Post Job Free
Sign in

Data Analysis, Machine Learning, SQL, Python

Location:
Worcester, MA
Posted:
July 12, 2020

Contact this candidate

Resume:

Weinan Zhi

*** ********* **** **** ********* MA *1602

508-***-**** adeivc@r.postjobfree.com https://github.com/WeinanZhi Education:

Worcester Polytechnic Institute (WPI), Worcester, MA Master of Science in Data Science, GPA: 3.63/4.0, May 2020 Courses: Deep Learning, Non-linear dimension Reduction Renmin University of China (RUC), Beijing, China

Bachelor of Management in Information Management and Information System, GPA: 3.17/4.0, June 2016 Skills:

Python, SQL, Machine Learning, Data Analysis, R, XML, Tableau, Java, Bayesian statistics Projects:

Plant Seed Prediction, WPI, August-December 2019

Implemented transfer learning of ResNet18/DenseNet121 pretrained layers to classify created balanced dataset of segmented and unsegmented plant seed images. Reached 87.6%/92.4% classification accuracy on average.

Pheumonia Detection on Chest X-ray, WPI, August-December 2019

Drastically reduced the dimension of Chest X-ray images by ISOMAP and Laplacian Eigenmaps algorithm to solve a classification problem on Pneumonia. Multiple layer perceptron and logistic regression models were applied as evaluation. Both of the two classifiers got 0.86 test F1 score. GoogleMap Image Geolocation, WPI, January-April 2019

Implemented transfer learning of ResNet50/VGG16 pretrained layers to geolocate the created balanced dataset of 5000 satellite images generated from Google Maps static API. Reached 45.6%/42.8% classification accuracy on country level.

Indexing Improvement on SimpleDB, WPI, August-December 2018

Optimized range query by implementing R Tree Algorithm.

Optimized hash index in SimpleDB by a linear hash index.

Designed ten query plans to evaluate the improvement performance. Experience:

GQP Contractor, Mapfre Insurance company, Feb 2020-May 2020

Time Series Analysis on Microsoft Office data. The project adopted ARIMA models as baseline models, PROPHET models or RNN models as improved models on Application Usage prediction despite the similar test errors on different indicators of each Application.

Employees Persona Clustering for improving communication efficiency. Used Activity Log and Detailed personal information to cluster employees into target personas in order to improve communication efficiency.

CEO Assistant, Shanghai Yutai Investment CO.LTD., August 2016-April 2018

Constructed the official website of the company and fully authorized to communicate with the outsourcing company with the potential launching IOS application. Organization:

Member, Mensa International, May 2015-Present



Contact this candidate