[WSDM'2024 Oral] "SSLRec: A Self-Supervised Learning Framework for Recommendation"

Working white paper for OpenUBA

This repository analyzes Instagram user engagement using SQL queries to extract insights into user behavior, marketing effectiveness, and platform trends, aiming to optimize strategies, improve retention, and ensure platform integrity.

Network analysis to detect social communities, uncover behavioral patterns, and support data-driven content targeting.

Task of DLUT Big Data Department & SensorsData, third semester. Cooperate with @Bellick

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.

A MATLAB-based interactive app that analyzes Disneyland attraction visit trends using geo-tagged Flickr data. Visualizes crowd flow, attraction popularity, and visit sequences across months and years using heatmaps, pie charts, bar graphs, and more.

SQL-based ecommerce performance analysis using Google BigQuery and the Google Analytics sample dataset, delivering insights on traffic, user behavior, revenue by channel, purchasing patterns, and ecommerce funnel conversion.

Social media analytics project

Behavioral analysis SDK that classifies browser sessions as human, bot, or LLM agent using client-side detection rules.

Official Inspectlet JavaScript SDKs — session recording, heatmaps, form analytics, and AI session insights for React, Next.js, and any browser app.

"Dataherd-Raika is a library designed to simulate large-scale user behavior datasets. It takes a single user event (like a click or keyword input) and, by applying simple probability distributions and custom variables, expands it into a vast dataset."

Event-driven feature flag API with deterministic rollouts, machine learning-assisted decisions, and safe fallback behavior.

Product analytics project focused on e-commerce conversion optimization and user funnel analysis.

End-to-end e-commerce user behavior analysis | 电商用户行为全链路分析

Deep learning-based analysis of mobile device usage and user behavior using ANN. Includes data cleaning, EDA, feature engineering, model training, and infographic visualization.

A Flask-based E-Commerce Targeting System that provides customer segmentation and personalized product recommendations. Users can upload structured interaction data for analysis, receive AI-driven recommendations, and gain insights into user behavior. The application is built with Flask, Pandas, Scikit-Learn, and integrates an interactive web inter

⚡ valmi.io reverse ETL (data activation) is the open source ( OSS ) data activation platform to load data from warehouses into Webhooks and SaaS tools like Klaviyo, Facebook Ads, Salesforce, Braze etc. Valmi.io Customer Data Platform (CDP) helps track and ingest user activity events from websites, shopify, serverside events. https://cloud.valmi.io

Automated end-to-end test scripts for a music library application using Cypress. The tests simulate a typical sequence of user actions—from logging in through updating the music library—to validate real-world use cases and the integrity of user workflows.

基于 SQL 的电商用户行为分析项目，包含流量来源分析、转化漏斗诊断、新老用户对比与运营优化建议。