Candidate Information Title Data Scientist Part-Time Target Location US-NC-Apex Email Available with paid plan Phone Available with paid plan	20,000+ Fresh Resumes Monthly     View Phone Numbers     Receive Resume E-mail Alerts     Post Jobs Free     Link your Free Jobs Page     ... and much more Register on Jobvertise Free
Search 2 million Resumes Keywords: City or Zip:	Related Resumes

HAOQI NIEMAIL AVAILABLE 919-279-5288 LinkedIn: https://LINKEDIN LINK AVAILABLE Greencard Holder EXPERIENCE Data Scientist, Leoforce LLC, Raleigh, NC 2021 - 2023 EDUCATION Ph.D., Electrical and Computer Engineering, North Carolina State University, Raleigh, NC 2016 - 2020 M.S., Electrical Engineering, Shanghai University, Shanghai 2012 - 2015 SKILLS Procient in Python (Numpy, Pandas, Scikit-learn, Keras), Java, R, SAS, C/C++, SQL, MATLAB, Git, Tableau, Microsoft Power BI and the most common development and data analysis tools. PROJECTS1. Predictive Maintenance (Python): Aug. 2023 - Nov. 2023(1) Is the asset working under normal condition: using anomaly detection techniques to detect the working status.(2) Will an asset fail within a certain time or some certain time period: Reformat the problem as Binary classication or multi classication problems and apply ML models and deep learning models to tell the results.(3) Remaining useful life for an asset: Reformat the problem as a regression problem, use piece-wise linear degradation model to process the training target and then use regression models and deep learning models to predict the remaining useful life.2. Roles Classication (Python): Jun. 2023 - Aug. 2023 Collaborate with a team member to work on the role classication problem. Given a job title, output will be the corre- sponding role. Explore to use Bert/Transformers and XLNet to do these text classication task. 3. Inventory Restocking (Python): Nov. 2022 - May. 2023 Discover the inventory restocking problem for a slab sales warehouse. Optimization of this problem would be ensuring the customer dont miss any sales opportunities without overstocking. Convert this problem into the prediction of their future sales so that they can replenish products at the right time and at the right place based on demand and projected sales. Found the cumulative sales of their history sales data is linear based on data analysis even though their daily sails amount more like a random number. Apply linear model to predict their PHONE NUMBER AVAILABLE days sales. 4. Light weight analyzer (Python): Jun. 2022 - Oct. 2022 Using fastText model to train the skills of the candidates and make the skills in the word vectors format. Find the similar skills by comparing the similarity between the skills by using the word vectors, and then OR join the similar skills to a Bollean search criteria to quickly nd the candidates with required skills. The results show that after adding this function, the sourced candidates increase around 5%.5. Movers Response (Python): Nov. 2021 - May. 2022 For the job candidates, we want to know how likely they response to the emails. Based on the historical data, analyze the features of the candidates like their job title, skills, working days for one job, skills, locations, etc and select some of them which contribute the most to their responses. Conduct several machine learning models to compare the results and apply the best one (Gradient Boosting Classier) to identify whether a candidate will response or not and assign the corresponding scores to the candidates.6. Title/Company scrubbing (Python): Apr. 2021 - Oct. 2021 As the previous spelling correction was pretty slow expecially when dealing with the cases where the edit distance is greater than 2. Apply symmetric spell algorithm to reduce the complexity of edit candidata generation and dictionary lookup by using deletes only instead of deletes + transposes + replaces + inserts. Results show that the new algorithm is six orders of magnitude faster (for edit distance=3) and language independent than the previous one. 7. Deep learning based Method for Wide-Area Control (Python, Matlab, SQL): Aug. 2019 - Jul. 2020(1) Propose a novel delay aware prediction control framework to address the data transmission delay of PMU measure- ments in modern power systems;(2) Apply machine/deep learning techniques to predict the PMU data;(3) Conduct the experiments by using LSTM models to predict future PMU data and recover the unrecevied ones. 8. Online Adaptive Wide-Area based on Dynamic Anomaly Delay Detection: Aug. 2018 - Jul. 2019(1) Discover the problem that the control performance would decrease as more communication drops during the wide- area communication;(2) Use a Kernal Density Estimation (KDE) based model to detect the anomaly delay which may potentially cause the sys- tem unstable;(3) Address the problem by tuning the control gains associated with the existing links to make the overall system perfor- mance remain approximated optimal.9. Evolutionary Algorithms and the Applications to Real-World Problems (Java, Matlab): Aug. 2012 - Jul. 2015(1) Present a novel meta-heuristic algorithm named human learning optimization, inspired by human learning mech- anisms, including individual, social and random learning operators to solve the binary-coded problems (tested on the benchmark functions) more eectively;(2) Compare the results with mainstream evolutionary algorithms on the well-know MKPs problems (8% - 15% improve- ment);(3) Apply these algorithms to real-world problems such as node placement of industrial wireless sensor networks and data-driven control;(4) Evolutionary algorithm based feature selection for biomedicine. 10. Online Tuning of Cloud-based Wide-Area Control (C++, Matlab): Aug. 2016 - Jul. 2018(1) Propose an optimized SDN controller to select a shortest path in transmitting the measured data;(2) Propose a constrained feedback control algorithm to ensure complete trust between the SDN controller and manage- ment applications;(3) Conduct sparse control algorithm together with SDN controller in the Exo-GENI cloud to test the validation and control performance of the power system control.AWARDS Best Conference Paper in the 2018 IEEE Power & Energy Society General Meeting. PUBLICATIONS1. Haoqi Ni, Aranya Chakrabortty, and Yufeng Xin. "Online Tuning of Cloud-based Wide-Area Controllers with Variations in Network Trac." 2019 IEEE Power & Energy Society General Meeting (PESGM). IEEE, 2019. 2. Haoqi Ni, Mohamed Rahouti, Aranya Chakrabortty, Kaiqi Xiong, and Yufeng Xin. "A distributed cloud-based wide-area controller with SDN-enabled delay optimization." 2018 IEEE Power & Energy Society General Meeting (PESGM), pp. 1-5. IEEE, 2018.3. Ruixin Yang, Junyi He, Mingyang Xu, Haoqi Ni, Paul Jones, and Nagiza Samatova. "An intelligent and hybrid weighted fuzzy time series model based on empirical mode decomposition for nancial markets forecasting." In Industrial Conference on Data Mining, pp. 104-118. Springer, Cham, 2018.4. Ling Wang, Lu An, Haoqi Ni, Wei Ye, Panos M. Pardalos, and Min-Rui Fei. "Pareto-based multi-objective node placement of industrial wireless sensor networks using binary dierential evolution harmony search." Advances in Manufacturing 4, no. 1 (2016): 66-78.5. Ling Wang, Haoqi Ni, Ruixin Yang, Panos M. Pardalos, Xin Du, and Minrui Fei. "An adaptive simplied human learning optimization algorithm." Information Sciences 320 (2015): 126-139. 6. Ling Wang, Haoqi Ni, Ruixin Yang, Panos M. Pardalos, Li Jia, and Minrui Fei. "Intelligent virtual reference feedback tuning and its application to heat treatment electric furnace control." Engineering applications of articial intelligence 46 (2015): 1-9.7. Ling Wang, Ruixin Yang, Haoqi Ni, Wei Ye, Minrui Fei, and Panos M. Pardalos. "A human learning optimization algorithm and its application to multi-dimensional knapsack problems." Applied Soft Computing 34 (2015): 736-743. 8. Ling Wang, Haoqi Ni, Weifeng Zhou, Panos M. Pardalos, Jiating Fang, and Minrui Fei. "MBPOA-based LQR controller and its application to the double-parallel inverted pendulum system." Engineering Applications of Articial Intelligence 36(2014): 262-268.9. Ling Wang, Haoqi Ni, Ruixin Yang, Minrui Fei, and Wei Ye. "A simple human learning optimization algorithm." In Compu- tational Intelligence, Networked Systems and Their Applications, pp. 56-65. Springer, Berlin, Heidelberg, 2014. 10. Xikun Wang, Lin Qian, Ling Wang, Muhammad Ilyas Menhas, Haoqi Ni, and Xin Du. "A Novel Deterministic Quantum Swarm Evolutionary Algorithm." In Computational Intelligence, Networked Systems and Their Applications, pp. 111-121. Springer, Berlin, Heidelberg, 2014.11. Ling Wang, Haoqi Ni, Ruixin Yang, Vijay Pappu, Michael B. Fenn, and Panos M. Pardalos. "Feature selection based on meta-heuristics for biomedicine." Optimization Methods and Software 29, no. 4 (2014): 703-719. PROFESSIONAL ACTIVITIESInvited Reviewer for Top Journals and Conferences in the eld of Computer Science and Electronic Engineering such as IEEE Transactions onPowerSystems;IEEE Transactions onControlSystemsTechnology;Applied Soft Computing; Knowledge-based System; IEEE PESGM 2018/2019/2020.