seantino

Use NumPy, PyTorch and Sox package in Python to preprocess and normalize the Samsung Bixby corpus data sets for NLP models training. Preprocessing the corpus dataset before the Fourier transforms decoding to ensure the process successful.

Proposed a novel XGBoost + Random Forest solution, I divided the tasks to my group members, collaborated with them to complete the new solution in a limited time. Use K-means clustering algorithm combined with Silhouette Coefficient to describe and analyze the value of user portraits of game products.

Design queries and procedures in MySQL Database. Carried out secondary research, including investigation on the industry and filtering of relevant companies, arranged the data in the given format.