Wednesday, May 25, 2022
Media contact: Jordan Bishop | Editor, Brand Management Department | 405-744-7193 | [email protected]
The Oklahoma State University Coalition for the Advancement of Digital Research and Education (CADRE) recently partnered with Dell Technologies and Intel to recognize the exceptional use of data science and computer science by two students.
This year in CADRE 2022, Khaled Mohammed Saifuddin — a Ph.D. candidate in the Department of Computer Science — won first place in the Dell Intel Student Award for Outstanding Use of Data Science and Computing for his research work with advisor Dr. Esra Akbas, “Drug Abuse Detection in Twitter-sphere: Graph-Based Approach”.
The rate of non-medical opioid use has increased markedly since the early 2000s. Recently, the US government declared a national emergency to slow the rate of drug-related (AD) death. In this research work, Khaled presented a unique graph-based model that can automatically detect AD from freely available social media data.
In the beginning, to achieve the goal, a large number of Twitter posts were collected based on a list of keywords that also included drug names and drug abuse terms. After that, textual data was represented as graphical data called text graphics, which can handle complex structures and capture local and global word-by-word co-occurrence.
Two different types of text graphs were constructed from tweets: document-level text graphs and corpus-level text graphs. Then, different graph neural networks were applied to get the representation of nodes and graphs.
Finally, the representations were fed to a machine learning classifier to classify whether a tweet was DA-related or not. Thus, the text classification problem was introduced as a node and graph classification problem.
The experimental result shows that the proposed model outperforms the state-of-the-art baseline models with a maximum accuracy of 96.4%, which is almost 20% better than the baselines.
Ishraque Zaman Borchon — a PhD student in mechanical and aerospace engineering — received second place Dell Intel Student Award for Outstanding Use of Data Science and Computing for his research, “Development of machine learning algorithms for the high-temperature upward-facing cavity receiver for Scheffler concentrators.
This technology begins with a dish, which serves to reserve energy from the sun by using curved mirrors to direct the energy to a receiver. The upward-facing receiver is responsible for capturing solar thermal energy in the most efficient and feasible way. The receiver design requires extensive calculations and is known to be expensive. A mistake in the design could cost millions of dollars. In his research work, Borshon created software to efficiently calculate the heat loss of a complex solar heat exchanger.
The first step in Borshon’s process was to identify the three main variables that contribute to convective heat loss, which were determined to be surface temperature, tilt angle, and flow rate.
Conduction is the process of losing heat through physical contact with another object or body. Convection is the process of heat loss by the movement of air or water molecules through the skin.
Data was collected and placed into different models to determine the most accurate method of finding convective heat loss. The data determined that the physical number-based model and the random forest model were the most suitable for development.
After collecting the data and creating the necessary models, Borshon created software where users can enter predictions for each of the different variables to see the predicted heat loss, nusselt number (measurement between heat transfer per convection and conduction) and the convective heat transfer coefficient. .
The heat loss forecasting software can be found here: share.streamlit.io/ishraque2008/random_forest_heat_transfer/main.py.
For more information about OSU’s Department of Computer Science, visit cas.okstate.edu/department_of_computer_science/. To learn more about OSU’s School of Mechanical and Aerospace Engineering, visit ceat.okstate.edu/mae/.
Story by: Bailey Sisk | [email protected]
Comments are closed.