17 results for “topic:umap-hdbscan”
Behavioral segmentation of open field in DeepLabCut, or B-SOID ("B-side"), is a pipeline that pairs unsupervised pattern recognition with supervised classification to achieve fast predictions of behaviors that are not predefined by users.
Browser-first text exploration, clustering, and semantic search. Use AI to analyze, search, and chat with your private documents without ever uploading them to the internet - all running on your own device.
Filtering the stockmarket into clusters with financial ratios built on quarterly snapshots over 5 years of data. The goal was to see the evolution paths of companies over time and find clusters that develop future winners. Methods used, PCA, UMAP, HDBSCAN, KMeans.
Developed an unsupervised pipeline that clusters financial headlines into 100+ meaningful topics using BERTopic and HDBSCAN. Integrated sentiment analysis to track market sentiment volatility and visualize topic trends across stock tickers and time. Combined multiple datasets to create a scalable, finance-ready insight system
Repository for fine-tuning and clustering sentence embeddings for Food Items
Implemented machine learning across HR, Sales, Marketing, and PR to improve decision-making. Used models like XGBoost, Prophet, LSTM, clustering, and NLP to enhance retention, forecasting, segmentation, and sentiment analysis for business growth.
User Clustering Pipelines with BERT Models on Long and Heterogeneous Tweets - BSc Thesis
Celem projektu było wykonanie analizy budżetów partycypacyjnych polskich miast pod kątem istniejących w nich wspólnych tematów, trendów, zależności.
Group project for the NUS module "IT1244 Artificial Intelligence: Technology and Impact"
Document Clustering, Summarisation and Visualisation on 20NewsGroup
DeepSeek & Public Opinion
An app that displays a daily-updated map on crime & crime hotspots/clusters in Philly.
User Clustering Pipelines with BERT Models on Long and Heterogeneous Tweets - BSc Thesis
By utilizing various features, the Project assists the organization in categorizing countries and determining which country the NGO should allocate funds to.
Adapted BERTopic pipeline for Topic Modeling the arXiv dataset
Bachelor thesis is submitted in fulfilment of the requirements for the Bachelor of Science degree in the Department of Business Analytics and Information Technologies at the Faculty of Applied Sciences of Ukrainian Catholic University.
Topic Modelling of News Dataset