Michael Graczyk

  • Citations Per Year
Learn More
In this paper, we present a machine learning algorithm to classify product listings posted to anonymous marketplaces. We classify these listings according to the type of product being sold. The categories are derived from the 12 product categories on a popular anonymous marketplace, Agora. Our algorithm is a combination of TF-IDF for feature extraction, PCA(More)
Wang and Jiang recently proposed an end-to-end neural system solving the Stanford Question Answering Dataset (SQuAD) task [1]. The model consists of their textual entailment model called Match-LSTM, and a new sequence model for finding answers in context paragraphs. We reimplemented the model using Keras and Tensorflow [2][3]. Although we were not able to(More)
  • 1