Lacon classy

Run model

python ./src/cli.py NGRAM_SVC 'pas maca test moist cat'

Run pretrained svm model with n-grams as feature vector.

Available model with different feature vecor are shown in Model performance secion.

Model performance

Name	F1	Recall	Precision	Accuracy	False Positive	False Negative	True Positive	True Negative
NGRAM_SVC	0.928	0.913	0.944	0.983	0.006	0.168	0.832	0.994
HAND_ENGINEERED_SVC	0.778	0.715	0.910	0.959	0.005	0.565	0.435	0.995
HAND_ENGINEERED_AND_NGRAMS_SVC	0.930	0.916	0.944	0.983	0.006	0.162	0.838	0.994
NGRAM_ADABOOST	0.912	0.894	0.932	0.980	0.008	0.205	0.795	0.992
HAND_ENGINEERED_ADABOOST	0.746	0.680	0.924	0.956	0.003	0.638	0.362	0.997
HAND_ENGINEERED_AND_NGRAMS_ADABOOST	0.903	0.886	0.923	0.978	0.009	0.220	0.780	0.991
NGRAM_KNN	0.837	0.796	0.894	0.965	0.009	0.398	0.602	0.991
HAND_ENGINEERED_KNN	0.773	0.711	0.908	0.958	0.005	0.574	0.426	0.995
HAND_ENGINEERED_AND_NGRAMS_KNN	0.861	0.822	0.913	0.970	0.008	0.348	0.652	0.992