🦆 Contextually-keyed word vectors

Amazon SageMaker Local Mode Examples

Using pre trained word embeddings (Fasttext, Word2Vec)

Incremental learning of word embeddings with context informativeness.

Wikidata embedding

ArWordVec is a collection of pre-trained word embedding model built from huge repository of Arabic tweets in different topics. The aim of these models is to support the community in their Arabic NLP-based research.

Ensemble PhoBERT with FastText Embedding to improve performance on Vietnamese Sentiment Analysis tasks.

Code to run LDA algorithm on Twitter/Foursquare scraped data.

A tool to view how Word2Vec represents words in your favourite books.

PageRank Algorithm

A web service that exposes semantic similarity search via a web GUI and a RESTful API.

Automnomously attempting a categorical summarization of a sparse, asymmetrical corpus in English language, by performing text classification - which is achieved by our intuitive sentence pair classification scenarios and usecases.

graphsense is a library to train and infer graph embedding based code suggestion models

Text classification is the task of assigning a set of predefined categories to free text. Text classifiers can be used to organize, structure, and categorize pretty much anything. For example, new articles can be organized by topics, support tickets can be organized by urgency, chat conversations can be organized by language, brand mentions can …

Natural Language Processing (traitement automatique du langage naturel )

This application is built for employers looking for candidates against a particular job description .

A simple project that creates a dataset of News Headlines with Primary Category, Secondary Category, Date, Day, Month,Year, Sentiment, SentimentPolarity, Emotion and Url. All News Headlines are scraped from punch newspaper and sorted into a csv file.

A prototype legal text search engine that uses a semantic search algorithm in order to find related keywords and sort the results by relevance.