Word Embeddings instead of One-hots: Using Transformers word embeddings to improve categorical feature encoding in financial fraud data

Detecting financial fraud is complex. One important source of information is textual data, such as the text within reports, descriptions, or comments. Two different words may have similar meanings but vary differently in their characters and length. In this project, we evaluate how the meaning of a word may be used to improve financial fraud […]

Read More
Geo-Distributed Serverless Computing Systems

The proposed research project aims to address a limitation in serverless computing, which is a popular technology for building scalable applications in the cloud. Currently, serverless functions are restricted to run within a specific region, which can lead to performance and cost implications when dealing with external requests from different geographical locations. The project proposes […]

Read More
Apprentissage automatique pour la résolution de problèmes de tournées de véhicules avec préférences implicites

Fastercom est une entreprise spécialisée dans le développement d’algorithmes et de logiciels permettant de résoudre différentes formes du problème de tournée de véhicules, le « Vehicle Routing Problem » (VRP). Les contraintes considérées sont généralement très diverses et varient selon les domaines des clients. L’objectif de ce projet est de concevoir une méthode visant à […]

Read More
Signalisation Ver-Mac : Localisation automatique de chantiers

Signalisation Ver-Mac est un fabricant d’équipement de signalisation routière pour les chantiers de construction (panneaux à message variable, senseur, etc.). Ver-Mac a plusieurs dizaines de milliers d’équipement en Amérique du nord qui sont connectés et sur lesquels ils collectent des données depuis plus de 10 ans. Ver-Mac a des chantiers intelligents où des données temps […]

Read More
Ethica CRO: Génération de modes opératoires normalisés assistée par NLP

Le présent projet de Ethica CRO explore le développement d’un prototype d’application pour les établissements de recherche canadiens qui intègre une solution d’intelligence artificielle (IA) au développement de modes opératoires normalisés (MON). Les ensembles de données de test dérivés des MONs de nombreux établissements de recherche canadiens conformes aux normes nationales du Canada (NNC) gouvernant […]

Read More
Efficient OLAP Queries on In-memory Databases with Frequent Updates

In this project, we will collaborate with Actenum our industry partner tackle the challenges of online analytics processing (OLAP) queries on in-memory databases that are updated frequently. Specifically, we will address the technical research issues about how to conduct effective and efficient OLAP on frequently updated data stored in a large in-memory database, and retain […]

Read More
Learning to organize and discover biomedical scientific literature

This project will investigate machine learning methods to organize and discover the vast literature on biomedical science. First, it will focus on named entity recognition–the task of finding and classifying entities in text documents–on large collections of abstracts and full-text of research papers and investigate semi-supervised learning methods to leverage large collections of unlabeled research […]

Read More
Insights and Recommendation Extraction using advances in Large Language Modeling

Businesses allocate a significant amount of financial resources and time to produce regular financial and managerial reports, which involve analyzing accounting data. These reports play a vital role in evaluating and comprehending the overall performance of the business, thereby aiding in making important decisions. Additionally, businesses often engage industry experts or consultants to derive actionable […]

Read More