ConGraT

Joint text-graph modeling for improved predictions [Archived]

Research

Many platforms and datasets contain text-attributed graphs, where nodes characterized by texts are connected by edges. Examples include social media users, posts and follow relationships, as well as academic articles and their citations. We’ve developed a method called ConGraT for jointly modeling the texts and nodes in such graphs. This method is versatile and can be used on various platforms like social media, articles, and web pages, without being tailored to any specific type or structure of data. It trains a model to align texts and the corresponding nodes within a shared embedding space, making for more efficient use of data for downstream tasks. Our experiments have shown that ConGraT performs better than existing methods in classifying and predicting links between texts and items.

Computational Social Science, NLP, Social Media Analytics  , Natural Language Processing, Artificial Intelligence, Graph Modeling
Read the paper

Home     >     Research     >     ConGraT

Collaborators

William Brannon

PhD candidate

Wonjune Kang

PhD candidate

Suyash Fulay

PhD candidate

Hang Jiang

Research Affiliate (Northeastern)

Brandon Roy

Head of Applied Research

Professor Deb Roy

Director & Principal Investigator

Jad Kabbara

Research Affiliate (IBM Research)

More Research

All Research

VoiceTrace

Community voices you can check

Research

Data Provenance, Communication, Sensemaking, Civic Engagement, Participatory Research, Deliberation

Campus Trade

Helping new MIT students discover the stories that make campus special

Pilots & Programs

Conversation Design, Listening, Design, Game Design, Playful Interaction

New Social Forms

Exploring new rituals, formats, and structures for coming together

Pilots & Programs

Communication, Listening, Design, Playful Interaction, Workplace Communication

Storyline

An auditable AI framework for tracing competing narratives across podcasts, conversations, and news

Research

Computational Social Science, Large Language Model, Communication, Sensemaking, Responsible AI, Narrative

realtalk@Boston

A new civic infrastructure in Boston grounded in dialogue as a way to building “civic muscle” of democracy

Pilots & Programs

Research, Sensemaking, Conversation Facilitation, Participatory Research, Listening

AI-Assisted Conversation Highlighting

Discovering semantically or emotionally salient moments in spoken discourse using LLMs

Research

Sensemaking, Generative AI, Human-Computer Interaction