataset having 50K movie reviews for natural language
processing or Text analytics. This is a dataset for binary sentiment classification. Dataset provides a set of 25,000 highly popular movie
reviews for training and 25,000 for testing.
For more dataset information and also access to the dataset, please go
through the following link:
https://ai.stanford.edu/~amaas/data/sentiment/
To Deliver:
Upload the following on the D2L:
1. A report that explains why you chose your model(traditional
classification or deep learning) and your results. Attach your plots and
add an explanation if you are also using plotting. Please also reference
any sources you use.
2. Your code. (you need to provide the code in 2 sections, 1 for training
and 1 for testing.)
Hint:
1. You can use the Keras imdb dataset. You can use only 10,000 words
for your model input. The following code helps you to do so:
You also need to vectorize your text inputs. To do so, you can use the
following method
Sentiment Analysis is a common NLP task that data scientists need to
perform. In this project, we want to do sentiment analysis on IMDB
movie reviews to figure out if a comment is positive or negative.
To do so, you need to create a model using either traditional
classification models or deep learning algorithms, to predict the number
of positive and negative reviews.
Your training accuracy should not be less than 87%, and the difference
between training and test accuracy should not exceed 2.5%.
By creating and implementing such a model, you would prevent
overfitting your model. You may use plotting to check whether your
model is overfitted (optional).
Note: you cannot predict whether your model has overfitted by
defining a difference threshold between your training and testing
accuracy. In this model, we experimented that this difference
threshold may show overfitting in the model.
Dataset:
IMDB dataset having 50K movie reviews for natural language
processing or Text analytics. This is a dataset for binary sentiment classification. Dataset provides a set of 25,000 highly popular movie
reviews for training and 25,000 for testing.
For more dataset information and also access to the dataset, please go
through the following link:
https://ai.stanford.edu/~amaas/data/sentiment/
To Deliver:
Upload the following on the D2L:
1. A report that explains why you chose your model(traditional
classification or deep learning) and your results. Attach your plots and
add an explanation if you are also using plotting. Please also reference
any sources you use.
2. Your code. (you need to provide the code in 2 sections, 1 for training
and 1 for testing.)
Hint:
1. You can use the Keras imdb dataset. You can use only 10,000 words
for your model input. The following code helps you to do so:
You also need to vectorize your text inputs. To do so, you can use the
following method