Skip to main content

Spring 2023 Final Projects

  • MIND News Recommendation System with OpenAI’s Pre-trained embeddings
  • English and Japanese Machine Translator
  • Polyglot Refinement: Neural Machine Translation for Low Resource Languages
  • Research Paper Summarization
  • Comparative Analysis of Adversarial Attacks on BERT-based NLP Models
  • Text Summarization of Academic Papers
  • Grammatical Error Correction using Recurrent Neural Networks
  • Syntactic Aware Cross Modality Alignment for Vision Language Navigation
  • Chatting Bot
  • Normalizing informal language using Deep Learning
  • Comparative Analysis of Deep Learning Models on Yahoo! Answers Dataset for Text Classification
  • Toxic Comment Classification
  • News Keywords Extraction Tool
  • The Role of Word Embeddings in the Task of Text Classification
  • Identify Key Information in Patient Notes from Medical Licensing Exams
  • Research Paper Title Generation based on the Abstract
  • CompareNet: Reproduction of Graph Neural Fake News Detection with External Knowledge
  • Exploring BERT-based Models’ Performance on SQuAD 2.0
  • Fake News Detection with various natural language processing models.
  • Fake News Classifier
  • Using Sentiment Classification to measure online perception
  • Comprehensive NLP Approaches to Plagiarism Detection
  • Aspect Based Sentiment Analysis For Restaurant Reviews
  • Identifying Sarcasm in Social Media Text using NLP Techniques
  • Comparison for Question-Answering System Performance using BERT, RoBERTa, ALBERT and XLNet Models

Fall 2022 Final Projects

  • Automatic essay evaluation
  • Predicting Quality of Text Generators
  • Sentiment Analysis of Customer Support Interactions on Twitter
  • Multi-Modal Emotion: Classifying Human Emotion in Video Segments
  • Toxic Comment Classification Optimization
  • Sentiment Analysis for IMDB Comments
  • Market Trend Inference By Sentiment Classification Of Financial News Headlines
  • Identification and Categorization of Offensive Language in Social Media
  • Logical Sentence Comparator
  • CipherSniffer
  • Synthesizing an Approach to QNLI Classification
  • Emotion Detection from Text using BERT-CNN
  • Spam detection of Amazon Product Review
  • Identifying and Classifying Hate Speech Using NLP
  • Bitcoin Sentiment Analysis
  • Book Genre Prediction
  • Understanding the relationship beween comments and code
  • Benchmarking QA Models
  • Fact Checking Public Health Claims
  • Fake News Stance Detection using Feature Combination and Reduction
  • Predicting TikTok Sentiment Analysis Based on Google Play Store Reviews
  • Sentiment Classification for Financial Speculation
  • Product Attribute Value Extraction via Question-Answering Approach
  • Sentiment Analysis on Movie Reviews
  • Towards a post-BERT NLP architecture
  • Snuffing Out Fake News Through Stance Detection
  • Identify toxic and harassing online comments
  • Sentiment Analysis on Movie Reviews
  • Sentiment Analysis of Movie Reviews: Positive or Negative
  • Generating MTG Cards With a RNN

Fall 2021 Final Projects

  • Comparison of Language Models on COVID-19 Tweets Text Classification
  • Reproduction and Analysis of Google’s Research Study on Natural Language Processing, Transfer Learning, and T5: the ‘Text-to-Text Transfer Transformer’
  • Aspect-level Sentiment analysis using Attention-based LSTM
  • Re-implementation of EmoNet and Further Experimentation
  • Prompt-based Text Classification Methods for Fake News Stance Detection
  • Toxic Comment Classification Challenge on Kaggle
  • Re-implementation of Bidirectional LSTM-CNNs-CRF
  • Predicting the Stock Market with News Articles using Neural Networks
  • Spam/Click-bait detection with RNN and BERT model
  • Toxic Comment Classification Challenge using NLP methods
  • Amazon Product Review Rating Prediction
  • Precise Question Classification for Applications in Customer Service
  • Toxic Comment Classification
  • Predicting Knowledge Graphs for Textual Games
  • Analyzing LSTM-based Classification Models Dependency on Spurious Correlations
  • Social Media Fake News Detection with Deep Learning
  • A Comparison of BERT-based Question Answering Models on SQuAD 2.0 and TriviaQA
  • Classification of Toxic Comments Using Naive Bayes and LSTM Models
  • Implementing Sentimental Analysis for Resort Tweets
  • Open-domain Question Answering using Wikipedia
  • CPA ChatBot
  • Exploring BERT-based Models’ Performance on SQuAD 2.0
  • Public Equity Social Sentiment Mining
  • Toxic Comment Classification
  • Comparison of Inject and Merge Encoder-decoder Models for Image Captioning
  • Robust Question Answering System
  • Using Convolutional Neural Network to Detect Fake News
  • Sentiment Analysis on Movie Reviews
  • Multi-task Learning Word Embedder
  • Fact-based Editing
  • Sentence Sentiment Analysis for IMDB Comments
  • Analysis of Adversarial Attacks against Email Spam Detection Models
  • Performing Sentiment Analysis on Reddit Comments Associated with the Stock Market and Studying Possible Price Correlations
  • Re-implement Primer: An efficient Transformer architecture
  • Comparing various transformer based architectures with applications in sentiment analysis
  • Korean Hate Speech: Classification on Online News Articles Comments
  • Emoji Prediction for Tweets
  • Toxic Comment Classification: An Evaluation of State-of-the-Art Models
  • Re-implementing Potash et al.’s GhostWriter: Using an LSTM for Automatic Rap Lyric Generation
  • Pre-trained BERT Evaluation On Image Captioning
  • Applying Sentiment Analysis on Movies Reviews Written in Chinese

Fall 2020 Final Projects

  • The detection of fake news and real news
  • Tackling Toxicity: Training Tools to Target Toxic Text
  • Analysis and Improvement of Three Top-Performing Models in Fake News Challenge
  • Comparison of Language Models on Twitter Sentiment Analysis
  • Evaluation of BERT models on Question Answering Tasks
  • Toxic Comment Classification Proposal
  • Analyze and Differentiate Application Feedback
  • Exploring Temporal Verb Embeddings in Script Event Prediction
  • Generate Sentiment Polarity of Tweets Using LSTM with Word2Vec/GloVe
  • Learning Language-to-Vision Mapping in Agent Navigation Task
  • Deep image captioning with local features
  • ITS: Improved Text Summarizer Based on TextRank
  • An implementation of the SQuAD Question Answering model
  • Fake News Detection with Deep Learning
  • Automatic Lyric Generation Model
  • Sentiment Analysis and Topic Classification of Newspaper Articles
  • Spam E-Mail Classification Comparison using Naïve Bayes and RNN Algorithms
  • Toxic Comment Classification Challenge
  • Product Review Classifier
  • Word -> Pronunciation (Grapheme to Phoneme) Translation
  • Sentiment Analysis of Public Opinion on COVID-19 Over Time
  • Sentiment Analysis on Movie Reviews
  • Deep Learning library for DPCNN - Text Categorization
  • Building and Improving an Abstractive Seq2Seq Text Summarization Model
  • Is Attention All You Need?
  • Fake News Challenge: Using Neural Networks for Stance Detection in News Articles
  • Stylistic Text to Speech Synthesis
  • Language guided 3D Shape Retrieval by Learning Joint Text-Shape Embeddings
  • A Hybridized Approach to Image Captioning
  • Text summarization from basic to advanced approaches
  • Toxic Comment Classification
  • Comment Abuse Tracker
  • IMDB sentiment Analysis
  • Seq2Seq COVID doctor Chatbot with Attention Mechanism and Unsupervised Data Augmentation

Spring 2020 Final Projects

  • Span-based End-to-End Semantic Role Labeling
  • Predicting the Usefullness of Yelp Reviews
  • Real or Not? NLP with Disaster Tweets
  • Social Robots: The Process of Translating from English to Body Language in a 3-D Vector Space
  • Joint Intent Detection and Slot Filling
  • Google QUEST Q&A Labeling
  • Tackling the Problem of Generalizing in End-to-End SLU Task
  • LangNMT: A Neural Machine Translation tool
  • Visual Question Answering Utilizing Semantic Similarity
  • Toxic/Abusive Comments Detection
  • NLP with Disaster Tweets: Predicting Which Tweets Are About Real Disasters and Which Are Not
  • Semantic Search of Speeches in Audio
  • Exploring Attention Variants of Question Answering Models on SQuAD
  • Emotional Intensity Classification
  • Toxic Comment Classification
  • Automated Patent Abstract Generator
  • Generating plausible movie reviews

Fall 2019 Final Projects

  • Positives and Negatives of Yelp Reviews
  • Ham and Spam Mobile Messages Classification
  • Automatic Tagging for Stack Overflow Queries
  • E-Ranked: A Deep Learning Based Product Search Relevance Tool
  • Fake News Detection with XLNet
  • Predicting Crude Oil Price using Real Time Price and Tweets from Twitter
  • Toxic comments evaluator with given words input
  • Headlines based SpamClickBait News Detection through NN method
  • An Enhanced Sketch2Scene Using Natural Language
  • A Simple Method for Defending Against Adversarial Attacks in NLP
  • An Ensemble Approach to Toxic Comment Classification
  • DDI extraction using CNNs and word2vec
  • Denoising Multi-Factor Attention Network for Open-Domain Question Answering
  • How I Met Your Feature: Interpretability Analysis of Question Relevance in Visual Question Answering
  • Identification and Classification of Toxic Comments with Deep Neural Networks
  • Learning children’s language model and apply it to the normal text
  • Offensive Word Classification
  • Performing sentiment analysis on movie reviews and twitter comments
  • Sentiment in Movie Reviews
  • Shape Retrieval using Shape Description Embedding
  • Irony detection in English tweets
  • Kaggle Toxic Comment Classification Challenge
  • Linking Public Opinion to Direct Action
  • Skinny Style Transfer
  • Predicting Emojis in Tweets using Sentiment Analysis
  • Sentences it Uttered, and Game Levels it Spoke
  • Solve Kaggle Question Answering Challenge with Bidirectional Encoder
  • Representations Model
  • Quizbowl
  • Toxic Comment Classification
  • Detecting Tweet Sentiment by Ordinal Classification
  • Applying sentiment analsis to predict drug interactions based on chemical structure
  • Classifying Toxic Online Comments
  • Emoji Prediction
  • Establishing Contextual Meaning From Tweets via Tokenization and Prepositional Phrase Attachment Analysis
  • Get to your point CNN !
  • Mining and Summarizing Customer Reviews
  • Obtaining word vectors for sentiment analysis: a synthesis of strategies
  • Text Sentiment Analysis
  • Text Style Transfer
  • Toxic Comment Classification
  • Twitter Sentiment Analysis System Yelp Dataset Task: Comment Rating Prediction