Predicting ocean-wave conditions using buoy data supplied to a hybrid RNN-LSTM neural network and machine learning models

Gulustan Dogan, Meghan Ford, Scott James

The ability to accurately predict ocean-wave conditions is paramount for many maritime activities. A framework comprising a bi-directional recurrent neural network (RNN) with long-short term memory (LSTM) cells (RNN-LSTM) was developed for timely and accurate predictions of ocean-wave conditions.

Human Activity Recognition Using Convolutional Neural Networks

Gulustan Dogan, Sinem Sena Ertas, İremnaz Cay

Using smartphone sensors to recognize human activity may be advantageous due to the abundant volume of data that can be obtained. In this paper, we propose a sensor data based deep learning approach for recognizing human activity.

Understanding the Effects on Balance for Elite Platform Divers Using Machine Learning

Gulustan Dogan, Blythe Layne, Seyda Ari, Morgan Glisson, Evan Kurpiewski, Michel Heijnen, Karl Ricanek

Machine learning plays a crucial role in our society’s efforts to combat injury to athletes.

Locomotion-transportation recognition via LSTM and GPS derived feature engineering from cell phone data

Gulustan Dogan, Jonathan Daniel Sturdivant, Seyda Ari, Evan Kurpiewski

This paper was put forth to test the notion of detecting forms of locomotion from various radio frequency data for the 2021 SHL recognition challenge.

Computational Thinking: A Pedagogical Approach Developed to Prepare UNCW Students for the Era of Artificial Intelligence

Gulustan Dogan, Yang Song, Damla Surek

We propose Computational Thinking (CT) as an innovative pedagogical approach with broad application. Research and current industry trends illustrate that students should have a solid computational thinking ability in order to have the skills required for future jobs in Artificial Intelligence.

911 4 COVID-19: Analyzing impact of covid-19 on 911 call behavior

Gulustan Dogan, Rachel Carroll; Gozde Merve Demirci

This paper explores the impact of COVID-19 on 911 Call behavior to help first responders develop effective solutions to emergent situations proactively. Correct prediction of call volume and call type helps first responders optimize resource allocation.

A Learning Analytics Case Study: Relation of Students’ Learning Approach to Online Learning Environment Behaviours

Gulustan Dogan, Seydi Alkan, Alper Bayazit, & Gozde Merve Demirci

The aim of this research is to examine the relation between the log data obtained from the learning platform prepared within the scope of this study and the learning approaches of the students.

Using artificial intelligence to predict fall-risk during adaptive locomotion in humans

Gulustan Dogan, Nouran Alotaibi, Elif Sahin, Sinem Sena Ertas, Iremnaz Cay, Şeref Recep Keskin, Michel JH Heijnen, Karl Ricanek

Falls are the third leading cause of unintentional injuries for ages 18-35 years according to the CDC; although there are many studies for falls in the elderly population, the causes and circumstances of falls for younger adults are understudied.

DNN and CNN Approach for Human Activity Recognition

Gulustan Dogan, Seref Recep Keskin, Aysenur Gençdoğmuş,Buse Yıldırım & Yusuf Öztürk

One of the common causes of low back pain is postural stress. When sitting or walking, poor posture may result in spinal dysfunction.

Where are you? Human activity recognition with smartphone sensor data

Gulustan Dogan, Iremnaz Cay, Sinem Sena Ertas, Şeref Recep Keskin, Nouran Alotaibi, and Elif Sahin

This research is mainly focused on recognizing modes of transportation (Still, Walk, Run, Bike, Car,Bus, Train, Subway) in a user-independent manner with an unknown phone position. The goal was to recognize the user's transportation (activities) from data coming from the phone of the “test” user while the location of that phone on the “test” user is not specified.

A Data-Driven Approach to Kinematic Analytics of Spinal Motion

Seref Recep Keskin, Aysenur Gencdogmus, Gulustan Dogan, Yusuf Ozturk

We employ deep learning and machine learning methods to study spine motion and postural stress using two sensors attached to lower back of a healthy subject while the subject is performing regular daily activities.

A Deep Learning Approach to Predict Learners' Primary Language in MOOCs

Ismail Duru, Ayse Saliha Sunar, Gulustan Dogan, Su White, Banu Diri

In this study, we aim to exploit deep learning for course dropout prediction of second language English speakers by using their posts to discussions gathered from a range of MOOCs without performing feature extraction.

Sentiment Analysis of Twitter Data with Deep Learning

Gozde Merve Demirci, Gulustan Dogan, Sule Itir Satoglu

We train a deep neural network on twitter data for sentiment analysis. We aim to see whether social media can help disaster management. Our findings indicate that social media can be used to understand sudden situations like disasters and may help to decide how to act.

Energy Efficient Smart Buildings: LSTM Neural Networks for Time Series Prediction

Idil Sulo, Seref Recep Keskin, Gulustan Dogan, Theodore Brown

We use the Long Short Term Memory (LSTM) neural network model to analyze the energy expenditures of the buildings that reside in the campuses of the City University of New York (CUNY).

Protru: a provenance-based trust architecture for wireless sensor networks

Gulustan Dogan

We design a distributed trust‐enhancing architecture using only local provenance during sensor fusion with a low communication overhead. Our network is cognitive in the sense that our system reacts automatically upon detecting low trust.