Jinane Harmouche

◦ Developing R&D Proposals: Two proposals for developing new technologies using AI got awarded by the Department of Energy.
◦ Leading R&D Machine Learning Projects: Developed end-to-end data analytic pipelines with prediction models involving data merging, conditioning and processing of raw drilling vibrations, measurements while drilling data and seismic data.
◦ Leading a Data Compression Project: Taking a principal investigator role on this project and implementing data compression algorithms for seismic and fiber optic data.
◦ Preparing Presentations & Reports: for clients and working closely with consultants.
◦ Serving as a Subject Matter Expert: on machine learning, data extraction and project management for the geophysicists and the engineers on my team.

Developing R&D proposals for new technologies using AI.
Two proposals were awarded by the Department of Energy.
Leading R&D machine learning projects.
Developed end-to-end data analytic pipelines with prediction models involving data merging, conditioning, and processing of raw drilling vibrations, measurements while drilling data, and seismic data.
Leading a data compression project.
Took a principal investigator role on the project and implemented data compression algorithms for seismic and fiber optic data.
Preparing presentations and reports for clients.
Worked closely with consultants.
Serving as a subject matter expert on machine learning, data extraction, and project management for geophysicists and engineers on the team.

Done behavioral data acquisition and analysis to assess learning and memory in Alzheimer’s mice models.
Conducted literature review on electrical brain stimulation for Alzheimer’s disease and the NIH stroke scale for its predictive and prognostic value.
Mentored data science and neuro-engineering graduate students at Rice University.

Developed NLP pipelines for content matching and text similarity between legal contracts.
Developed NLP pipelines for contracts clause classification.
Developed NLP models for detecting legal decisions that have been overruled using datasets of overruled and overruling legal cases.
Trained and fine-tuned deep learning models on large datasets using AWS machine learning and storage services.
Worked closely with subject matter experts to define requirements, solicit feedback, and establish product enhancement cycles.
Acquired a 3 months training and a certificate in reinforcement learning.

Developed association rules mining models for condition monitoring of water distribution pipelines using built-in sensors in smart fire hydrants.
Developed unsupervised signal processing and statistical features extraction methods for monitoring large gearbox systems using vibration data.
Applied the methods to the Terminal Link Train of Toronto Pearson Airport.
Mentored civil engineering graduate students at the University of Waterloo.

Conducted exploratory data analysis and regression using disparate historical environmental time-series data.
Evaluated data usefulness in the early prediction of sewer overflows using publicly available data for US states and counties.

Proposed improvements to the quality and reliability of information provided by hotel booking websites, particularly Tripadvisor.com, to facilitate users decision making.
Developed a proof-of-concept workflow and pipeline.
Added user selection fields for reviews’ time frame.
Webscraped website data of Tripadvisor.com.
Built NLP machine learning models.
Visualized results of hotels’ ranking by sentiments and reviews’ changes overtime.

Developed a novel semi-supervised time-frequency analysis and signal decomposition technique for non-stationary time-series.
Contributed to the development of a Matlab toolbox for time-series decomposition, the ASTRES Toolbox.