FURI | Spring 2021

Understanding Machine vs Human Generated Text in News Articles

Data icon, disabled. Four grey bars arranged like a vertical bar chart.

The aim of this research is to create a model that can take a text input to classify the content of the message into two categories, Machine Generated and Human Written. This research uses Hierarchical Attention Networks (HAN) trained with PolitiFact data to create the predictive model. This model can be run by websites for helping prevent disinformation. In the future, HAN can be trained to infer not only from words and sentences but also from the paragraphs, which can potentially provide more insight.

View the poster

Student researcher

Mertay Dayanc

Computer science

Hometown: Ankara, Incek, Turkey

Graduation date: Fall 2021

More projects from the current symposium

William Boutin

Mechanical engineering

Flat Punch vs Spherical Indenter Probes for Fracture Testing of GaAs Single Crystals Utilized in Microelectronics and Semiconductor Industries

Investigating the effects of diverse nanoindentation probes can enhance the reliability and sustainability of semiconductor materials.

Mentor: Masoud Yekani Fard

Program: FURI

Fuad Hossain

Electrical engineering

Integration of Virtual and Augmented Reality with Nanoscale Data Visualization for Enhanced Semiconductor Manufacturing

Virtual reality and augmented reality will improve fault detection and quality control in semiconductor manufacturing, leading to more efficient advanced technology production.

Mentor: Umberto Celano

Program: FURI

Nguyen Michael Do

Electrical engineering

TCAD Simulations Aided by Physics-Informed Neural Networks

Using physics-informed neural network aided by TCAD simulation data will free up computational resources and runtime.

Mentor: Kexin Li

Program: FURI