Predict starts people give to a product, using exist data like summary, text, product ID, User ID...
This project is a in class project in Kaggle.
To run code, please place train.csv under the main directory as main.py
Requirement: python3.5 gensim tensorflow(optional)
Structure:
- main.py
- classes: utils, dictionary, manipulate_comment_file, machien_learning_classifier, normal_classifiers
- Classifier.py
- classes: classifier(Interface class), classifier_statistic, classifier_product_average, classifier_semantic_analysis_highest, classifier_semantic_analysis_average
- ProjectReport.docx
- A report about this project. More details, but don't copy this directly to yours.
- Submission and History.zip
- A submission and history record. Use mean root-square error to evaluate it.
- Result.jpg
- Record all method I tried and score it get. The lower the better.
