Revenue Growth Prediction - ON-983

Project type: Research
Desired discipline(s): Computer science, Mathematical Sciences, Mathematics
Company: PureFacts Financial Solutions Inc.
Project Length: Longer than 1 year
Preferred start date: As soon as possible.
Language requirement: English
Location(s): Toronto, ON, Canada
No. of positions: 5
Desired education level: Postdoctoral fellow
Open to applicants registered at an institution outside of Canada: No

About the company: 

PureFacts is a niche SaaS company and provider of wealth and asset management solutions globally, with staff who speak 18 languages in 13 countries around the world. We are razor-focused on the technology our clients need to improve their business and client outcomes.

Describe the project.: 

The main goal of the project is to develop a predictive machine learning model that can accurately forecast the revenue growth potential of financial advisors. This final product aims to enhance the decision-making process within financial institutions by identifying high-performing advisors and understanding the factors that contribute to their success. The model should simplify procedures and maximize revenue generation by focusing on a select number of promising advisors.

The candidate will be responsible for the following tasks:
- Exploration of Available Data: Investigate and identify relevant data sources necessary for building the predictive model.
- Data Extraction and Structuring: Extract the identified data and organize it in a human-understandable format, ensuring it is suitable for analysis.
- Thorough Data Analysis and Feature Engineering: Perform comprehensive analysis of the prepared data to extract meaningful features and select the most relevant ones for the model.
- Design and Development of the Machine Learning Solution: Develop and implement a machine learning-based solution that can effectively predict the revenue growth of financial advisors. This includes building, training, and validating the model.

The methodologies and techniques to be used in this project include:
- ETL (Extract, Transform, Load): For data extraction, transformation, and loading into a suitable format for analysis.
- Feature Extraction/Selection: Identifying and selecting the most relevant features that contribute to the revenue growth prediction.
- Time Series Analysis: To analyze patterns and trends over time, which may impact revenue predictions.
- Recurrent Neural Networks (RNN): A type of neural network suited for sequential data, useful for time-series forecasting.
- Gated Recurrent Units (GRU): A variant of RNNs that can capture long-term dependencies more effectively, potentially improving prediction accuracy.

Required expertise/skills: 

Profeciency in python. Thorogh understanding of machine learning principles and tichniques, i.e. model training/prediction, performance measuring, feature analysis, featurization, statistics analysis, etc. Knoweledge/experiece with ML packages e.g. sklearn, tensoflow, etc. Working knoweledge of SQL. Basic understanding of code control (GIT or SVN).