João Sedoc

Department of Computer and Information Science

School of Engineering and Applied Science

University of Pennsylvania

My research interests are in Non-task driven dialog agents (Chatbots), Natural Language Processing, Time Series Analysis, and Deep Learning.

I am primarily working with Lyle Ungar as my thesis advisor and collaborating with Dean Foster, Jordan Rodu, and Jean Gallier.

My current projects are

ChatEval: A System for Chatbot Evaluation

Tree to Tree Learning

Knowledge Representation in Neural Systems

Extending Word Embeddings to Sentences

Spectral Estimation of Hierarchical factorial hidden Markov models

Predicate and Word Clustering using Signed Spectral Graph Methods