/spam-prediction-model

Using the SMS Spam Collection Dataset from Kaggle. Extracted features using CountVectorizer, used a Logistic Regression pipeline and RandomForestClassifier pipeline and then cross-validation. Used a pre-trained word embedding model spaCy to compare cross-validation f1, recall and precision scores.

Primary LanguageJupyter Notebook

spam-prediction-model

Using the SMS Spam Collection Dataset from Kaggle. Extracted features using CountVectorizer, used a Logistic Regression pipeline and RandomForestClassifier pipeline and then cross-validation. Used a pre-trained word embedding model spaCy to compare cross-validation f1, recall and precision scores.