Cereja is a bundle of useful functions we don't want to rewrite and .. just pure fun!

中文文本相似度计算器

Scrape job websites into a single spreadsheet with no duplicates.

Search anything, instantly

Klasifikasi Berita Online pada KOMPAS untuk mata kuliah Pencarian dan Penambangan Web menggunakan metode Logistic Regression

Specialized Intelligent Text Matching and Correction Engine

This project focuses on the development of a Recurrent Neural Network (RNN) model using Gated Recurrent Units (GRUs) for Twitter sentiment analysis, along with hyperparameter tuning. The performance of the RNN-GRU model is compared against two pre-existing models

LOcal Search Engine

This is a Full Stack project that incorporates Next.js & React, Big Data, Data Engineering, AI/ML, and NLP. It is meant to be YouTube video recommender for Computer Science students.

中文文本数据去重小工具

CorpusReader_TFIDF is a custom Python class designed to calculate TF-IDF (Term Frequency-Inverse Document Frequency) for documents in a corpus. It was developed as part of an assignment for the Introduction to Natural La

Analyzing and classifying French tweets related to global warming and drought using NLP and Machine Learning. - Analyse et classification des tweets français parlant du réchauffement climatique et de la sécheresse en utilisant le traitement du langage naturel (NLP) et l'apprentissage automatique.

Busca por posts no Bluesky usando TFIDF para classificar relevância dos resultados

This project is a Cocktail Recommender System that utilizes TF-IDF (Term Frequency-Inverse Document Frequency) to recommend cocktails based on user queries.

An internet search engine written mostly in python. Currently TF-IDF based.

Book recommendation system and reader rating analysis.

This project aims to develop a sentiment classifier model using an equally distributed subset of kaggle’s Twitter Sentiment Analysis dataset. The objective is to classify each tweet into one of the sentiment classes using various machine learning models, and optimize their performance through hyperparameter tuning and evaluation techniques.

智源 - 看山杯 专家发现算法大赛 2019-传统方法

NLP use cases using popular solutions: Frequency Embeddings, Word embedding (word2vec, doc2vec, Glove), RNN,LSTM, Transformers-BERT, Sentence_Transformers etc. PyTorch

Developed BERT, LSTM, TFIDF, and Word2Vec models to analyze social media data, extracting service aspects and sentiments from a custom dataset. Provided actionable insights to telecom operators for customer satisfaction and competitive analysis.