EEG based person Identification using the PhysioNet EEG Motor Movement/Imagery Dataset

This project is a FastAPI backend service that uses a BiLSTM model to classify stress levels and emotions from user's input. It also provides emotion-based video recommendations and feedback via Vertex AI. Try our streamlit app below:

Emotion detection web app using BiLSTM + Flask with Vercel-optimized serverless deployment and remote inference fallback.

In this repository, we deal with the task of implementing Natural Language Inferencing (NLI) using the SNLI dataset. Different methods such as BiLSTM, BiGRU, Attention models and Logistic Regression are experimented.

Predicting the likelihood of a job post to be fake using shallow and deep learning algorithms

Sentence Segementor basically helps in classifying sentence boundaries using Named Entity Recognition method. The classifier has been found useful in application like machine translation, sentence classification, NMT etc.

Continuous Vietnamese Sign Language Recognition using a Correlation Network and a Large Language Model for Gloss Sentence Rephrasing

Sentiment analysis web app using BiLSTM on the Amazon Polarity dataset. Built with TensorFlow and Flask for real-time sentiment prediction.

Comparison of multimodal models for Emotion Detection on IEMOCAP

Our project aims at classifying the given news articles as fake or true based on the content using few-shot learning models with fewer data.

Predictive and Prescriptive Analytics Using LSTM techniques

My undergraduate thesis program, Aspect-Based Sentiment Analysis Towards Matket Place Application Review Using Bidirectional Long Short-Term Memory used Python, Keras and Tensorflow

Implementation and evaluation of RNN, LSTM, GRU, and BiLSTM models for Named Entity Recognition on news data.

中山大学 自然语言处理项目：中文语言模型。Keras实现，BiLSTM框架。

This repository implements a Bidirectional Long Short Term Memory (BiLSTM) for performing Parts-of-Speech (POS) Tagging on Assamese-English code-mixed texts.

An experiment conducted on Twitter, using Bag of Words and BiLSTM algorithms, focus solving three questions.

Label slots in user queries

Context-dependent model pipeline (BERT-CNN-BiLSTM) for Sentiment Analysis on 2 different datasets.

Code repository for 2020 Fall COMP4901K Group Project: COVID-19 Named Entity Recognition