Xiaoxu Lu

me

A second year master student at School of Computer Science, Carnegie Mellon University.

A software developer and a data science engineer.

  • Carnegie Mellon University

    2014.08 - Present

    GPA: 3.6/4.0

    Master of Science

    Language Technologies Institute

    School of Computer Science

    Graduate Teaching Assistant

    Coureses: Meachine Learning, Text Mining, Signal Processing, Big Data Analytics

  • Shandong University

    2010.09 - 2014.06

    GPA: 89/100

    Bachelor of Engineering

    Department of Computer Science and Technology

    Outstanding Graduate Student Award

    Courses: Data Structure, Computer Architecture, Operating System, Databases

  • University of California,

    Los Angeles

    2013.08 - 2014.09

    Exchange student for Summer Session

    Courses: Game Theory, Business Communication

  • National University of Singapore

    2012.08

    Summer Program on Economic and Enterprise Development


2015.06 - 2015.08

Global Merchant Repository Team, Data Platform Division

Implemented a merchant classifier based on transaction locations and customers frequently shopped zip codes.

Built features using MAD(Median Absolute Deviation), WGE(Weighted Gaussian Entropy), and DBSCAN(Density Based Spatial Clustering with Application to Noise) to train the classifier.

Explored how customers are dispersed regarding the merchant location.

Achieved an average accuracy of 92% by cross validation.

2015.11

Implemented a rating prediction system based on Yelp reviews, employing Multi-class Logistic Regression.

Applied Feature Hash Model on review texts and mini-Batch Stochastic Gradient Descent for Logistic Regression.

Achieved 60% on accuracy and 0.76 on Root Mean Square Error.

2015.04 - 2015.05

Implemented a rating prediction system based on Yelp reviews, employing Multi-class Logistic Regression.

Applied Feature Hash Model on review texts and mini-Batch Stochastic Gradient Descent for Logistic Regression.

Achieved 60% on accuracy and 0.76 on Root Mean Square Error.

2014.10 - 2015.05

Implemented a software that enables medical practitioners to practice skills in real-world scenario simulations.

Lead a group of four people responsible for Engine Package for Backend development.

Developed under Scrum and Agile methodology.

2015.03

Implemented a human face detection system employing AdaBoost algorithm.

Extracted Eigenfaces from LFWCrop database and built weak classifiers for AdaBoost utilizing Eigenfaces.

2014.10 - 2014.12

Implemented a biomedical question answering system on Apache UIMA framework.

Applied evidence-based retrieval and natural language processing technologies.

Built UIMA pipeline, retrieved related documents and snippets on queries built from given questions, extracted answers and evaluated all the results.