Md Parvez Mollah
Graduate Research Assistant
Md Parvez Mollah
About Me
Hi, I am Parvez, a Ph.D. Candidate at the Dept. of Computer Science in the University of New Mexico. My Ph.D. supervisor is Prof. Dr. Abdullah Mueen. My research interests are Temporal Data Mining, Machine Learning, Roadside LiDAR, and Intelligent Trasnportation System.
- Nickname Supto
- Current Residence Albuquerque, New Mexico, USA
- Email parvez@unm.edu
- Phone +1 505 835 4107
Things I Love To Do
Resume
Education
Aug. 2018 - Present
University of New Mexico, USAMS/PhD in Computer Science (Expected graduation - 2021/2023)
CGPA: 4.08/4.00, Graduate Courses: Algorithms & Data Structure, Data Mining Techniques, Advanced Operating Systems, Geometric & Probalistic Methods in CS, Intro to Cybersecurity, Intro Theory of Computation
Jan. 2012 - Apr. 2016
University of Dhaka, BDBSc in Computer Science & Engineering
CGPA: 3.59/4.00
Experience
May. 2019 - Present
University of New Mexico, USAResearch Assistant
Research interests: Time series analysis, Data Mining, Machine Learning, Smart Grid.
Currently working on roadside LiDAR data to enable functionalities of intelligent transportation system. Developed an efficient compression method for roadside LiDAR data so that it
can be sent to the cloud in real-time over 5G (published in CIKM 2022) .
Developed an efficient algorithm to summarize large time series datasets (published in ICDM 2021).
Implemented a Long Short-Term Memory (LSTM) networks model for forecasting solar power generation based on weather information.
Presented this work as a poster titled ”Forecasting Left-over Solar Capacity of Household Solar Panels” at the 26th NSF EPSCoR National Conference, 2019.
May. 2022 - Aug. 2022
Meta Platforms, Inc., USASoftware Engineer Intern (ML)
Developed an algorithm to measure the impact of training data on production Automatic Speech Recognition (ASR) models. Achieved 25-35% model performance improvement by training the model with high-valued data measured by the algorithm. Also, reduced the training set size by 50% while maintaining similar performance to the baseline model.
Aug. 2018 - May. 2019
University of New Mexico, USATeaching Assistant
Courses: Introduction to Database, Introduction to Computer Architecture and Organization.
Responsibilities include grading assignments and projects, teaching lab classes etc.
Feb. 2018 - Jul. 2018
AnyConnect Inc., BDSoftware Engineer
Developed iOS applications for communicating with Internet-of-Things(IoT) enabled devices.
Developed C++ libraries for the communication procedures between two IoT devices.
Dec. 2016 - Feb. 2018
IPvision Canada Inc., BDSoftware Engineer
Worked as a member of Big Data team; working platform- Java.
Designed and developed database layer of RingID’s channel and newsfeed features in Cassandra.
Developed several utility programs e.g., Database Schema Matcher, Query Parser etc.
Technical Skills
Programming Languages: C, C++, Java, Matlab, Python, Objective C
Databases: MySQL, Oracle, Cassandra
Tools & Libraries: Git, Keras, TensorFlow, PyTorch, OpenCV, OpenGL
Programming Contest Career
Knowledges
- Time Series Analysis
- Data Mining
- Machine Learning
- Dynamic Programming
- Data Structures
- Graph Theory
- Number Theory
Publications
J. Ferdous, M. P. Mollah, M. A. Razzaque, M. M. Hassan, A. Alamri, G. Fortino and M. Zhou, "Optimal Dynamic Pricing for Trading-Off User Utility and Operator Profit in Smart Grid", in IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 50, no. 2, pp. 455-467, Feb. 2020, doi: 10.1109/TSMC.2017.2764442.
Blog
No posts yet.