nmf-matrix-factorization

Audio Source Separation using the Non Negative Matrix Multiplication

individual Fair Nonnegative Matrix Tri-Factorization

Meta-Lingo is a comprehensive desktop application designed for corpus linguistics research. Built with modern technologies (Electron + React + Python FastAPI), it provides powerful tools for multimodal corpus management, linguistic analysis, and annotation.

Deconvolution with Iterative Completion for Estimating cellular Proportions from RNA-seq data.

Building a Collaborative Filtering based Recommender system using e-commerce data.

A Project on Topic Modeling using alogoriths like LSA/LSI, LDA, NMF on RACE dataset

Code and associated data relating to Kowalski, MacGregor, Long, Bell, and Cronin, "Automated Library Generation and Serendipity Quantification enables Diverse Discovery in Coordination Chemistry", JACS, 2022

Fair graph learning: A fair and interpretable clustering algorithm based on proportional group representation to ensure fairness and transparency in applications like graph clustering, and community discovery

Codebase and Data for the work on Network diffusion-based approach for survival prediction and identification of biomarkers using multi-omics data of Papillary Renal Cell Carcinoma

Machine learning model using NLP topic modeling to automatically classify customer complaints based on products and services for improved customer service in the financial industry.

Unsupervised classification of products based on their text description (NLP) or image (computer vision)

Detailed sentiment analysis (overall and aspect based sentiment analysis) on major Singapore attractions.

Code to recreate results from the "Multi-view biclustering via non-negative matrix tri-factorisation" paper.

This repository classifies Goodreads Fantasy book reviews into subgenres using advanced topic modeling techniques like NMF, LDA, and BERTopic. A dataset of 2M English-language reviews is analyzed, with topics compared to predefined subgenres using cosine similarity. Heatmaps and summaries visualize the results.

Survey Insights Engine

This project demonstrates the application of Non-Negative Matrix Factorization (NMF) for topic modeling on a dataset of abstracts

Reproducible EE8104 project – cross-corpus speech-emotion recognition on TESS + RAVDESS using MFCC extraction, PCA/NMF compression, and CNN / SVM baselines (91–92 % top accuracy).

15 hand postures from 8-channel surface EMG. I audited my own evaluation and found the split was leaking overlapping windows between train and test — the honest number is 14.5 points below what I first reported. 83.7% within-subject, 32.3% cross-subject.

This project aims to identify key topics in Quora questions related to popular applications. We'll use Non-Negative Matrix Factorization (NMF) to extract meaningful themes and patterns from this data.