Data Analyst Python

Jiacheng Lu

************@*****.*** 408-***-**** San Mateo, CA

EXPERIENCE:

Rural Commercial Bank of Zhangjiagang

Data Analyst Intern

Suzhou, China

Jun 2016 – Sep 2017

• Collected, and analyzed data on established contact approaches that increased customer retention by 21% in CRM.

• Drove sales to upgrade with in-depth analysis on subscriptions and customer behaviors, impacting customer conversion and overall client experience. Built a customer churn model using logistic regression based on existed data warehouse.

• Increased renewal rate by 18% by leading client service relationships, helping target the right customers at the right time. East Star Mortgage

Data Analyst

San Gabriel, CA, US

Aug 2020 –Present

• Implemented ETL process for data from 1k+ data sources and demonstrated effective data integration by using SQL and Python data which collected from title company which we worked with and our own data warehouse.

• Provided support for executing mortgage or loan applications by assessing financial risks and ad campaign choosing.

• Identified 2K+ potential customers by defining key metrics of high-end customers through assisting senior leaders to understand the business and pulling and analyzing 160K+ records from multiple resources using SQL. EDUCATION:

Northeastern University, Silicon Valley

Master of Data Science GPA: 3.6/4.0

Silicon Valley, CA

Jan 2019 – May 2020

Relevant Coursework: Predictive Analytics, Data Mining, Data Visualization, Machine Learning, Data Modeling, Leadership in Analytics, Data Warehouse and SQL, Risk Management Analytics, Probability Theory & Statistics, Analytics Systems Technology Nanjing University of Finance and Economics

Bachelor of Management in finance GPA: 3.5/4.0

Nanjing, China

Sep 2014 – Jun 2018

Relevant Coursework: Accounting, Finance, Financial Management, Risk Management, Auditing, Financial Analytics Honors & Awards: Excellence Scholarship, Excellent advanced individual, Organized a school e-sports competition with sponsor PROJECTS:

Telecom customer Churn Prediction

Machine learning & Python

Silicon Valley, CA

Apr 2019 – May 2019

• Trialed different Machine Learning algorithms, such as Logistic Regression (with Lasso & Ridge), Decision Tree, KNN Classifier and Random Forest Classifier to predict potential customer churn. The dataset about Telco customers extracted from IBM.

• Dealt with the missing value and detected outliers using box plot then explored data by visualizing data using seaborn package.

• Selected key features using Lasso and improved the Lasso model using Gridsearch to tune the parameters. Then evaluated the performance using Confusion Matrix and ROC curve. The final F1 Score is 0.61 and the AUC is 0.82.

• Evaluated ROC performance on all models then found the random forest classifier performs best with AUC 0.861 among all. Data Visualization using COVID-19 data

Data Visualization & Python

Silicon Valley, CA

Mar 2020 – Apr 2020

• Scraped data from Twitter using a package called ‘GetOldTweets’ to make sure enough Tweets for analysis. Selected 10 cities which have a large population and number of confirmed cases such as New York City, Boston, Chicago, Los Angeles and so on.

• Did text preprocessing to improve the performance of NLP models. Used the package “Textblob” to assign each Tweet a sentiment score called “polarity” to measure how positive or how negative the review is. Then graphed data to see the trend.

• Used Latent Dirichlet Allocation via Mallet (LdaMallet) to extract topics from all Tweets and then combined the results from sentiment analysis and topic modeling to analyze which topics will contribute to people’s emotions.

• Supported suggestions to people who feel mentally discomfort, ex: Help others, Order food, Stop playing games all day Mobile Ad Campaigns Performance Analysis – Google Analysis Machine learning & Python

Silicon Valley, CA

Jan 2020 – Mar 2020

• Evaluated the performance of ads distributed by Mobiground and predicted the conversion with cost, budget, hour of day, impressions, CTR from Google Analytics for more efficient mobile distribution and identifying opportunities to improve ROI.

• Performed EDA on the dataset and created Tableau dashboards to understand Mobiground’s overall Ads performance.

• Tested multiple machine learning models and selected a stacking model including 26 ensemble models and 2 deep net models to predict CTR with a MAE of 3.92, a MSE of 71.27 and a R squared of 0.98.

• Improved the conversion rate by 4% via distributing mobile contents during identified peak usage time within the budget. SKILLS:

Skills: SQL, Python (Numpy, Pandas, Matplotlib, Seaborn, Scikit-Learn), R, Tableau, R Shinny, Spark, MS Office, Matlab Specialties: Data Analysis, Machine Learning, Predictive Modeling, Statistical Analysis, Interactive Data Visualization, Dashboard

Contact this candidate