hdbscan-clustering

The thesis presents the parallelisation of a state-of-the art clustering algorithm, FISHDBC. This objective has been achived by improving the main data structures and components of the algorithm: HNSW, MST and HDBSCAN. My contribution is based on a lock-free strategy, completely wrote in Python.

tgmm: A Gaussian Mixture Model based on Expectation-Maximisation implemented with PyTorch

Repository documenting the code to the paper "From Clusters to Graphs – Toward a Scalable Viewing of News Videos"

Turn news streams into searchable events and source-linked timelines. Python library, FastAPI service, and local-first SQLite storage.

Welcome to the CDP Knowledge Assistant! This chatbot leverages cutting-edge Natural Language Processing (NLP) techniques to help users retrieve and answer queries from documentation of CDP platforms such as Segment, mParticle, Lytics, and Zeotap.

I built an end-to-end customer churn segregation and prediction project.

A lightweight, configurable toolkit for unsupervised clustering of short or long texts. It builds vector embeddings for your data, discovers dense regions with HDBSCAN, optionally reduces dimensionality for visualization (e.g., t‑SNE), and exposes graph-friendly outputs for downstream exploration.

User Clustering Pipelines with BERT Models on Long and Heterogeneous Tweets - BSc Thesis

2023F-Machine Learning Term Project: Topic Modelling on Amazon Review

Autonomous AI incident intelligence & remediation platform. Ingests distributed telemetry with Kafka, clusters anomalies via HDBSCAN & Redis, performs autonomous root-cause code diagnostics with LangGraph, and delivers real-time SRE command center with HITL approval

Deterministic HDBSCAN* clustering of geospatial points using shortest-path distance on spatial networks.

Bioinformatics, data wrangling, machine learning, and statistical analysis of microscopy and proteomics data in Dr. Martha Bhattacharya's lab at the University of Arizona

Модуль 6. Навчання без вчителя.  Кластерізація. KMeans. Principal Component Analysis

User Clustering Pipelines with BERT Models on Long and Heterogeneous Tweets - BSc Thesis

AI-powered intelligence aggregation platform (internship) - ingestion pipelines, LLM clustering with HDBSCAN, PostgreSQL vector search, and a production Row-Level Security model with PII detection.