User Behavior Insights standard schema specification

This is a total solution for user behavior tracking from collecting to analysing data.

Growth Hacking knowledge system（黑客增长知识体系）.

Gem that tracks suspicious activity within your user sessions.

Build your own analytics - A single library to grabs every click, touch, page-view, and fill — forever.

This repository contains scripts to carry out anomaly detection in the CLUE log data set.

A Convenient Field Study Toolkit (for our WWW21 paper Towards a Better Understanding of Query Reformulation Behavior in Web Search)

Our CIKM21 Paper "Incorporating Query Reformulating Behavior into Web Search Evaluation"

This dashboard presents a comprehensive behavioral analysis of DEX Traders and Liquidity Providers in the Avalanche ecosystem. Using Flipside’s Intelligence Driven Growth (IDG) methodology, it explores user journeys, engagement, platform usage, and liquidity contributions across major Avalanche DEXs from May 2024 to April 2025.

Custom BI Tool (WP Plugin) – "Mixpanel Tracker" developed for Product Analysts, UX Researchers, and Business Owners to track user behavior, gain business insights, and optimize product performance.

SENTINEL is an immersive insider threat detection and training platform designed for security analysts, SOC teams, and IT professionals. Featuring a fully simulated UEBA (User and Entity Behavior Analytics) environment, interactive threat simulations, and comprehensive indicators of compromise (IOCs) library🔒👨🏿‍💻.

Unsupervised ML system detecting anomalous user activity, deployed as a FastAPI service with Docker and Nginx.

Analyzed Citi Bike data to uncover hidden issues and trends in ride duration, peak usage, and station demand, providing recommendations to optimize bike availability.

Examples of analytics business projects I worked on as freelance consultant

Exploratory Data Analysis (EDA) on Fitbit users' data to uncover trends in activity and health metrics.

This analysis focuses on understanding user behavior and conversion performance in an e-commerce mobile application using event-level data.

This project employs XGBoost regression and XGBoost classifier model to predict user order and user churn on online travel agency data. Reach 97% prediction accuracy.

User behaviour analysis on a public bike-share dataset

This is Digital Case Study of Clique Bait food Store, which offers comprehensive insights into user behaviour, campaign effectiveness, and product performance through SQL Queries, providing actionable insights for optimization and growth strategies.

This project analyzes customer behavior in a large eCommerce dataset using advanced machine learning and deep learning models. The goal is to predict user actions (view, add to cart, purchase) within online shopping sessions to help businesses better understand customer journeys and improve sales strategies.