Albert Guo Email & Phone Number

Shenzhen, Guangdong, China

Use data analysis methods to help the Industrial and Commercial Bank of China improve the insufficient product penetration for corporate customers.Use data analysis methods to help ICBC improve the reach rate of corporate customers.

New York, United States

Preprocess the credit card raw dataset containing 500,000 rows and 11 columns.Perform feature engineering on the dataset according to RFM principle (Recency, Frequency and Monetary value) combined with credit card scenario, and generate 15 effective feature factors.Use ECOD, PCA, Autoencoder and iTree algorithms to model 15 feature factors respectively and find out potential credit card default records.Remind credit card company of potential default users and risk management.

New York, United States

New York, United States

Conduct Porter’s Five Forces analysis on the cement industry where Argos is located, to understand the external and internal environment of the industry.Preprocess Argos’ raw dataset containing 2 million rows and 100 columns. Classify users by their purchase amount and purchase frequency, and use Tableau to display user information on a map.Use K Means clustering algorithm and K-prototype clustering algorithm to classify users. By drawing silhouette score, we choose to set K as 9, then group customers and propose different business plans according to the characteristics of customers in each group.

New York, United States

Use the New York Times API to obtain the New York Times Best Sellers data, and save the data to the MongoDB database in Json file format. Get the keywords entered by the user (book title, author name, genre) from the front-end web page, and transfer the keywords to MongoDB. According to the clustering algorithm to calculate the similarity to recommend the most suitable books for this user.Return recommendation results to user by using Flask library in python.

New York, United States

Collect data from 8602+ housing customers in 91 dimensions (geographical location, evaluation, housing area, number of bedrooms, etc.) from the Airbnb database, and build Machine Learning Models in R to Predict the Housing Prices.Compare the RMSE between the prediction results and the test set data in the logistic regression model, the decision tree model and the XGBoost model to select the most suitable model parameters and the optimal model.Finally select the XGBoost model as the optimal model, and prediction results ranked ninth.

United States

Henan, China

1. Collect CSI 300 index data by the API interface on Jupyter Notebook.2. Collect investors’ comments on the stock market on the different investment platforms. Do natural language processing to the comments, such as performing word segmentation by Jieba library and stop word processing. Create positive sentiment thesaurus and negative sentiment thesaurus, and calculate the investors’ sentiment. Then do a correlation analysis between the sentiment index and CSI 300 index.3. Collect stock index data and use PCA (Principal Component Analysis) to select the stock index with the greatest contribution.4. Based on the above stock indicators, use the K-Mean clustering method to screen out high-quality stock. 5. Construct the selected stocks into an investment portfolio, use SVM to predict the future trend of the portfolio, and form a quantitative timing strategy to obtain an excess return of 5% higher than the CSI 300 Index.

Master Of Science - Ms, Financial Engineering, 89/100

Activities and Societies: Financial Market and Data Financial Econometrics Derivative Pricing Stochastic Modeling Machine Learning.

Master Of Science - Ms, Computer Science, General

Bachelor Of Technology - Btech, Computer Science

Bachelor'S Degree, Finance, General(Minor)

Bachelor'S Degree, Computer Science

Activities and Societies: Student Union Learning Department Vice President of Chess Association

Secondary School, Science Department