Efficient and Accurate Project Management Analysis Platform using Active and Few-Shot Learning Techniques for Advanced Insights and Predictions

This grant application, aims to develop an efficient and accurate project management analysis platform based on active and few-shot learning techniques. The main objective of our platform is to provide advanced insights and predictions related to project activities, teams, and budgets, using recent advancements in natural language processing (NLP), active and few-shot learning technologies. By automating most of the data science processes, our platform aims to reduce the need for manual data annotation, thereby minimizing human error and increasing the efficiency of the analysis. Our platform consists of two main subsystems: a log management and annotation subsystem and a model training and prediction subsystem. The log management and annotation subsystem will provide tools for efficiently managing and annotating project management logs, including NLP-based methods for automatically extracting relevant information from the logs. The model training and prediction subsystem will leverage active and few-shot learning techniques to train machine learning models that can predict project-related risks, identify over- or under- budgeting and data quality of project-related information. By combining these two subsystems, the platform will provide organizations with a powerful tool for making data-driven decisions and improving project management performance..

Faculty Supervisor:

Essam Mansour

Student:

Partner:

Banque Nationale du Canada

Discipline:

Computer science

Sector:

Finance and Insurance; Professional, scientific and technical services

University:

Concordia University

Program: