About
I'm a graduate student in Computer Science at San Jose State University, pursuing the passion for Machine Learning, Web Development and etc. Before coming to SJSU, I recieved my master degree at Columbia University in 2015. I received my bachelor degree from Southeast University (Nanjing, China) in 2013.
And I'm blessed with a fabulous family. Me and my husband are the proud parents of two kids. In my free time, I enjoy hiking, cooking, painting and traveling.
- Skills: Machine Learning, NLP, TensorFlow, Pandas, Data Analysis and Database, Multi-threading Programming
- Languages: C/C++, Python, Java, JavaScript, MySQL, HTML5, CSS, PHP
- Systems: Linux, OS X
Blogs
This section is to keep track of some of my tech blogs and stories of my personal life.
1. Literature Review, 12/10/2020
I did a literature survey based on object detection (R-CNN, Fast R-CNN, Faster R-CNN, Mask R-CNN). This review begins with a brief introduction on the four fundamental visual recognition problems in the field of computer vision and compare traditional methods with deep learning based methods. Then, I mainly focus on object detection architectures: R-CNN, Fast R-CNN, Faster R-CNN and one instance segmentation architecture: Mask R-CNN. In each section, I introduce model workflow, then summarize the achievements and analyze the disadvantages. Finally, I compare these architectures and write down some of my thoughts.
2. Movie reviews sentiment analysis with Naive bayesian classifier, 11/2021
I implemented a Naïve Bayes classifier in Python with picked features(based on mutual information) and Add-one smoothing from scratch and did not use off-the-shelf software. Then I used Pandas to read all data into a data frame (tokenize reviews to a list of words, remove stopwords, stem words), then split the data into training set and test set. The most interesting part would be training model: calculate log prior and log likelihood for picked features. And then test model: predict test set reviews’ sentiment and improve the model.