An advanced hybrid recommendation system that combines collaborative filtering and content-based filtering approaches, enhanced with temporal awareness and contextual personalization

Spider - web crawler and local wordlist processor to generate frequency sorted wordlist / ngrams

Go metrics for calculating string similarity and other string utility functions

Demonstrates, at very small scale, how a language model is trained (bigram language model).

Demonstrates, at very small scale, how a language model is trained (bigram language model).

Demonstrates, at very small scale, how a language model is trained (unigram language model).

Demonstrates, at very small scale, how a language model is trained (unigram language model).

Simple n-gram statistical language model that illustrates some concepts used in modern LLMs, but without heavy machinery like neural networks.

A statistical 5-gram language model implemented in Python, trained on Sherlock Holmes stories by Arthur Conan Doyle (Project Gutenberg) to generate text in the author’s writing style.

^[S]\s$NLP(^\[Super\]\s$Natural Language Processing) Note. https://en.wikipedia.org/wiki/Natural_language_processing

A fluency_based evaluation tool for Chinese Grammatical Error Correction (CGEC)

A minimum viable Markov gibberish generator in 32 lines of Python, inspired by the legendary Mark V. Shaney program of 1980s

High-performance hybrid search engine for PHP. submodule@https://github.com/umaarov/goat-dev

Fuzzy / approximate string similarity metrics for PHP, JavaScript, Python

sms spam classification project(svm, tf-idf, n-gram)

Hybrid Hangman-solving agent combining Hidden Markov Models and Reinforcement Learning for intelligent, probabilistic word guessing.

3 AI projects: Search Algorithms (DFS, Bidirectional), LRTA* Pathfinding, and NLP Trigram Model in Python & Java

AI-powered Chrome Extension that autocompletes your emails in real time using an N-Gram Language Model — built with FastAPI and Python. Compose smarter and faster, locally and privately.

NLP Text Classification & Language Modeling: From-scratch implementations of Naive Bayes and Logistic Regression for emotion classification, plus N-gram models for text generation. Built with NumPy and compared against scikit-learn. Features custom evaluation metrics, confusion matrices, and comprehensive visualizations.

Built statistical N-gram language models from scratch to explore tokenization, training, probability modeling, and text generation; achieved a 97.6% perplexity improvement from unigram to 4-gram models.