Web-Logs-Unsupervised-and-Supervised-Learning-Association-Rule-Mining-ARIMA-Prediction

Web Logs Data Unsupervised, Supervised Learning, Association Rule Mining & ARIMA Prediction. Web Crawling of citation information from Google Scholar.

Part I - Data Analytics — Web Log Data

1. Data ETL

• Load Data
• Feature Selection

2. Unsupervised learning
3. Supervised learning

• Data Preparation
• Logistic Regression
• K-fold Cross Validation

4. Association Rule Mining

Part II - Web Crawling

5. Crawl the professor Gang Li citation information from 2003 to 2021
6. Train Arima to predict the 2018 to 2020 citation

• Train Arima Model
• Predicting the citation and Calculate the RMSE
• Draw the visualization to compare

7. Conduct the Grid Search with parameter selection and then predict the 2021 and 2022

• Grid Search
• Select the best parameter values and Predict for 2021 and 2022