sequential-pattern-mining

Seq2Pat: Sequence-to-Pattern Generation Library

Data Mining project 2020/2021 @ University of Pisa

ERMiner: Sequential Pattern Mining algorithm for Sequential Rules generation and Event Prediction

Implementation of Sequential Pattern mining using Time interval weights

The project dives into transaction records of an online retail business to uncover hidden relationships between products. The overall goal is a data-driven approach to enhance the customer shopping experience, improve loyalty, boost profitability, tailor marketing strategies, and optimize inventory management via strategic business decisions.

SpotiFIRE makes Spotify playlists with the given topic keywords using PrefixSpan sequential pattern mining algorithm.

An implementation of sequential pattern mining algorithm with focus on a change in state.

Offical repo of the paper "A novel methodological framework for the analysis of health trajectories and survival outcomes in heart failure patients" (ICLR 2024)

Implementation and Evaluation of popular Data Mining Algorithms such as Association Rules Mining and Collaborative Filtering.

A Rigorous Sampling-Based Approach for Sequential Pattern Mining

An application of time series data discretization and episode mining techniques on stock price data.

Mining regular high utility sequential patterns by integrating sparse co-occurrence structures with look-ahead utility bounds — reference implementation and reproducible experiment pipeline.

Reference implementation and experiment harness for RHUSPM-Par: parallel regular high-utility sequential pattern mining with recursive dynamic load balancing on shared-memory multi-core machines.

Sequential pattern mining with a from-scratch GSP algorithm, benchmarked against PrefixSpan across 9 parameter configurations — execution time, memory usage, and pattern extraction analysis

Social Media Emotion Transition Analysis and Prediction using Reddit data and sequential pattern mining.

Configurable frequent and sequential pattern-mining pipelines with Apriori, FP-Growth, PrefixSpan, and GSP.

Implementation of algorithm ERMiner: Sequential Rule Mining using Equivalence Classes.

Adaptive Incremental Target Regular Sequential Pattern Mining (Ada-IncTaSPM): Java reference implementation and benchmark harness accompanying the submitted paper. MIT licensed.

Reference implementation and experimental artifacts for Pre-HUSPM-Adaptive, an adaptive pre-large algorithm for incremental high-utility sequential pattern mining.