![](img/saeid.jpg)
I am Third Year Ph.D Student at department of Computer Science , University Of Georgia. I have started my PhD studies in august 2015 . I am working with Prof. Hamid Arabnia and Prof. Juan B. Gutierrez. My research focuses on Machine Learning and Deep Learning, DNA Computing , Optimization , Data Privacy , Probabilistic Model .
Ph.D. in Computer Science, Department of Computer Science, The University of Georgia, 2015 - present
M.Sc. in Applied Mathematics ( Operations Research ), The University of Arak, Iran, 2005 - 2007
B.Sc. in Applied Mathematics, The University of Kashan, Iran, 2000 - 2005
My research interests include, but not limited to the following areas:
Machine Learning
Deep Learning (RNN,CNN,Atiention model,LSTM,GAN)
Text Mining
Natural Language Processing
DNA Computing
Optimization
Data Privacy
Google Scholar
ResearchGate
My Current research Projects:
Image classifier for snakes
In this project, we created a dataset with around 20 000 images from poison and non-poison snakes. In short, the accuracy was around 65 which with some techniques we increased it up to around 90. We still working on it to produce a mobile app for it which can detect 10 poison snakes and around 40 non-poison snakes in the USA. We will submit a paper to ICMLA 2018.Learning privately from different datasets with Expectation–maximization (EM)
Suppose we have different datasets. we collect those data and predict the model for all datasets. But we need to label data. We can use maximum vote and regression, but they are too sensitive about noise. In the paper “ Learning Privately from Multiparty Data Jihun Hamm, Paul Cao, Mikhail Belkin ICML 2016 “ they proposed to use logistic regret ion to preserve privacy and they showed that if the number of datasets increases, sensitivity will decrease. In this project, we assigned weight to datasets base on size of datasets and accuracy. After that, we used expectation maximization to label data.Produce SPARQL based on the description.
We use a dataset with 5000 entries . each row has a SPARQL and its description. We created an encoder-decoder LASTM to convert description to SPARQL. This model is like language translation. The accuracy of this model was not acceptable because when the length of sentence increase, the performance will decrease. In this step, we want to implement LSTM with attention to keep performance while length increase.The List of NSF or NIH projects I was involved with:
ALICE
ALICE stands for Adaptive Learning for Interdisciplinary Collaborative Environments. ALICE is an information system that designs a learning environment for every student based on her/his interaction with the system. ALICE is under active development.
Malaria Host-Pathogen Interaction Center (MaHPIC)
The MaHPIC combines Emory investigators’ interdisciplinary experience in malaria research, metabolomics, lipidomics and human and non-human primate immunology and pathogenesis with UGA’s expertise in pathogen bioinformatics and large database systems, and Georgia Tech’s experience in mathematical modeling and systems biology. The CDC has provided support in proteomics and malaria research, including nonhuman primate and vector/mosquito infections. NIAID Grant HHSN272201200031C
Acute Models of Malaria to Study Experimental Resilience (THoR’s HAMMER)
A research partnership within DARPA’s THoR (Technologies for Host Resilience) program, the HAMMER project uniquely focuses on malaria and its effects on human and non-human primate hosts. This project will explore the molecular mechanisms of resilience, susceptibility and resistance of non-human primate hosts when challenged with a malaria infection. DARPA contract #W911NF-16-C-0008
Solving Minimum k-supplier in Adleman-Lipton model
Saeid Safaei, Vahid Safaei, Elizabeth D. Trippe, Karen Aguar, Hamid R. Arabnia
4th Annual Conf. on Computational Science & Computational Intelligence (CSCI'17) | Dec 14-16, 2017 | Las Vegas, Nevada, USA
A Vision for Health Informatics: Introducing the SKED Framework An Extensible Architecture for Scientific Knowledge Extraction from Data
Juan B. Gutierrez, Elizabeth D. Trippe, Jacob B. Aguilar, Yi H. Yan, Mustafa V. Nural, Jessica A. Brady, Mehdi Assefi, Saeid Safaei, Mehdi Allahyari, Seyedamin Pouriyeh, Mary R. Galinski, Jessica C. Kissinger.
bigdas: Big Data Analytics as a Service - bigdas@KDD2017, August 14, 2017, Halifax, Canada.
A Brief Survey of Text Mining: Classification, Clustering and Extraction Techniques
Mehdi Allahyari, Seyedamin Pouriyeh, Mehdi Assefi, Saeid Safaei, Elizabeth D. Trippe, Juan B. Gutierrez, Krys Kochut.
bigdas: Big Data Analytics as a Service - bigdas@KDD2017, August 14, 2017, Halifax, Canada.
Text Summarization Techniques: A Brief Survey
Mehdi Allahyari, Seyedamin Pouriyeh, Mehdi Assefi, Saeid Safaei, Elizabeth D. Trippe, Juan B. Gutierrez, Krys Kochut.
arXiv preprint arXiv:1707.02268v1. July 2017.
Solving Minimum Beta-vertex Separator Problems in the Adleman-Lipton Model.
Dolati, Ardashir, Mehdi Sohrabi Haghighat, Saeed Safaei, and Hajar Mozaffar.
Proceedings of the 2008 International Conference on Foundations of Computer Science, FCS 2008, July 14-17, 2008, Las Vegas, Nevada, USA. CSREA Press 2008, ISBN 1-60132-066-3 pp. 97-101.
Fast Parallel Molecular Solution to the Maximum Triangle Packing Problem on Massively Parallel Bio-Computing.
Dalvand, Babak, Saeed Safaei, and Mojtaba Nazari.
Proceedings of the 2009 International Conference on Foundations of Computer Science, FCS 2009, July 13-16, 2009, Las Vegas, Nevada, USA. CSREA Press 2009, ISBN 1-60132-103-1 pp. 169-173.
Molecular Solutions for the Minimum Edge Dominating Set Problem on DNA-based Supercomputing
Safaei, Saeed, Babak Dalvand, Babak Esmaeili, and Vahid Safaei.
Proceedings of the 2009 International Conference on Foundations of Computer Science, FCS 2009, July 13-16, 2009, Las Vegas, Nevada, USA. CSREA Press 2009, ISBN 1-60132-103-1 pp. 32-36.
Solving bin-packing problem in the adleman–lipton model
Dolati, Ardeshir, Mehdi S. Haghighat, and Saeed Safaee.
In the First Conference and Workshop on Mathematical Chemistry, vol. 1, pp. 70-74. 2008
Molecular Solutions for the Bin-packing and Minimum Makespan Scheduling Problems on DNA-based Supercomputing
Safaei, Saeid, Babak Dalvand, Zahra Derakhshandeh, and Vahid Safaei.
Proceedings of the International Conference on Parallel and Distributed Processing Techniques and Applications, PDPTA 2009, Las Vegas, Nevada, USA, July 13-17, 2009, 2 Volumes. CSREA Press 2009, ISBN 1-60132-123-6 pp. 512-516
Molecular solutions for the maximum K-Facility dispersion problem on DNA-based supercomputing.
Safaei, Nozar, Babak Dalvand, Saeed Safaei, and Vahid Safaei.
Proceedings of the 2011 International Conference on Foundations of Computer Science, FCS 2011, July 2011, Las Vegas, Nevada, USA. CSREA Press 2011.
Molecular solutions for the maximum K-Facility dispersion problem on DNA-based supercomputing.
Safaei, Nozar, Babak Dalvand, Saeed Safaei, and Vahid Safaei.
Proceedings of the 2011 International Conference on Foundations of Computer Science, FCS 2011, July 2011, Las Vegas, Nevada, USA. CSREA Press 2011.
Solving Minimum K-Center Problem in the Adleman? Lipton Model.
Safaei, Saeed, Hajar Mozaffar, and Babak Esmaeili.
Proceedings of the 2008 International Conference on Foundations of Computer Science, FCS 2008, July 14-17, 2008, Las Vegas, Nevada, USA. CSREA Press 2008, ISBN 1-60132-066-3 pp. 97-101. pp. 251-255
Solving several kinds of constrained shortest path problems via DNA computations.
Sohrabi-Haghighat, Mahdi, Ardeshir Dolati, and Saeed Safaee.
Applied mathematics in Engineering, Management and Technology 3(1) 2015:527-533
Considerations on Interdisciplinary Instruction and Design Influenced by Adaptive Learning. A Case Study Involving Biology, Computer Science, Mathematics, and Statistics
Karen Aguar, Charles C. Sanchez, Diego Boada Beltran, Saeid Safaei, Mehdi Asefi, Jonathan Arnold, Pedro Portes, Hamid R. Arabnia, Juan B. Gutierrez
arXiv preprint arXiv:1703.06010 (2017).
email : ssa@uga.edu
address : Pound Hall,105 Foster Rd Athens, GA 30606