Research Article
Ensemble Machine Learning Model for Classification of Spam Product Reviews
Table 5
Ten best features selected by Information Gain for Yelp Dataset.
| S. no | Features | Description |
| 1 | rev_rating | Reviews rating | 2 | stdev_revApp_rating | Standard deviation of review rating and rating application | 3 | stdev_revrating_avgrevratingapp | Standard deviation of review rating and average review rating application | 4 | avg_cosine_similarity_text | Average cosine similarity in review text | 5 | polarity_text | Polarity of review text | 6 | rev_pos_ascend | Reviews part of speech in ascending order | 7 | rev_pos_descend | Reviews part of speech in descending order | 8 | avg_levenshtein_dist_text | Average Levenshtein distance between reviews text | 9 | automated_readability_index_text | Automated readability index (ARI) of review body | 10 | avg_num_letters_per_word | Average number of letters per word in review body |
|
|