zshwuhan

adatreenat

This software contains three closed tree mining algorithms: an incremental one IncTreeNat, a sliding-window based one, WinTreeNat, and finally one that mines closed trees adaptively from data streams, AdaTreeNat. Closed patterns are powerful representatives of frequent patterns, since they eliminate redundant information.

adversarial

Code and hyperparameters for the paper "Generative Adversarial Networks"

adwin

ADWIN is an adaptive sliding window algorithm for detecting change and keeping updated statistics from a data stream, and use it as a black-box in place or counters in learning and mining algorithms initially not designed for drifting data.

ai-ml-hospital-admissions

Machine learning techniques used to predict hospital room admissions.

ai-ml-rtkgers

A hyperplane regression tree learning algorithm (RTKGERS).

An-Online-Social-Network-based-Recommendation-System

BLSTM-for-supervised-sequence-recognition

General Purpose Implementation of LSTM extended from OCROPUS to recognize any sequential data

bnprec

Collaborative-Filtering-by-Probabilistic-Matrix-Factorization

A recommendation system (in Python) based on Probabilistic Matrix Factorization algorithm and tested on 100k ratings dataset from MovieLens

deepnet

Implementation of some deep learning algorithms.

Diff-FSPM

Mining Frequent Sequential Patterns under Differential Privacy

dshw2

Homework 2 (frequent pattern mining) of Data Structure Practice course

DynamicLDA

Dynamic Topic Modeling and Topic Chains of Reuters News Articles using SCVB0

EventPredictionV2

fuzzy_time_series_predictor

Implementation of several proposed algorithms to fuzzy time series prediction

GENet

The deep Learning network based on Graph Embedding

graphchi_mf_topn_index

TopN index for matrix factorization based graphchi

Hierarchical-topic-detection

Group documents in a hierarchy of clusters

iRoughSet

An incremental learning tookit based on rough set theory

JNMF

Matlab implementation of Joint Non-negative Matrix Factorization

M3F

Mixed Membership Matrix Factorization

MVMT_MapReduce

Multi-View Multi-Task Machine Learning Algorithm designed for a Hadoop MapReduce job to operate on large datasets with multiple views and problems with multiple tasks.

NetClus-1

Ranking-Based Clustering of Heterogeneous Information Networks With Star Network Schema NetClus Algorthms

nn_pred

Deep learning for time-series prediction.

NNForMLL

Neural Network Models for Multi-label learning

np-divs

Nonparametric nearest-neighbor-based estimators for divergences between distributions

outlier-detect

Code that implements the novel outlier detection algorithms from my Ph.D. dissertation.

pr-graph

Graph-based Posterior Regularization for Structured Prediction

symnmf

TimeTrace

This is an Android app developed for time management, user behavior analysis, and app recommendation.