文件名称：30天精读MBA
文件大小：3.69MB
文件格式：EPUB
更新时间：2022-01-09 11:22:56
MBA
We design and experiment with an innovative way to automatically generate product profiles from Amazon reviews. Using NLP, we extract opinions from each review, clusters them by their orientation through an unsupervised learning (k-means). From these clustered opinions, we estimate the product profiling kernel θ and the pricing kernel λ. Finally, we optimize/update the word polarity by minimizing the prediction error (supervised learning). While the trained model perform only slightly better than random guessing, the interim outputs and the estimated parameters seems to provide useful information while showing a possibility for improvement.