Developed a ML model with 81% accuracy to improve the performance of e-mail marketing.

• Extracted important textual features of email titles by natural language processing (NLP).
• Applied SMOTE algorithm to solve the problem of data imbalance.
• Constructed the classification model to predict whether an email is spam or not before sending it.
• Concluded the word blacklist. Let marketing team avoid to use these words in the email title.
• The accuracy rate achieved 81% by optimizing the SVM model parameters through Grid search.

The dataset is not disclosed because the data involves the confidential information of the company.