Data Analysis Internship | Statistical data collecting, reorganizing and processing Resume - Hire IT People

TECHNICAL SKILLS

PROFESSIONAL EXPERIENCE

Confidential

Data Analysis Internship | Statistical data collecting, reorganizing and processing

Responsibilities:

Design insurance scheme according to customer’s requirements, determined the premium amount according to client’s number and job nature.
Collected and reorganized claim data and made statistical judgments
Created database based on Access, applied statistical theory to analyze and improve the effectiveness of formulation process for group insurance

Confidential

Big Data Analytics: Expedia Hotel Recommendations

Responsibilities:

Used R and SparkR to build and operate the algorithm for processing a dataset over 4 gigabytes from Kaggle website
Built recommender model using user searching statistics by calculating the distance between searching and booking record which constructed by a value of combination of variables and their formed weight based on Random Forest
Achieved a high accuracy (70%) when giving top 10 likely stay hotel for users

Confidential

Database and SQL: A Database and Website for an Online Second - hand Book Store

Responsibilities:

Designed ER-diagram and related relational schema that allow users create accounts, sell/buy books, and add reviews etc.
Mapped our E/R diagram, including constraints, into a SQL schema using PostgreSQL on Microsoft Azure
Programed in python and flask to build a web application and used python package SQLAlchemy to connect the database

Confidential

Computational Statistics with R: Predictive Model for 8 th Grade Students’ Science IRT Score

Responsibilities:

Trained a model to predict students’ science IRT score among 28 chosen interested independent variables in four aspects
Attained a set of predictors based on different methods, namely subset selection, lasso, ridge and random forest
Compared models by cross-validation and obtained the optimal model for future prediction

Confidential

The Comparison of Five Classification Methods by Monte Carlo Simulation

Responsibilities:

Simulated data using Monte Carlo method in two different scenarios, namely linear and non-linear Bayes ’ boundary in R
Compared Logistic Regression, LDA, QDA, 1-NN and 10-NN using cross-validation by mean squared prediction error

Confidential

Multivariate Analysis: Effects of Sibling Size on Students’ Academic Achievement

Responsibilities:

Applied MANOVA method for comparing multivariate dependent academic performance variables of students with different sibling size, verified the null assumption of evenly performance of different groups of students
Utilized discriminant analysis to further search in which direction did the difference of different type of students lies on

Confidential

Data Mining: Algorithm Implementation for Different Methods in R

Responsibilities:

Analyzed face image data by pixmap library in R using PCA and KNN methods to retrieve the characteristics of each image
Implemented algorithms of Naive Bayes, CART and Logistic Regression to identify author utilizing the Federalist Papers Dataset